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
|
|---|---|---|---|---|---|
Most genes in bacteria are experimentally uncharacterized and cannot be annotated with a specific function . Given the great diversity of bacteria and the ease of genome sequencing , high-throughput approaches to identify gene function experimentally are needed . Here , we use pools of tagged transposon mutants in the metal-reducing bacterium Shewanella oneidensis MR-1 to probe the mutant fitness of 3 , 355 genes in 121 diverse conditions including different growth substrates , alternative electron acceptors , stresses , and motility . We find that 2 , 350 genes have a pattern of fitness that is significantly different from random and 1 , 230 of these genes ( 37% of our total assayed genes ) have enough signal to show strong biological correlations . We find that genes in all functional categories have phenotypes , including hundreds of hypotheticals , and that potentially redundant genes ( over 50% amino acid identity to another gene in the genome ) are also likely to have distinct phenotypes . Using fitness patterns , we were able to propose specific molecular functions for 40 genes or operons that lacked specific annotations or had incomplete annotations . In one example , we demonstrate that the previously hypothetical gene SO_3749 encodes a functional acetylornithine deacetylase , thus filling a missing step in S . oneidensis metabolism . Additionally , we demonstrate that the orphan histidine kinase SO_2742 and orphan response regulator SO_2648 form a signal transduction pathway that activates expression of acetyl-CoA synthase and is required for S . oneidensis to grow on acetate as a carbon source . Lastly , we demonstrate that gene expression and mutant fitness are poorly correlated and that mutant fitness generates more confident predictions of gene function than does gene expression . The approach described here can be applied generally to create large-scale gene-phenotype maps for evidence-based annotation of gene function in prokaryotes .
Advances in sequencing technology have ushered in a new era of bacterial genomics . At low cost , a single individual can sequence the complete genome of a bacterial isolate in less than a week . While this explosion in genome sequencing and gene discovery is breathtaking , it also serves as a reminder that for most bacteria , we do not know the function for most of the genes in the genome [1] . Even in the model bacterium Escherichia coli , which has been extensively studied for decades , there are hundreds of genes that are poorly annotated or entirely hypothetical [2] . Therefore , it is critical that methods for systematically elucidating gene function in microbial genomes are developed [3] . The current paradigm is that newly sequenced bacterial genomes go through a computational annotation pipeline that predicts gene structure and putative function . The latter is predicted from sequence homology to known gene families , protein domains , and characterized enzymes . However , given that most experimentally characterized genes derive from a small number of bacteria representing a tiny fraction of prokaryotic diversity , there are a large number of gene families that have never been experimentally characterized and hence computational annotations are useless beyond “conserved domain” or “conserved hypothetical . ” Furthermore , due to weaker sequence conservation , computational annotations of gene function in microbial species get progressively less reliable the further one moves away from well-studied model bacteria such as E . coli and Bacillus subtilis [4] . Lastly , there are classes of genes ( for instance , transcription factors [5] and transport proteins [6] ) for which homology-based annotations are either vague or unreliable . Taken together , computational predictions alone , while a necessary first step towards genome annotation , are not sufficient to meet the growing challenge of assigning function to the millions of genes identified by DNA sequencing . One attractive approach to characterize genes on a global scale is via the analysis of large-scale mutant collections . Mutants provide insight into gene function by providing a direct link between genotype and a cellular phenotype . By correlating genes with their phenotypes , a specific gene function can often be inferred [7] , [8] , [9] , [10] . In the post-genome era , a number of microorganisms have been subjected to large-scale mutagenesis and phenotyping efforts . In bacteria , genome-wide mutant collections have been constructed for several bacteria , using either targeted methods [11] , [12] , [13] or random transposon mutagenesis [14] , [15] , [16] . One attractive approach to characterize genes on a global scale is via the analysis of large-scale mutant collections . Mutants provide insight into gene function by providing a direct link between genotype and a cellular phenotype . By correlating genes with their phenotypes , a specific gene function can often be inferred [7] , [8] , [9] , [10] . In the post-genome era , a number of microorganisms have been subjected to large-scale mutagenesis and phenotyping efforts . In bacteria , genome-wide mutant collections have been constructed for several bacteria , using either targeted methods [11] , [12] , [13] or random transposon mutagenesis [14] , [15] , [16] . Regardless of how mutant strains are generated , a key challenge is the quantitative analysis of the mutant collections across the diverse range of conditions necessary to identify phenotypes for the majority of genes in the genome [17] . The phenotypes of mutant collections can be assayed in high-throughput either as individual strains or in pooled , competitive fitness assays . In a recent example of the former approach , the individual mutant strains of the E . coli KEIO deletion collection were assayed in hundreds of growth conditions using an agar-based colony size assay [10] . At least one phenotype was identified for ∼50% of E . coli genes using this individual strain assay . Conversely , the use of pooled assays to measure mutant phenotypes is best exemplified in Saccharomyces cerevisiae . Each yeast deletion strain contains a unique DNA tag ( or barcode ) sequence , that enables the pooling and competitive fitness profiling of thousands of strains in parallel [8] . Similar to individual strain assays , competitive pool assays provide a relative measure of strain fitness . Nevertheless , the use of competitive fitness assays with DNA tags has two primary advantages . The first is that genome-wide mutant collections are pooled in a single tube , thereby simplifying experimental setup , increasing throughput , and reducing issues related to strain contamination . More importantly , the tag-based pooled fitness assay provides a highly quantitative measure of strain fitness regardless of whether a microarray [18] , [19] or sequencing [20] is used to measure tag abundance . Shewanella oneidensis MR-1 is a Gram-negative γ-proteobacterium isolated from freshwater lake sediment [21] . Like most other members of the Shewanella genus , S . oneidensis MR-1 ( hereafter abbreviated MR-1 ) can use a wide variety of terminal electron acceptors , including both soluble and solid metals . As such , MR-1 has received attention for its potential roles in the bioremediation of heavy metals and energy generation via fuel cells [22] . The computationally annotated MR-1 genome contains 4 , 318 protein-coding genes on its main chromosome and an additional 149 protein-coding genes on a single megaplasmid [23] , [24] . Based on orthology relationships ( bidirectional best BLAST hits ) , MR-1 shares 1 , 639 genes ( 37% ) with the γ-proteobacterium E . coli . A total of 1 , 655 genes ( 37% ) in the MR-1 genome are annotated as hypothetical , with 83% ( 1 , 371 ) of these genes not having orthologs in E . coli . Here we describe the functional characterization of the MR-1 genome via the generation and phenotypic analysis of a large transposon mutant collection . Using a DNA tag-based pooled fitness assay , we assayed mutant fitness for 3 , 355 nonessential genes in 121 diverse metabolic , redox , stress , survival , and motility conditions . In addition to identifying phenotypes for over 2 , 000 genes , we demonstrate that mutant fitness profiles can be used to infer specific functions for genes and operons , a subset of which we confirm experimentally . Furthermore , we demonstrate that the correlation between gene expression and mutant fitness is poor in bacteria , thus underscoring the need to complement transcriptomics with mutant phenotyping . Our strain collection and fitness dataset are valuable resources for studying microbial metal reduction and for microbiology in general , given that many previously uncharacterized MR-1 genes have orthologs in diverse bacteria .
To provide a deep functional characterization of the MR-1 genome by analysis of mutant phenotypes , we expanded our previously characterized transposon library [19] by mapping an additional 17 , 301 mutants . The total MR-1 transposon collection consists of 24 , 688 archived strains and represents mutants in 3 , 447 unique genes ( Table S1 ) . Genes without a transposon insertion are potentially essential for viability . Starting from this premise , we classified 336 MR-1 genes as ExpectedEssential because their orthologs are essential in E . coli [25] , [26] and 67 NewEssential genes that were not expected from studies of E . coli , including 12 that are not orthologs of essential genes in any of 14 other bacteria ( DEG version 5 . 4 ) [27] . The 67 NewEssential genes include those of the ATP synthase complex , gluconeogenesis , biotin synthesis , and phosphate transport ( see Text S1 for full analysis; Table S2 for full essential gene list ) . To enable large-scale phenotypic analysis of the MR-1 mutant collection , we engineered our transposons to contain TagModules . A TagModule contains two unique 20 bp DNA tags ( or barcodes ) , termed the uptag and downtag , each flanked by common PCR priming sites . In a system identical to that used for the yeast deletion collection , thousands of mutant strains , each carrying a unique TagModule , can be pooled together and competitively grown in a condition of interest [9] , [19] . We calculated the fitness of each strain as the log2 ratio of the signal for the tag after growth in that condition relative to the start of the experiment ( Figure 1A ) . Negative log2 ratios indicate that the given strain has a fitness defect relative to the median strain; positive log2 ratios indicate a fitness advantage . We constructed two pools , the upPool and dnPool , for the phenotypic analysis of MR-1 mutants ( Figure 1A ) . These pools contain transposon mutants in 3 , 355 MR-1 genes ( Figure S1 ) . To assess the potential of mutant fitness to annotate gene function , we assayed the fitness of the pooled strains in 195 experiments and 121 different conditions ( see Table S3 ) . These conditions include aerobic experiments in defined minimal media with one of 26 different sole sources of carbon , 20 for nitrogen , 8 for sulfur , and 5 for phosphorous . Given the unique activity of Shewanella with respect to anaerobic respiration , we also profiled the pool fitness under anaerobic conditions with different electron acceptors including iron ( III ) citrate , manganese ( IV ) oxide , fumarate , dimethyl sulfoxide ( DMSO ) , and nitrate . Additionally , we profiled a number of stresses including metals , salts , temperature , pH , stationary phase , and heat shock . Lastly , we assayed motility using a soft agar assay . A heatmap of our entire fitness compendium is presented in Figure 1B . We used a number of tests to validate the technical and biological consistency of our mutant fitness dataset including the correlation of related strains or genes , the fitness of expected auxotrophs in rich versus minimal media , the relationship between fitness values and growth rates when grown individually , and complementation of mutant phenotypes . First , we compared the fitness for ( 1 ) identical mutant strains contained in both pools but assayed with different tags ( uptag for upPool and downtag for dnPool ) and ( 2 ) different transposon mutants for the same gene ( Figure 1C ) . For the identical mutant strains , the fitness values are highly correlated when assayed in the upPool versus the dnPool ( r = 0 . 96; Pearson correlation ) , which confirms that a single uptag or downtag of a TagModule is sufficient for measuring fitness . The fitness correlation is slightly lower ( r = 0 . 89; Pearson correlation ) for different mutants in the same gene ( Figure 1C ) . Next , we examined the cofitness ( correlation of fitness ) of operons , cotranscribed groups of genes that are often functionally related [28] . We defined the fitness of a gene as the average of relevant strain fitness values ( see Materials and Methods ) . As illustrated in Figure 1D , our data confirms the expectation that the fitness of genes within an operon are positively correlated in a single condition ( r = 0 . 73; Pearson correlation ) . For each of our 195 pool fitness experiments , we used both the operon pair fitness correlation ( as in Figure 1D ) and the fitness correlation of identical strains in the upPool and dnPool ( red triangles in Figure 1C ) as computational metrics of experimental quality ( Figure 1E ) . To validate the biological consistency of our pool fitness assay , we compared gene fitness on DL-lactate minimal media to LB rich media to confirm MR-1 auxotrophs predicted by TIGR roles [29] and flux balance analysis [30] . As illustrated in Figure S2 , most predicted auxotroph genes have severe fitness defects in minimal media and normal growth in rich media , which demonstrates that our pooled fitness results are biologically meaningful . For example , of 60 auxotrophs predicted by both TIGR and the flux balance analysis , 88% have fitness of −2 or less in minimal media , versus just 3% in LB rich media . Lastly , we compared the pooled fitness data of MR-1 grown in minimal media to single strain growth data of E . coli deletion strains grown in minimal media [11] to examine the phenotypic consistency of orthologous auxotrophic genes ( Figure S3 ) . The majority of experimentally verified MR-1 auxotrophs are also E . coli auxotrophs . However , there are a number of disagreements due primarily to redundancy ( presence of isozymes or alternative pathways ) in MR-1 but not E . coli ( glnA , metL , purA ) or vice versa ( panE , asnB , trxB , gltBD , argI ) . MR-1 auxotrophs without E . coli orthologs include diverged purC , aroQ , and folK genes and SO_3749 , a new arginine synthesis gene ( see below ) . It is important to note that fitness as used in this study reflects the abundance of tags in a pooled assay and not an absolute growth rate for each individual strain . Hence , our pooled fitness assay does not distinguish whether less fit strains are due to slower doubling times or longer lag phases . To test this , we measured growth for 48 individual mutant strains in DL-lactate minimal media . As illustrated in Figure 1F , we observe a positive correlation ( r = 0 . 73; Spearman correlation ) between pool relative fitness values and single strain growth rates . This highlights the quantitative nature of our competitive growth assay , as previously observed in yeast [18] . Additionally , most mutants with fitness defects in the pool assay are due to slower doubling times rather than extended lag phases ( Figure 1F ) . To verify that our mutant phenotypes are caused by a single transposon insertion , we complemented the fitness defect of 10 genes , 7 of which are annotated as hypothetical ( Figure 2 ) . In each instance , we complemented the mutant phenotype by introducing an intact copy of the gene on a plasmid . These results demonstrate that poorly characterized genes have phenotypes in a diverse range of conditions and that a single transposon insertion was responsible for the fitness defect . Of note , SO_3257 is adjacent to a large flagellar/chemotaxis gene cluster in MR-1 and is a distant homolog of Vibrio cholerae flgO [31] ( 24% identity but at a conserved location ) ( Figure 2A ) . SO_0274 ( ppc ) encodes phosphoenolpyruvate carboxylase and is absolutely required for growth on minimal media with DL-lactate as a carbon source ( Figure 2D ) . Interestingly , a flux balance model predicts that the loss of ppc will lead to only a 5% reduction in growth rate [30] . This discrepancy between a model prediction and an experimental result highlights the utility of fitness profiling to discover unexpected roles for characterized enzymes . Overall , these complementation results and the fitness agreement of independent mutants in the same gene ( Figure 1B ) suggest that secondary mutations are not a significant factor in our experiments , allowing us to use pooled fitness results as a proxy for single mutant phenotypes . Lastly , we examined the potential role of polarity in our fitness dataset . Polarity , whereby a mutation in an upstream gene in an operon causes a loss of expression of downstream genes , is a concern in all transposon mutant studies . To address this issue , we looked for instances in which only the upstream or downstream gene of an operon pair has a strong fitness defect ( fitness <−2 vs . >−1 ) . If there are strong polar effects , then we should rarely see cases where only the downstream gene in the pair is sick , because mutating the upstream gene should usually impair the downstream gene , but not vice versa . Across our entire fitness compendium , upstream-only sick occurs only about 50% more often than downstream-only sick ( 3 , 333 vs . 2 , 233 cases , P<1e-15 , binomial test ) suggesting that most within operon fitness correlations reflect similar biological roles and are not simply an artifact of polarity . To further support the notion that polarity is not a dominant factor in our results , we successfully complemented the fitness defects of 4 upstream genes in operons ( SO_0887 , SO_1333 , SO_3257 , SO_4485; Figure 2 and Table 1 ) . Our fitness dataset provides an opportunity to explore general principles related to the phenotypic importance of single genes across a large number of diverse conditions . First , we sought to determine how many genes have at least one phenotype ( positive or negative ) . We used the 14 control experiments ( independent pool recovery experiments from the freezer or “start” in Figure 1A; grey bar in Figure 1B , black circles in Figure 1E ) to transform a test statistic , which quantifies the consistency of the various measurements of a gene's fitness , to a Z score ( see Materials and Methods for details ) . In other words , the Z scores from the control experiments follow the standard normal distribution and Z scores in the other experiments indicate their level of statistical significance . To test whether a gene's pattern of fitness is consistent with it having no phenotype , we used a chi-squared test to combine the Z scores . By combining the Z scores from all experiments , we were able to detect significant fitness patterns for genes with mild phenotypes in many conditions . At P<0 . 001 ( chi-squared test ) , we find that 2 , 350 genes have a statistically significant fitness pattern . Therefore , 70% of the MR-1 nonessential genome ( 2 , 350 genes with a phenotype out of 3 , 355 assayed ) has a phenotypic consequence when disrupted by a transposon . Nevertheless , many of these fitness patterns are weak and it is difficult to place these genes into pathways because the phenotypes are too subtle . To identify those genes with a strong fitness pattern , we used two approaches . First , we used an arbitrary cutoff and selected genes within the top one-third of chi-squared values as having strong fitness patterns . As shown in Figure 3A , these genes with higher chi-squared values tend to show greater cofitness ( correlation of fitness ) with other genes in their operons . Second , as the chi-squared test may miss genes that have a fitness defect in only one condition , we also considered genes with highly significant defects in one condition ( P<0 . 01 after Bonferonni correction , corresponding to Z<−3 . 88 ) . Together , these two tests gave us 1 , 230 genes with strong patterns , and the remainder of the section will focus on these genes . We next explored the properties of these 1 , 230 genes with a strong fitness pattern . When examined in the context of COG function codes [32] , all classes of genes have significant phenotypes ( Figure 3B ) . However , genes in characterized COG families are rather more likely to have a phenotype than genes with only general predictions or unknown roles ( code R or S in Figure 3B ) . In addition , genes involved in “replication , recombination , and repair” ( code L ) are also less likely to have a phenotype , which may reflect the choice of conditions that we profiled or genetic buffering ( redundancy ) in these pathways . To address redundancy in a more systematic way across our entire dataset , we asked whether genes that are similar to other genes in MR-1 ( paralogs ) have phenotypes . We find that genes with a highly similar paralog ( over 50% amino acid identity ) are slightly less likely to have a phenotype than unique genes ( Figure 3B ) . This suggests that paralogs often function under specific conditions and therefore do not always provide functional redundancy . This finding supports recent observations that most c-type cytochromes in MR-1 , despite frequent sequence similarity to one another , have a detectable phenotype when deleted [33] . Lastly , “core” hypothetical genes ( without a known family ) that are conserved across diverse Shewanella genomes are more likely to have a phenotype than other hypotheticals ( Figure 3B ) . This suggests that poorly characterized core genes , which partly define what it means to be in the Shewanella genus [34] , are functionally more important than other poorly characterized genes . For the 1 , 230 genes with a strong fitness pattern , we determined how many experiments resulted in a significant phenotype . We find that most of these genes have a clear phenotype in a small number of experiments ( Figure 3C ) . The median of these genes has a fitness defect in 9 experiments and positive fitness in 1 experiment . By contrast , some genes have strong fitness patterns in a large number of experiments . One example is the genes of the general secretory pathway , as illustrated in Figure 1B . The individual general secretory pathway genes have a complex , highly correlated fitness pattern and their absence leads to strong fitness defects or positive fitness in the majority of tested conditions . One of the main aims of this study was to use mutant fitness to annotate gene function . Using the approaches described below , we predicted specific functional annotations for 40 genes/operons with either poor or incomplete annotations including 17 enzymes , 10 transporters/efflux pumps , 4 transcriptional regulators/signaling proteins , 4 electron transport proteins , and 5 other proteins ( Table 1 for list; Text S2 for detailed rationale ) . Even in retrospect , few of these annotations could have been made by homology alone . First , we identified those genes with a strong fitness defect in only one or a few conditions , as an annotation of a specific function in these instances is often easier than if a gene has pleiotropic effects . Second , we identified groups of genes with high cofitness across the entire compendium . These genes are more likely to be functionally related and allow for functional annotation if one or more of the genes in the cofitness cluster are characterized [7] . Given these genes and their phenotypes , we used comparative genomics and prior experimental data from MR-1 and other bacteria to predict specific functions . We split our evidence-based annotations into three categories , “new” , “expanded” , and “confirmed” , to reflect prior knowledge and genes that have multiple functions . Selected examples of evidence-based annotations and their experimental validation are described below . Some of our gene annotations reflect specific new functions for hypothetical genes . For example , the MR-1 genome does not have an annotated enzyme for synthesizing ornithine from N-acetyl-ornithine , a necessary step in arginine biosynthesis . Given that MR-1 is capable of synthesizing arginine , we sought to identify the missing enzyme . The N-acetyl-ornithine to ornithine reaction is encoded in E . coli by argE , an N-acetyl-ornithine deacetylase , and in B . subtilis by argJ , an ornithine acetyltransferase . Given that MR-1 does not have homologs to either argE or argJ , we examined our fitness data for uncharacterized genes with high cofitness with known arginine biosynthesis genes to identify the missing enzyme . Using this approach , we identified high cofitness between mutants in the hypothetical gene SO_3749 and mutants in other arginine biosynthesis genes ( Figure 4A ) . The outlier in Figure 4A , argD , participates in other metabolic processes including lysine biosynthesis that may contribute to its complicated fitness pattern . SO_3749 contains a hydrolase domain , so we predict that its activity resembles that of ArgE rather than the transacetylase ArgJ . To verify that SO_3749 encodes the missing step in arginine biosynthesis , we performed cross-species complementation assays . First , we complemented the minimal media growth deficiency of a MR-1 SO_3749 transposon mutant with the E . coli argE gene ( Figure 4B ) . In addition , we complemented the growth defect of an E . coli argE deletion strain with a plasmid expressing the SO_3749 gene ( Figure 4C ) . Despite lacking any detectable homology to E . coli argE , our results demonstrate that the hypothetical gene SO_3749 encodes a functional N-acetyl-ornithine deacetylase ( or ornithine acetyltransferase ) . We next used our fitness compendium to uncover new two-component signal transduction pathways . In MR-1 , as in many bacteria , the majority of histidine kinases ( HK ) and their cognate response regulators ( RR ) are cotranscribed in operons [35] . However , there are often instances of “orphan” HKs and RRs in the genome for which the cognate partners are unknown . Using cofitness as a metric for functional interactions , we find that two orphan HK-RR pairs have highly correlated fitness . In the first example , hybrid HK SO_3457 and RR SO_1860 have high cofitness ( r = 0 . 83; Pearson correlation ) and are required for motility , growth in high and low pH , and growth on a large number of carbon and nitrogen substrates . The E . coli orthologs of these genes , barA and uvrY , form a two-component system that ultimately regulates the global regulator CsrA [36] . Gene expression experiments with SO_1860 and SO_3457 mutant strains suggest that CsrA is not a target of this two-component system in Shewanella ( data not shown ) demonstrating that this system may regulate different genes . There inferences were confirmed by a very recent report on barA and uvrY in MR-1 [37] . In a second example , we find that HK SO_2742 and RR SO_2648 have high cofitness ( r = 0 . 85; Pearson correlation ) and are required for optimal growth in minimal media with acetate , propionate , or butyrate as carbon sources ( Figure 5A ) . To further demonstrate a functional relationship between SO_2742 and SO_2648 and to look for target genes of the RR , we assayed transcript levels in mutants of both the HK and RR after transfer to minimal media with acetate as a carbon source . We found that the expression of genes in these mutants is highly correlated further suggesting a direct biochemical interaction between the HK and RR ( Figure 5B ) . Lastly , we identify acs ( SO_2743; encoding acetyl-coA synthetase ) , which is divergently transcribed from the adjacent SO_2742 , as significantly downregulated in both mutants ( Figure 5C ) . In E . coli , acs activates acetate to acetyl-CoA and an acs mutant grows poorly in acetate containing media [38] . Based on this data , we propose that the MR-1 two-component system SO_2742 and SO_2648 senses carboxylates and that one of its primary targets is acs . A combination of mutant fitness profiles and comparative genomics is a powerful method for predicting gene function . For instance , mutants in the hypothetical gene SO_1913 have high co-fitness with mutants in the general secretory pathway ( Figure 1B ) . To infer the potential role of SO_1913 in the general secretory pathway , we looked for previous experimental evidence from homologs in other species . Using this approach , we identified a positional ortholog in Shewanella benthica KT99 ( KT99_05357 ) that is similar to the type III secretion chaperone yscW ( PF09619 ) [39] . Additionally , we found that some homologs of SO_1913 in other species are fused with an uncharacterized meta/hslJ domain ( for example , Lferr_0115 from Acidithiobacillus ferrooxidans ATCC 53993 ) that is predicted to be associated with heat shock ( and hence chaperone ) proteins . Based on our mutant fitness data and the comparative analysis , we propose that SO_1913 is a chaperone for the general secretory pathway . In a second example , we identified two conserved hypothetical genes ( SO_3259-SO_3260 ) that are required for motility . These genes are located adjacent to a large flagellar/chemotaxis gene cluster suggesting that SO_3259-SO_3260 may play a direct role in motility . To identity a putative function for SO_3259-SO_3260 , we analyzed homologs with experimental evidence in other species . We find that SO_3259 is similar to the flagellar modification genes pseD and pseE from Campylobacter jejuni , that are involved in decorating the flagellum with sugars and are required for full motility [40] . Therefore , we propose that the SO_3259 and SO_3260 participate in the modification of the flagellum by some sugar and that this modification is required for motility . In a final example , we find that the previously hypothetical gene SO_0444 is required for growth in ZnSO4 or CuCl2 stress conditions . SO_0444 is in an operon with and predicted to be regulated by SO_0443 [41] , a putative ortholog of the E . coli zinc-responsive transcriptional regulator ZntR [42] . Given that SO_0444 is predicted to be a membrane protein , we propose that this previously hypothetical gene encodes a copper/zinc efflux protein . In addition to new annotations , analysis of mutant fitness profiles can directly confirm previous predictions of gene function and also expand our current understanding of a single gene's activity . The operon SO_1427-SO_1432 encodes a DMSO reductase that functions under anaerobic conditions [43] . In addition to its essential role in DMSO reduction , we find that mutants in SO_1427-SO_1432 are also impaired in their ability to use manganese oxide as a terminal electron acceptor , thus expanding the known substrate range of this reductase . Given that MR-1 was isolated under manganese reducing conditions [21] , the finding that a DMSO reductase plays a role in manganese oxide reduction is particularly interesting . Whole-genome gene expression profiling using microarrays is standard practice in microbiology . Typically , these studies are designed to detect differential expression between two conditions ( i . e . , treatment versus control or mutant versus wild-type ) in an effort to identify those genes whose expression is under regulation ( for example , see [44] ) . The expectation in these experiments is often that genes with differential expression are more likely to be functionally important ( and hence have a fitness consequence when mutated ) . However , the extent to which this is true , that is the correlation between differential gene expression and mutant fitness , has to our knowledge not been systematically investigated in bacteria . To address this issue we assayed transcript levels for wild-type MR-1 grown in rich media ( LB ) and in minimal media with DL-lactate , acetate , or N-acetyl-glucosamine ( NAG ) as carbon sources and compared differential expression to differential fitness values obtained from identical growth conditions . The genome-wide correlation between mutant fitness and gene expression , while statistically significant , is weak ( Figure 6 ) . The only genes that “make sense” , which we define as those that are upregulated in condition A relative to condition B and also important for fitness in condition A relative to condition B , are a few key genes that directly contribute to the utilization of the carbon substrate . For instance , genes of the NAG utilization operon are important for fitness and are upregulated in NAG-containing media ( Figure 6B ) . However , in every comparison , there are genes with fitness defects when mutated in a single condition whose expression is not differentially regulated under the same condition . Conversely , in all comparisons there are many genes that are differentially expressed but have no fitness consequence . Our findings are similar to those in yeast where the correlation between mutant fitness and gene expression is also weak [8] , [45] . Therefore , the lack of correlation between gene expression and mutant fitness is a general trend . Theories that might explain the lack of correlation include standby expression [46] , anticipatory control [47] , [48] , suboptimal control of recently acquired genes [49] , or post-transcriptional control . Given the extensive use of gene expression profiling in bacteria and the relative lack of large-scale mutant fitness studies , it is important to examine the information content of both assays . In particular , for a single organism , we compared the ability of large expression and mutant fitness datasets to predict gene function and gene regulation . For the mutant fitness dataset , we used the data from the 195 experiments described in this paper . For gene expression , we used the data from 371 experiments contained in the MicrobesOnline website [50] . These gene expression experiments are derived from this study and others ( for example; [51] , [52] ) and represent a diverse range of conditions comparable to what we describe here for mutant fitness . For gene function predictions , we tested the ability of the large-scale expression and/or mutant fitness datasets to place genes into pathways . Of 3 , 247 genes that were in both datasets , the TIGRFam database assigned 618 genes to 79 different “subroles” such as “Amino acid synthesis: Aspartate family” by homology [29] . To classify the genes into subroles from the data , we used a standard machine learning technique called random forests that is resistant to overfitting and that estimates the confidence of its predictions . We trained the classifier using the 618 genes that have assigned subroles , and then predicted the TIGR subroles for all 3 , 247 genes given the data ( We used 10-fold operon-wise cross-validation to avoid overfitting the genes with known subroles , see Materials and Methods ) . For the 618 genes with known subroles , most of the predicted subroles did not match TIGRFam annotations , but predictions with confidence values above 50% were likely to be correct ( Figure 7 ) . We find that gene expression gives more correct gene function predictions than mutant fitness but fitness gives more high-confidence predictions ( Figure 7A and 7B ) . Furthermore , combining the fitness and expression datasets together did not increase the confidence of the predictions based on mutant fitness alone ( Figure 7C ) . The higher confidence of mutant fitness-based predictions confirms our intuition that mutant phentotypes give more direct information about gene function than gene expression patterns do . To determine the ability of gene expression and mutant fitness to predict gene regulation , we examined the expression and fitness correlation of MR-1 transcription factor-target gene pairs from the RegPrecise database [41] . We find that transcription factor-target gene pairs tend to have higher coexpression than cofitness correlations , relative to shuffled gene pairs ( Figure 8 ) . Interestingly , pairs with high correlation in one dataset tend not be highly correlated in the other ( Pearson correlation r = 0 . 09; even this correlation disappears if co-transcribed transcription factor-gene pairs are excluded from the analysis ) . Thus , the two datatypes , expression and fitness , should be complementary for analyzing gene regulation . To illustrate this point , we examined our fitness dataset and identified strong cofitness ( r = 0 . 56; Pearson correlation ) between the uncharacterized transcription factor SO_1916 and its divergently transcribed neighbor gene SO_1917 , a putative efflux pump . Conversely , the expression correlation between SO_1916 and SO_1917 is weak ( r = 0 . 13; Pearson correlation ) . SO_1916 and SO_1917 are required for optimal growth under certain anaerobic conditions with DMSO as an electron acceptor ( Figure 2E ) . To demonstrate that SO_1916 regulates SO_1917 , we performed gene expression experiments on SO_1916 mutants after transfer to anaerobic conditions with DMSO as an electron acceptor ( Figure S4 ) . Relative to wild-type , we found a DMSO-specific down regulation of SO_1917 in two independent SO_1916 mutant strains ( Figure S4 ) . These data suggest that SO_1916 is a neighbor regulator that activates SO_1917 expression and that the activity of the efflux pump encoded by SO_1917 is necessary when DMSO is an electron acceptor . Overall , we conclude that large-scale gene expression and mutant fitness datasets provide complementary information . Mutant fitness gives higher confidence predictions of gene function and is better suited to annotating the function of genes with a low false-positive rate ( Figure 7 ) . Conversely , the analysis of expression datasets is a better methodology for elucidating gene regulation ( Figure 8 ) . However , it is important to note that fitness and expression datasets are not limited to gene function annotation and gene regulation , respectively . Gene expression is an established method for predicting gene function [53] albeit at lower confidence than mutant fitness ( Figure 7B ) . Conversely , as described above for SO_1916 , mutant fitness can lead to regulatory insights that may be missed by expression profiling alone .
The systematic determination of gene function across the diversity of bacteria is a major challenge in microbiology . Previous studies have demonstrated the utility of high-throughput mutagenesis and phenotyping strategies to annotate bacterial gene function [10] , [54] . Nevertheless , these studies were only able to assign specific functions to a small number of genes . To demonstrate the utility of mutant fitness for annotating gene function on a larger scale in a non-model bacterium , we described the construction of a near-complete set of archived mutants in Shewanella oneidensis MR-1 and the fitness profiling of this collection in over 100 diverse conditions . The entire fitness dataset is available on the MicrobesOnline website [50] and for download ( http://genomics . lbl . gov/supplemental/MR1fitness2011/ ) . In addition to identifying a strong phenotype pattern for 1 , 230 genes , of which 627 ( 51% ) lack an E . coli ortholog and 282 ( 23% ) are annotated as hypothetical , we used gene fitness profiles to derive ‘evidence-based’ gene annotations for 40 genes/operons and we verified some of these functions experimentally . On average , each reannotated gene has a tree-ortholog in 97 ( 5% ) of 1 , 828 other bacterial genomes in MicrobesOnline [50] . Indeed , most bacterial genomes ( 73% ) contain at least one tree-ortholog of these genes , which illustrates that our mutant fitness data and the evidence-based annotations are relevant to most bacteria . The approach presented here for MR-1 uses artificial DNA tags engineered into transposons and a pooled growth assay in order to determine relative fitness of each mutant strain . DNA tags , as best exemplified in yeast and in this study , are a powerful method for generating quantitative fitness data using a simple experimental assay and sample-processing step [55] . Nevertheless , the approach presented here requires archived strains and the mutant pool size is limited to the number of unique TagModules that are available . Archiving the mutant strains is beneficial for follow-up studies , in particular when an efficient system for constructing targeted mutations , such as recombineering in E . coli [56] , is not available . Archived strains also enable additional unpooled assays , for example studies of protein localization [57] or metabolomics [58] . However , given the number of poorly characterized microbial species and the continuing drop in sequencing costs , it is likely that future fitness datasets will be generated using pool-based approaches that do not require strain archiving such as HITS [59] , TraDIS [60] , Tn-seq [61] , [62] , and TRMR [63] . In addition to pool-based approaches described above , the high-throughput imaging of clonal mutants has also been used to generate a large-scale gene-phenotype map in bacteria , as recently demonstrated in E . coli [10] . To compare the two methodologies , we examined the operon fitness correlations from our tag-based MR-1 dataset and the colony size-based E . coli dataset . We find that the operon fitness correlations are significantly better for the MR-1 compared to the E . coli dataset , even for matched conditions ( Figure S5 ) . Given that polar effects should be similar in both assays ( a dominant drug marker marks both the targeted E . coli deletions and the MR-1 transposon insertions ) , we conclude that the DNA tag-based assay is more quantitative and better suited for identifying small fitness defects . However , we found that some MR-1 pool experiments ( for example , succinate as a carbon source ) consistently gave low-quality fitness data by our metrics , presumably because a handful of strains take over the population and skew the relative fitness values ( data not shown ) . A similar phenomenon has been observed in pooled transposon mutant studies in E . coli [64] , suggesting that certain conditions are best assayed as single mutants rather than pool-based assays . Regardless of which method is used to generate large gene-phenotype maps in bacteria , the challenges associated with this data are common . Foremost , most genes in the genome either have a weak phenotype or no phenotype at all and therefore predicting functions using mutant fitness patterns is not possible . Given that strong phenotypes are easier to assay and interpret , one pressing question is what is necessary to identify strong phenotypes for all genes in a bacterial genome ? One option is to profile the single gene mutation collection under a more diverse set of laboratory conditions , including a large number of chemical inhibitors with different modes of action , as in yeast [17] and E . coli [10] . An alternative is to assay the mutant collection in more ecologically relevant conditions under the hypothesis that some genes may perform environment-specific functions ( for instance , cell-cell communication with a different species ) that are difficult to recapitulate in the laboratory . To lend credence to this hypothesis , a study looking at the survival of MR-1 mutants in sediment identified phenotypes for some genes that we failed to find a phenotype for using our more standard laboratory assays [65] . A third option is to systematically construct double mutations , similar to that described in E . coli [66] , [67] , under the hypothesis that bacterial species have functionally redundant genes and pathways . Each of the above approaches assumes that each gene actually has a functional consequence to the cell . However , a microbial genome is a snapshot in evolutionary time and some genes are under weak selection and are in the process of being lost [68] . It is unlikely that a strong phenotype , if any at all , will be found for all of these genes . Studies such as ours are part of a systematic effort to move from sequencing bacterial genomes to understanding the function of all genes . However , mutant fitness as described in this study is only one piece of experimental evidence for predicting gene function . Furthermore , proving mutant fitness-based annotations requires additional investigation including metabolite and enzymatic assays at the level of single genes . To improve and extend gene annotations to a greater percentage of the genome , it is clear that additional pieces of evidence such as protein-protein interactions [69] , [70] , biochemical activity , and metabolomics [71] will be necessary to supplement the existing large-scale mutant fitness and gene expression datasets , which are currently easier to generate and are sure to proliferate in the near future . The integration of diverse data types for a single bacterium should provide insight into the function for many uncharacterized bacterial genes for which we have strong phenotypes but no functional predictions . Ultimately , the functional annotation of a greater number of sequenced microbial genes promises to aid future efforts in drug discovery , bioengineering , and biotechnology .
S . oneidensis MR-1 was purchased from ATCC ( catalog number 700550 ) . E . coli conjugation donor strain WM3064 was a gift of William Metcalf ( U . of Illinois ) . The E . coli ΔargE strain was obtained from the KEIO collection [11] . See Table S4 for the strains used in this study . All strains were commonly cultured in Luria-Bertani broth ( LB ) with appropriate antibiotic selection; kanamycin ( 50 µg/ml ) for MR-1 transposon mutants and for the E . coli ΔargE strain , and gentamicin ( 15 µg/ml ) for complementation strains containing plasmid pBBR1-MCS5 . To grow the diaminopimelic acid ( dap ) auxotroph WM3064 , dap was added to the media at a final concentration of 300 µM . Our standard MR-1 minimal media contained salts ( per liter: 1 . 5 g NH4Cl , 0 . 1 g KCl , 1 . 75 g NaCl , 0 . 61 g MgCl2-6H20 , 0 . 6 g NaH2PO4 ) , 30 mM PIPES buffer , 20 mM DL-lactate , Wolfe's vitamins , and Wolfe's minerals . For anaerobic minimal media , we added one of the following electron acceptors: fumarate ( 30 mM ) , dimethyl sulfoxide ( 20 mM ) , iron ( III ) citrate ( 10 mM ) , manganese ( IV ) oxide ( 30 mM ) , trimethylamine N-oxide ( 10 mM ) , nitrate ( 5 mM ) , or cobalt ( III ) -EDTA ( 5 mM ) . For anaerobic experiments , manganese oxide [72] and cobalt ( III ) -EDTA [73] were prepared as described . For pool experiments with alternative nutrient sources , we replaced the DL-lactate with a different carbon source , the NH4Cl with a different nitrogen source , both with a single carbon/nitrogen source , or the NaH2PO4 with an alternative phosphorous source . For alternative sulfur sources , we replaced all sulfate containing minerals in the Wolfe's mineral mixture with non-sulfur containing versions such that the added sulfur source served as the sole source . MR-1 was typically grown at 30°C; E . coli was grown at 37°C . We previously reported the generation and preliminary analysis of a library of 7 , 387 transposon insertion mutants in MR-1 [19] . To achieve greater coverage of the genome , we mapped an additional 17 , 301 mutants using a two-step arbitrary PCR and sequencing protocol , as described [19] . Briefly , each TagModule contains two unique 20 bp DNA sequences , the uptag and downtag , each flanked by common PCR priming sites . The TagModules are cloned into a Gateway entry vector and can be readily transferred to any Gateway compatible destination vector via the LR clonase reaction ( Invitrogen ) . We transferred the TagModules into two different transposon vectors that are active in MR-1 , the Tn5-based pRL27 [74] and mariner-based pMiniHimar_RB1 [75] . The tagged transposons were introduced into MR-1 by conjugation with an E . coli WM3064 donor strain carrying the appropriate suicide vector . We used a two-step arbitrary PCR and sequencing protocol to simultaneously map the transposon insertion location and identify the TagModule . All mutants were stored as glycerol stocks in either 96-well or 384-well plates . Overall , we observe transposon insertion biases both on the main 5 MB chromosome and on the 161 kB megaplasmid ( Figure S6 ) . Analysis of the insertion preferences for Tn5 and mariner indicates that their insertion biases are not equal ( Figure S7 ) , thus illustrating the benefit of using multiple transposons to achieve maximal coverage . Additionally , we observed a greater than 4-fold increase in mapped megaplasmid insertions relative to the expectation ( based on size ) . Given that the megaplasmid appears to be equal in copy number to the main chromosome ( data not shown ) , we speculate that the megaplasmid is more accessible to transposon mutagenesis . The total MR-1 transposon collection consists of 24 , 688 archived strains and represents mutants in 3 , 447 unique genes ( Table S1 for full list ) . We constructed two mutant pools , the upPool and dnPool , by mixing equal volumes of overnight LB cultures for each strain . These pools were designed such that each strain's TagModule is unique within that pool and to achieve maximal coverage of the genome . The upPool contains 4 , 058 strains whose tags are detected at a threshold ∼5× over background in representative start hybridizations . Conversely , the dnPool has 3 , 977 strains that meet the same detection criteria . A total of 165 strains across both pools ( ∼2% of the overall number of strains that we attempted to pool ) were not detected at 5× over background and fitness values were not calculated for these strains . The 2% undetected strains are primarily due to sample tracking errors , slow growth of the mutant strain during the process of pool construction , and mutations in the TagModule ( data not shown ) . Taking into account only those strains that are detectable , our two pools contain 5 , 680 unique transposon mutants ( 2 , 420 mutant strains are in both pools ) and represent transposon insertions in 3 , 345 unique protein-coding genes . For 1 , 675 genes , two or more independent mutants are contained in the pools ( Figure S1 ) . For most pooled fitness experiments , we performed a single experiment on each of the upPool and dnPool . Individual aliquots of each pool were frozen at −80°C in glycerol ( 10% v/v ) . Prior to initiating a pool experiment , a single freezer aliquot of each pool was grown in LB aerobically at 30°C to mid log phase ( OD600 = ∼2 . 0 ) . At this point , we collected a sample ( ∼1×109 cells ) that we term the “start” . The start sample represents the time 0 of the experiment and is the control experiment that we compare all of our growth conditions to . The same recovered cells were typically used to inoculate the “condition” media at a starting OD600 of 0 . 01 or 0 . 02 . For standard liquid media conditions , we typically collected condition samples after the cultures reached saturated growth , representing between 3 and 9 population doublings . Some conditions , such as LB , reach a high density and have more population doublings than certain minimal media conditions , such as butyrate as the sole source of carbon . Liquid growth pool experiments were done in a number of formats . Aerobic minimal media experiments were done in 10 mL volumes with shaking at 200 rpm . Anaerobic experiments were conducted in hungate tubes with shaking at 200 rpm . Anaerobic experiments were set up in an anaerobic chamber ( Coy ) with a gas mix of 5% H2 , 10% CO2 , and 85% N2 . Certain stress experiments in LB were performed in 1 mL volumes in the wells of a 24-well microplate . For these experiments , the microplate was grown in a Tecan Infinite F200 reader to measure the amount of growth inhibition caused by the stress . Our target stress concentration resulted in a ∼50% reduction of the growth rate . For swimming motility experiments , we pipetted ∼1×108 cells from the start culture into the matrix of an LB soft agar plate ( 0 . 25% w/v agar ) and incubated the plate at 30°C . After 1 or 2 days , we removed cells from the outer ring ( i . e . the motile cells ) using a sterile razor . For heat shock survival experiments , we incubated the aliquots of the start cells in a 42°C water bath for different amounts of time . After incubation , we used some of the cells for measuring viability by serial dilution and plating on LB . The remainder of the cells was used to inoculate a fresh tube of LB that was grown overnight ( to avoid the possibility of detecting tags from dead cells ) . The tags from the overnight sample were hybridized and used as a measure of cell survival after heat shock . For cold survival experiments , we used the same method as for heat shock except we incubated the cells at 4°C rather than 42°C . We followed a similar method for stationary phase survival . However , in this instance , we left the pools in LB at 30°C for days after the cultures reached saturation . Again , we used some of the cells for determining viability by plating on LB plates; additional cells were used to inoculate fresh LB media . The overnight growth of these fresh cultures was used for tag array hybridization and serve as a measure of mutant survival in stationary phase . Further details on the media used and the growth conditions for each of the 195 pool experiments are contained in Table S3 . Plots showing the survival of MR-1 cells after heat shock , cold incubation , and stationary phase are contained in Figure S8 . Genomic DNA was isolated from each sample using either the DNeasy blood tissue kit with optional RNase treatment ( Qiagen ) or with a QIAxtractor genomic DNA robot ( Qiagen ) . Approximately 100 ng of genomic DNA was used as a template to amplify the uptags from the upPool samples and the downtags from the dnPool samples using previously described primers and PCR conditions [55] . We combined uptag and downtag PCR products ( 10 µL of each ) and hybridized to a single GenFlex 16K_v2 microarray ( Affymetrix ) that contains the tag complement sequences . Microarrays were hybridized , washed , labeled , and scanned as described [55] . Using Affymetrix . CEL files as a starting point , we first averaged the log2 intensities across the 5 replicate probes for each tag to obtain values for each uptag and downtag . We computed the difference ( the log ratio ) between these values for the condition array and the start array , to give a fitness value for each strain . Sometimes we used an average of start arrays from other experiments instead of hybridizing a start array from that actual experiment; log-levels in these independent start experiments were highly correlated ( r> = 0 . 95 ) . We removed strains with the lowest 2% of levels in this average start array from the analysis . We normalized the fitness values for the strains so that the median fitness for each pool and for each chromosome ( the main chromosome and the megaplasmid ) was zero . After this normalization , some of our experiments showed significant effects based on which 96-well plate the strain had been grown in while we were preparing our pools . So , for all experiments , we also set the median fitness of each of these groups of 96 strains to zero . See Table S5 for all strain fitness data . We computed fitness values for each gene by averaging the fitness values for all of the insertions in that gene . If a gene had one or more “good” insertions ( an insertion within the central 5–80% portion of the gene ) , then we used only those good insertions to compute the average . See Table S6 for all gene fitness data . To estimate the reliability of each fitness value , we took advantage of our 14 control experiments ( measurements of the start pools after independent recoveries from the freezer ) and the fact that we have more than one fitness measurement for most genes ( i . e . , more than one strain , or the single strain for the gene is in both pools ) . We used an approach similar to that of Efron et al . [76] . We first computed a t-like test statistic , which was:where x are the measurement ( s ) for the gene , μ is their average , n is the number of measurements , and Ψ = median ( STD ( x ) ) , that is , the median across all genes with more than one measurement of the standard deviation of that gene's measurements . We transformed our test statistic into Z scores that follow a normal distribution in the absence of biological signal by using the control experiments; we transformed the distribution separately for genes with 1 , 2 , 3 , or > = 4 measurements . We used conditions that we had repeated to verify that these Z scores were appropriate ( i . e . , the rate of discordant outliers with high |Z| values was about the same as expected by chance; data not shown ) . See Table S7 for Z score data . Aerobic single strain growth assays were performed in 96-well microplates in a Tecan Sunrise plate reader at 30°C with readings every 15 minutes . Anaerobic single strain growth assays were performed in a DTX880 plate reader ( Beckman ) housed in an anaerobic chamber ( Coy ) at 30°C with readings every 30 minutes . All microplate growth assays contained 150 µL per well at a starting OD600 of 0 . 02 . Before all single strain growth assays , we isolated a single colony of the mutant strain and confirmed the expected location of the transposon insertion by PCR with a transposon specific primer and a genome primer . We calculated doubling times for growth curves using a logistic algorithm implemented in R . To calculate the relative growth rate for transposon mutants in Figure 1C , we divided the doubling time of the transposon mutant by the doubling time of wild-type MR-1 . All single strain growth assays were performed a minimum of three times . We measured gene expression in wild-type MR-1 and in single transposon mutant strains . Cells were harvested after RNAprotect treatment ( Qiagen ) in either early exponential growth or one hour after transfer to an experimental media . For the transfer experiments , all cultures were initially cultured in DL-lactate minimal media to early exponential phase prior to transfer to the experimental media . For all experiments , we collected ∼2×109 cells , isolated total RNA with a RNeasy mini kit ( Qiagen ) , and synthesized Alexa Fluor 555 labeled cDNA with the SuperScript Plus Indirect cDNA Labeling Module ( Invitrogen ) . Labeled cDNA was hybridized to custom Nimblegen oligonucleotide microarrays according to the manufacturer's instructions . We used two gene expression microarray designs , a 4-plex microarray that allows hybridizing 4 samples to different regions of a single slide , and also a 12-plex microarray . The 4-plex microarray had 66 , 228 probes , which reduced to 61 , 178 after removing potential cross-hybridizing probes with BLAT , or roughly 15 probes per gene . The 12-plex microarray had 42 , 598 probes , which reduced to 40 , 881 after removal of cross-hybridizing probes . After removing potential cross-hybridizing probes , we made the distribution of the condition match that of the control ( i . e . , using quantile normalization ) , used local regression ( lowess ) to eliminate any bias in the log2 ratio by probe intensity , and set the median normalized log2 levels of the probes to zero ( separately for each scaffold ) . The final log2 ratio for each gene was the average of these normalized values for its probes; we removed values for genes with less than 4 measurements . These data are available on MicrobesOnline . To verify the quality of each experiment , we checked the correlation of log ratios between adjacent genes that are predicted to be in the same operon and the average absolute difference within these pairs . Most comparisons had operon correlations above 0 . 8; some experiments had lower correlations but also had low average differences , indicating that the correlation was low because there was less biological signal to detect . To complement the mutant phenotypes of single transposon mutant strains , we introduced an intact copy of the mutated gene on the broad-range plasmid pBBR1MCS5 , which contains a gentamicin resistance marker [77] . If possible , we used the native promoter of the gene to drive expression . Otherwise , we relied on the activity of the pBBR1MCS5 lac promoter to express the complementation gene . The complementation plasmids were constructed with circular polymerase extension cloning ( CPEC ) [78] using primers 5′-GCTCTAGAACTAGTGGATCCCCC ( N ) and 5′- GATATCGAATTCCTGCAGCCC ( N ) where ( N ) represents genome-specific primer sequences used to amplify the gene from genomic DNA . The underlined regions represent sequences homologous to the pBBR1MCS5 backbone . To prepare the vector for CPEC , we amplified pBBR1MCS5 with primers 5′-GGGCTGCAGGAATTCGATATC and 5′- GGGGGATCCACTAGTTCTAGAGC . Following amplification , the template vector was digested with DpnI . All PCR and CPEC reactions were performed with Phusion high fidelity DNA polymerase ( New England Biolabs ) . Both the vector and insert were gel purified with the Zymoclean Gel Recovery Kit ( Zymo Research ) . The CPEC reaction consisted of 50 ng of linearized pBBR1MCS5 , 25 ng of the complementation gene PCR product , and was cycled with the following protocol: initial denaturation for 15 seconds at 98°C , 4 cycles of 98°C for 30 seconds , 55°C for 30 seconds , and 72°C for 3 . 5 minutes , and a final extension at 72°C for 3 minutes . The CPEC reaction was transformed into chemically competent TOP10 cells ( Invitrogen ) and plated on LB with gentamicin . Complementation constructs were sequence-verified with primers comp_for_seq , comp_rev_seq , and in some instances , additional internal gene primers ( see Table S8 for full list of primer sequences used in this study ) . Complementation plasmids were transformed into E . coli donor strain WM3064 and delivered into MR-1 via conjugation . Single MR-1 colonies carrying the complementation plasmids were selected on LB plates with gentamicin and assayed for growth in a microplate format as described above . See Table S9 for full list of complementation plasmids used in this study . TIGR roles and subroles are associated with some gene families in the TIGRFam database [29] . We obtained TIGRFam assignments from MicrobesOnline along with their associated roles and subroles . To exclude functional assignments that were not specific , we removed assignments without a subrole or that matched the strings “unknown” or “other” . This left us with 978 assignments . To predict the functional classification of genes from fitness and/or expression data , we used a standard implementation of random forests ( randomForest 4 . 5–36 in R-2 . 11 ) and default settings . A random forest is a collection of decision trees , each of which makes their own predictions . The random forest's prediction is that predicted by the largest number of decision trees , and the confidence of the prediction is the proportion of trees in the forest that make that prediction . Before applying the random forest , we subtracted the mean from each experiment and replaced missing values with zeroes . To assess the predictions without being biased by the training data , we used 10-fold cross-validation: we trained the classifier on 90% of the genes with known subroles and then made predictions for the remaining 10% of the genes . We repeated this 10 times so that we had predictions for every gene . Because genes in the same operon tend to be functionally related , we ensured that genes in the training set ( the 90% ) were not in the same operons as the genes we made predictions for . | Many computationally predicted gene annotations in bacteria are incomplete or wrong . Consequently , experimental methods to systematically determine gene function in bacteria are required . Here , we describe a genetic approach to meet this challenge . We constructed a large transposon mutant library in the metal-reducing bacterium Shewanella oneidensis MR-1 and profiled the fitness of this collection in more than 100 diverse experimental conditions . In addition to identifying a phenotype for more than 2 , 000 genes , we demonstrate that mutant fitness profiles can be used to assign “evidence-based” gene annotations for enzymes , signaling proteins , transporters , and transcription factors , a subset of which we verify experimentally . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"bacteriology",
"microbial",
"metabolism",
"functional",
"genomics",
"microbiology",
"gene",
"function",
"genome",
"analysis",
"tools",
"gene",
"expression",
"biology",
"systems",
"biology",
"genetic",
"screens",
"genetics",
"genomics",
"gene",
"prediction",
"genetics",
"and",
"genomics"
] | 2011 | Evidence-Based Annotation of Gene Function in Shewanella oneidensis MR-1 Using Genome-Wide Fitness Profiling across 121 Conditions |
The isolation of neutralizing monoclonal antibodies ( nmAbs ) against the Zika virus ( ZIKV ) might lead to novel preventative strategies for infections in at-risk individuals , primarily pregnant women . Here we describe the characterization of human mAbs from the plasmablasts of an acutely infected patient . One of the 18 mAbs had the unusual feature of binding to and neutralizing ZIKV despite not appearing to have been diversified by affinity maturation . This mAb neutralized ZIKV ( Neut50 ~ 2 μg/ml ) but did not react with any of the four dengue virus serotypes . Except for the expected junctional diversity created by the joining of the V- ( D ) -J genes , there was no deviation from immunoglobulin germline genes . This is a rare example of a human mAb with neutralizing activity in the absence of detectable somatic hypermutation . Importantly , binding of this mAb to ZIKV was specifically inhibited by human plasma from ZIKV-exposed individuals , suggesting that it may be of value in a diagnostic setting .
Zika virus ( ZIKV ) belongs to the genus Flavivirus of the Flaviviridae family and is related to dengue virus ( DENV ) , yellow fever virus ( YFV ) , Japanese encephalitis virus ( JEV ) , and west Nile virus ( WNV ) [1] . The globally distributed mosquito species of the Aedes genus , vectors for many Flavivirus , can also transmit ZIKV [2 , 3] . However , ZIKV remained a relatively minor and obscure cause of human disease for most of the second half of the 20th century and was featured in a limited number of scientific reports . In fact , it was not until 2007 that autochthonous human infection was described outside Africa and continental Asia—in the Federated States of Micronesia [4–6] . Since then , reports from Brazil have chronicled a rapidly spreading epidemic that co-exists with DENV and chikungunya virus ( CHIKV ) . The epidemic has spread north with mosquito-borne transmission being reported in many nations of the Americas as far north as Mexico and southern Florida [7–9] . More ominously , ZIKV has been implicated as the causative agent in fetal developmental problems [10 , 11] . There are reports of ZIKV-associated birth defects , primarily brain abnormalities and microcephaly in infants born to mothers infected with ZIKV [12] . Virus has been recovered from amniotic fluid , placental , and brain tissues [13–21] . ZIKV infection has been classified as an ongoing threat by the World Health Organization . In the United States , the Centers for Disease Control and Prevention has issued guidance for the management of the infection in the general population , pregnant women , and infants [22–24] . Due to recent reports of sexually transmitted ZIKV infection , the CDC has also developed guidelines for prevention of this mode of transmission [22–27] . More recently , ZIKV transmission has also been described in Miami , Florida [28] , suggesting that autochthonous spread could occur in any region of the U . S . inhabited by Aedes spp . Treatment of a variety of human ailments using mAbs is revolutionizing our ability to ameliorate human suffering . For infectious disease , the Ebola epidemic highlighted the potential utility of a cocktail of three neutralizing ( n ) mAbs that block infection by the Ebola virus [29] . Most convincingly , the administration of a single nmAb up to five days post infectious virus exposure prevents the development of disease in Ebola-infected macaques [30] . Because mAbs can be engineered to prevent antibody-dependent enhancement by incorporating the L234A and L235A ( LALA ) mutations which reduce FcγR binding [31] , they are a promising intervention in flaviviral therapies . Our long-term goal is to use a cocktail of LALA-mutated nmAbs to prevent ZIKV infection in at-risk individuals , primarily pregnant women . Therapeutic nmAbs must be potent in order to be clinically viable , and most nmAb isolation strategies are based on the identification of high-titer , antigen-selected repertoires . Somatic hypermutation ( SHM ) in germinal center ( GC ) B cells provides the basis for selection of B cells producing Abs with increased affinity—a hallmark of the adaptive humoral response . This feature is conserved among mammals , highlighting the importance of Ab affinity enhancement for evolutionary fitness [32] . Thus , it is unsurprising that the vast majority of human Abs in the memory immunoglobulin ( Ig ) G pool have undergone affinity maturation and have , on average , 10–26 nucleotide substitutions from precursor genes [33] . The contribution of SHM to Ab-mediated viral neutralization is particularly clear for the chronically-induced broadly neutralizing antibodies to HIV [34–37] . Reversion of these anti-HIV nmAbs to precursor germline antibodies results in a drastic reduction or complete loss of viral neutralization [38–41] . Although mutated mAbs are found after secondary DENV infection , the role of these mutations in acute virus-neutralization and clearance is less clear [42–45] . Still , the prevalent thought is that antiviral Ab response involves the engagement of poor- or non-neutralizing germline clones generated by V ( D ) J rearrangement , followed by SHM-mediated refinement in germinal centers to enhance neutralization potency . Here we describe the isolation of 18 plasmablast-derived human mAbs , sorted 12 days post onset of symptoms from a ZIKV-patient in São Paulo , Brazil . The patient reported a previous history of dengue infection and yellow fever vaccination ( Table 1 ) . A few of the isolated Abs neutralized ZIKV , most of them at relatively high concentrations . Interestingly , one of these mAbs ( P1F12 ) exhibited no nucleotide mutations when compared to its corresponding germline sequences , but still recognized a ZIKV immunodominant epitope and neutralized the virus . These results suggest unforeseen roles for GC-independent responses against ZIKV and possibly other viruses .
Blood samples were collected from volunteer 533 , a 56-year-old woman who reported a pruriginous skin rash that started six days prior to the beginning of acute neurological deficits suggestive of GBS . ZIKV infection was confirmed by a positive real-time reverse-transcriptase PCR assay for ZIKV RNA in urine samples collected at days 11 and 12 after the onset of the first rash symptoms . Blood and cerebrospinal fluid were negative for ZIKV RNA . Previous history of a single dengue infection and yellow fever immunization were also reported . Peripheral blood mononuclear cells ( PBMCs ) were obtained from blood samples collected 12 days post onset of symptoms . Blood samples from patient 533 were obtained after signing a written consent form approved by the University of São Paulo’s Institutional Review Board ( CAPPesq 0652/09 ) . Anonymized plasma samples from volunteers in Brazil and US were obtained from naïve and convalescent subjects with RT-PCR-confirmed ZIKV or DENV infection ( S2 Table ) . Four volunteers donated samples post yellow fever vaccination . We determined the frequency of plasmablasts in circulation by cytometric analysis of PBMCs obtained from blood collected in acid citrate dextrose ( ACD ) using a Ficoll-Paque ( GE Lifesciences ) gradient . Briefly , we stained fresh PBMC samples ( 1 x 106 cells , room temperature , in the dark ) , with 100 μl of a cocktail containing the following fluorophore-antibody conjugates: phycoerythrin ( PE ) -CF594 anti-human CD3 ( clone UCHT1; Becton Dickinson [BD] ) , PE-CF594 anti-human CD14 ( clone MφP9; BD ) , Allophycocyanin ( APC ) -Cyanine ( Cy ) 7 anti-human CD19 ( clone SJ25C1; BD ) , Peridinin Chlorophyll Protein Complex ( PerCP ) anti-human CD20 ( clone L27; BD ) , APC anti-human CD27 antibody ( clone O323; Biolegend ) , Fluorescein isothiocyanate ( FITC ) anti-human CD38 ( clone HB7; BD ) , PE anti-human CD138 ( clone MI15 , BD ) . We also included the fixable viability dye LIVE/DEAD® Fixable Red Dead Cell Stain Kit ( Life Technologies ) in the staining mix , in order to discriminate between live and dead cells . After 30 min , we washed the cells twice with FACS buffer ( PBS , 0 . 5% FBS , 2 mM EDTA ) , resuspended with a PBS 1x solution , and stored at 4°C until acquisition on the same day . Samples were acquired using a BD FACSAria IIu flow cytometer and analyzed using FlowJo 9 ( FlowJo ) . The plasmablast population was defined as live CD19+ CD3- CD14- CD20- CD27+ CD38+ cells ( see gating and sort strategy in S1 Fig ) . Using this same plasmablast staining , fresh PBMC samples ( 5 x 106 cells ) were sorted on a BD FACSAria II flow cytometer . Single plasmablast cells were sorted into 96-well plates containing a lysis buffer designed to extract and preserve the RNA ( 250 mM Tris-HCl pH 8 . 3 , 375 mM KCl , 15 mM MgCl2 , 6 . 25 mM DTT , 250 ng/well yeast tRNA , Life Technologies; 20 U RNAse inhibitor , New England Biolabs [NEB]; 0 . 0625 μl/well IGEPAL CA-630 , Sigma ) . After sorting , the RNA plates were immediately frozen in dry ice for subsequent cloning of the Ab chains . We conducted reverse transcription followed by a nested PCR to amplify the variable region of the Immunoglobulin ( Ig ) chains using described protocols with minor modifications [46] . Briefly , cDNA was synthesized in a 25 μl reaction using the original sort plates . Each reaction contained 1 μl of 150 ng random hexamer ( IDT ) , 2 μl of 10 mM dNTP ( Life Technologies ) , 1 μl of SuperScript III Reverse Transcriptase ( Life technologies ) , 1 μl molecular biology grade water , and 20 μl of single sorted cell sample in lysis buffer ( described above ) . The reverse transcription reaction was performed at 42 ˚C for 10 min , 25 ˚C for 10 min , 50 ˚C for 60 min , 94 ˚C for 10 min . After the reaction was completed , cDNA was stored at -20 ˚C . Heavy and light chains were amplified in three different nested PCR reactions , using a mix of 5’ V-specific primers with matching 3’ primers to the constant regions of IgG , IgL , and IgK . PCR reactions were conducted using HotStarTaq Plus DNA Polymerase ( Qiagen ) . The second set of PCR reactions was carried out with primers redesigned to incorporate restriction sites compatible with subcloning into rhesus IgG1 expression vectors , instead of the original human vectors [46] . We sequenced the amplified and cloned products using primers complementary to the Ig constant regions . Sequences were analyzed using IgBLAST and IMGT/V-QUEST to identify V ( D ) J gene rearrangements , as well as SHM levels [47 , 48] . We expressed mAbs in Expi293F ( ThermoFisher ) human cell lines . The plasmids encoding heavy and light chains were co-transfected using the ExpiFectamine 293 Transfection kit ( A14525 , ThermoFisher ) . After 5–6 days , we harvested the secreted mAb in the supernatant . Ig concentration in the supernatant was determined by an anti-rhesus IgG ELISA , before we proceeded with the functional assays . For the experiments with purified mAbs , we used Protein A Plus ( Pierce ) -containing columns to remove the impurities . The concentration of purified protein was determined by measurement of absorbance at 280 nm ( NanoDrop , Thermo Scientific ) . P1F12 binding was determined by both virus capture assay ( VCA ) and recombinant ( r ) E ELISAs . The VCA plates were coated overnight with the mouse-anti-Flavivirus monoclonal antibody 4G2 ( clone D1-4G2-4-15 , EMD Millipore ) followed by incubation with viral stocks ( ZIKV or DENV ) . The rE ELISA plates were coated with ZIKV E Protein ( MyBiosource , MBS596001 ) diluted to 5 μg / ml in PBS . After the coating step , the plates were washed with PBS and mAb samples diluted to 1 μg / ml were added to designated wells and incubated for 1 h at 37 ˚C . Subsequently , the plates were washed and detection was carried out using a goat anti-human IgG HRP secondary Ab ( Southern Biotech ) , which was added to all wells at a dilution of 1:10 , 000 . Following a 1 h incubation at 37C , the plates were washed and developed with TMB substrate at room temperature for 3–4 min . The plates were developed with TMB substrate at room temperature for 3–4 min . The reaction was stopped with TMB solution and absorbance was read at 450 nm . The neutralizing potency of the mAbs was measured using a flow cytometry-based assay [49 , 50] . In brief , recombinant mAbs ( transfection supernatant or purified ) were diluted and pre-incubated with ZIKV ( Paraiba ) or the reference DENV serotypes in a final volume of 220 μL for 1 h at 37 ˚C . The virus and mAb mixture ( 100 μL ) was added onto wells of a 24-well plate of 100% confluent Vero cell monolayers in duplicate . A new seed of Vero cells ( CCL-81TM ) was obtained from the American Type Culture Collection ( ATCC ) repository for this study . The inoculum was incubated in a 37 ˚C incubator at 5% CO2 for one hour with agitation of the plates every 15 min . After one hour , the virus and mAb-containing supernatants were aspirated and the wells were washed with media . Fresh media was then added and the plates were incubated for a total of 24 hours . Cells were trypsinized with 0 . 5% trypsin ( Life Technologies ) , fixed ( BD cytofix ) , and permeabilized ( BD cytoperm ) . Viral infection was detected with the 4G2 antibody ( Millipore ) recognizing ZIKV or DENV , followed by staining with an anti-mouse IgG2a APC fluorophore-conjugated secondary reagent ( Biolegend ) . The concentration to achieve half-maximal neutralization ( Neut50 ) was calculated using a nonlinear regression analysis with Prism 7 . 0 software ( GraphPad Software , Inc . ) . The following strains were used in our neutralization assays: ZIKV Paraiba 2015 ( KX280026 . 1 ) , DENV1-West Pac ( U88535 . 1 ) , DENV2-NGC ( AF038403 . 1 ) , DENV3-Sleman/78 ( AY648961 ) , and DENV4-Dominica ( AF326573 . 1 ) PRNTs were conducted as previously described [51] . Briefly , purified P1F12 was serially diluted in OptiMEM supplemented with 2% human serum albumin ( VWR ) , 2% fetal bovine serum , and gentamicin . ZIKV Paraiba 2015 was diluted to a final concentration of ~500–1000 PFU / mL in the same diluent added to equal volumes of the diluted Ab . The virus/mAb mixture was incubated at 37 ˚C for 30 min . Cell culture medium was removed from 90% confluent monolayer cultures of Vero cells on 24-well plates and 100 μl of the virus/Ab mixture was transferred onto duplicate cell monolayers . Cell monolayers were incubated for 60 min at 37 ˚C and overlaid with 1% methylcellulose in OptiMEM supplemented with 2% FBS 2mM glutamine + 50 μg / ml gentamicin . Samples were incubated at 37 ˚C for four days after which plaques were visualized by immunoperoxidase staining , and a 50% plaque-reduction neutralization titer was calculated . Inhibition of P1F12 mAb binding was determined by ELISA . To begin , the ELISA plate was coated with mouse anti-Flavivirus monoclonal antibody 4G2 ( EMD Millipore ) diluted 1:1 , 000 in carbonate binding buffer and incubated overnight at 4 ˚C . The next day , the plate was washed five times with PBS-Tween20 and wells were blocked with 5% skim milk in PBS for 1h at 37 ˚C . After the block step , the plate was washed and virus samples were added to designated wells for 1h incubation at room temperature . Subsequently , the plate was washed with PBS only and corresponding blocking plasma samples were added for 1h at 37 ˚C . Following the plasma block , the plate was washed and P1F12 was added to corresponding wells for 1 h at 37 ˚C . P1F12 was detected using a rhesus IgG-specific antibody ( mouse anti-monkey IgG-HRP clone SB108a; Southern Biotech ) . Thereafter , the plate was washed and wells were developed with TMB substrate at room temperature for 3–5 min before the reaction was stopped with TMB Stop Solution . Absorbance was determined at 450 nm .
We isolated plasmablasts from patient 533 who presented with suspected Guillain-Barré syndrome ( GBS ) ( Table 1 ) ( first day of symptoms = D0 ) . The patient had a previous history of dengue infection and yellow fever vaccination ( Table 1 ) . The previously healthy 56-year-old woman presented to the emergency room ( D6 ) reporting a progressive paresthesia mainly in the extremities of her hands , along with acute , intermittent pain in her left forearm during the previous four days . At physical examination , the patient presented with a grade IV asymmetric muscular weakness and hypoesthesia in her left limbs , with abolished deep tendon reflexes in the lower limbs . A mild weakness of her left facial muscles was also noted . The patient reported no respiratory disorders and no hoarseness , and no signs of dysautonomia were detected at the clinical evaluation . Fever , conjunctivitis , and myalgia or joint pain were absent during the illness . Afterwards , the patient was hospitalized with a clinical diagnosis of GBS , for which an intravenous human Ig ( IVIG ) treatment was initiated at a dosage of 0 . 4 g / kg / day for 5 days . Cerebrospinal fluid analysis and an electroneuromyogram were performed on fourth ( D10 ) and fifth ( D11 ) days after neurological symptom onset , respectively; the results were within normal limits . The electroneuromyogram was repeated on the 15th day of neurological symptoms , but no significant abnormalities were noted despite the persisting weakness in the patient’s left leg and arm . During the treatment with IVIG , the patient presented with transient worsening of her hemiparesis , but progressively recovered over the course of weeks after discharge from the hospital . At 32 days post-neurological symptom onset ( D38 ) , a physical exam revealed significant improvement of muscular strength and abolished deep tendon reflexes in the lower limbs . The remittent skin rash cleared completely 10 days after its initial emergence . Blood , cerebrospinal fluid and urine samples were collected on the 5th day of neurological symptoms ( D11 ) for detection of ZIKV by RT-PCR . The urine sample was ZIKV-positive by PCR , while blood and cerebrospinal fluid were negative . A saliva sample collected on D15 was negative for ZIKV . We isolated plasmablasts from peripheral blood mononuclear cells ( PBMCs ) collected on day 12 ( Table 1 ) . From wells containing single-sorted cells , we amplified , cloned , and sequenced heavy and light Ab chains using 5’ primers complementary to the V gene segments and a 3’ primer annealing to the constant IgG region [46] . This resulted in 18 paired heavy and light chains ( S1 Table ) . Eight of the 18 mAbs bound to ZIKV ( Fig 1 ) . Seven of these mAbs exhibited cross-reactivity to one or more of the DENV serotypes , and a single mAb–P1F12–bound exclusively to ZIKV . Interestingly , two mAbs bound to DENV but not ZIKV . We tested the neutralization potency of the ZIKV-specific P1F12 mAb in a flow-based neutralization assay and a plaque reduction neutralization test ( PRNT ) and found that it neutralized ZIKV at approximately 2 μg / ml ( PRNT50 ) ( Fig 2 ) . Analysis of the isolated antibody variable ( V ) domain sequences revealed five mAbs with average gene mutation levels ( 10–26 nucleotide modifications ) , two mAbs with over 30 nucleotide substitutions , and 11 mAbs with unusually low levels of SHM for isotype-switched mAbs ( lower than 10 changes ) ( S1 Table ) . The most highly mutated mAbs ( P1B08 and P1C03 ) were not ZIKV-specific by binding ( S1 Table ) . In fact , the eight ZIKV-binding mAbs had the lowest SHM levels , including four mAbs lacking clearly recognizable mutations when compared with putative heavy and light chain germline precursors ( S1 Table , Fig 3 ) . Except for junctional diversity , the ZIKV-neutralizer P1F12 mAb heavy chain did not exhibit signs of antigen-selected Ig diversification . P1F12 had an identical sequence to the Ig heavy chain variable ( IGHV ) genes segment IGHV3-7*01 up to the amino acid C105 , prior to the CDR-H3 ( International Immunogenetics Information System [IMGT] ) [52] . However , position G106–the site of the junction between IGHV and the IGH diversity ( IGHD ) genes–differed from the germline reference . Interestingly , this region is part of a segment ( N1 ) with non-germline nucleotides corresponding to six amino acids identified between the IGHV and IGHD genes ( Fig 3C ) . This segment is likely the result of N nucleotide additions generated during B cell Ig gene rearrangement , prior to antigen selection . Because of the lack of mutations elsewhere in the sequences , it is likely that the R106G substitution was also generated during this developmental step . The downstream sequence corresponding to the junction between IGHD3-22*01 and the IGH joining ( IGHJ ) IGHJ6*02 genes also revealed similar nucleotide insertions . Likewise , the Kappa ( K ) chain junction between the IGKV1-8*01 and IGKJ4*01 genes also contained one insertion . Although we cannot rule out the possibility of SHM-mediated nucleotide changes in the N insertion regions , no mutation was identified in the remainder of the regions of the heavy and light chains . Thus , the P1F12 mAb is likely very close or identical to the original V- ( D ) -J gene rearrangement in the naïve B cell before antigen contact . To investigate whether P1F12 recognizes an immunodominant ZIKV epitope , we used a serological blocking assay . In brief , this assay detects the presence of competing Abs that can inhibit the P1F12 mAb binding to its epitope . Because P1F12 did not bind to recombinant E protein ( Fig 4 ) we used whole virus in our binding assays . We captured ZIKV on the plate using the 4G2 mAb ( pan-Flavivirus ) , and incubated ZIKV with plasma from patients with diverse histories of DENV and ZIKV exposure ( S2 Table ) . We added unlabeled P1F12 ( engineered with rhesus IgG1 constant regions ) and detected binding of the mAb using a HRP-labeled mouse anti-rhesus mAb ( Fig 5 ) . Nine of ten plasma samples from individuals that had been infected with ZIKV blocked the binding of P1F12 in a blinded test ( Fig 5 , S2 Table ) . Similar blocking activity was observed regardless of whether individuals had been previously infected with DENV or had been vaccinated for yellow fever . In contrast , little or no blocking activity was observed by DENV+ plasma in the absence of prior ZIKV exposure ( Fig 5 ) . Furthermore , this recognition was specific in that it was not observed in 14 of 14 DENV-only infected individuals . Thus , the P1F12 serological blocking assay accurately predicted previous ZIKV exposure , as confirmed by RT-PCR , in all but one of the patient plasma samples tested . Although this patient , donor 1302 , had a positive urine RT-PCR result for ZIKV , plasma from 1302 did not block P1F12 binding to ZIKV ( S2 Table ) . Interestingly , the plasma did not exhibit detectable ZIKV-neutralizing activity , suggesting that this patient did not mount a measurable antibody response against ZIKV . In conclusion , only the plasma that inhibited ZIKV infection of Vero cells contained P1F12-blocking antibodies .
Here we show that a IgG mAb with no detectable SHM was generated against ZIKV early in infection . Remarkably , despite being germline-encoded , this mAb is ZIKV-specific and does not bind to any of the four DENV serotypes . Furthermore , this mAb not only neutralizes ZIKV , but it also binds to an immunodominant epitope on the virus . Remarkably , despite being germline-encoded , P1F12 binds specifically to ZIKV and does not cross-react with any of the four DENV serotypes . Our results also suggest that P1F12 recognizes a unique epitope on ZIKV . It is unclear how this Ab developed such specificity without SHM . Finally , these findings suggest that affinity maturation is not necessary for the generation of isotype switched virus-neutralizing Abs . Low levels of SHM in Abs possessing neutralizing activity have been previously reported in mice and humans [53–55] , supporting the idea that germline-encoded mAbs can indeed neutralize . Abs with low levels of SHM have also been reported during the acute phase of human DENV infection , but it was not clear that these Abs contributed to the antiviral neutralization activity [56] . In studies in mice , VSV-specific mAbs lacking SHM have been isolated previously [53] . Interestingly , secondary , but not primary , mouse Abs against VSV had mutations [57] . Furthermore , the reversion of these mutated Abs to non-mutated precursors reduced , but did not abrogate , VSV binding and neutralizing activity . The binding differences between the mutated and germline Abs were much less pronounced than might be expected [57] . Additionally , mice that cannot conduct SHM due to AID knockout still mounted neutralizing Ab responses against Friend virus , a strain of murine leukemia virus [55] . It has been suggested that these Abs lacking extensive SHM undergo a GC-independent developmental pathway [58] , although the mechanistic basis for this phenomenon remains to be elucidated . Rapid , GC-independent responses might be particularly relevant in the control of acute cytopathic viruses [55 , 58 , 59] . The GC-independent Abs would arise quickly after infection and then curtail viral replication , preventing virus-mediated damage [60] . Even more provocatively , Hangartner et al . have argued that cytopathic viruses specifically evolved to retain binding to these germline sequences to decrease host lethality and increase fitness . On the other hand , chronic viruses may have evolved to avoid germline-binding and development of neutralizing responses to persist [60] . So far , these hypotheses remain unsubstantiated by the lack of evidence for strictly germline neutralizing Ab responses in humans . While our experiments were not specifically designed to detect GC-independent responses , it seems likely that the isotype-switched P1F12 originated directly from a germline precursor . We isolated P1F12 from a ZIKV-infected individual that developed neurological complications compatible with GBS and was treated with IVIG . Underlying factors that influence the potential association of GBS and ZIKV infection might involve an autoimmune process , which could influence the development of immune responses [61] . Additionally , IVIG may have had a role in the selection of the Ab responses mounted by peripheral B cell repertoires [62] . This is unlikely , however , since the patient initiated IVIG treatment on the same day that the plasmablasts were isolated . It is possible , then , that GBS or IVIG-treatment influenced the development of P1F12 . These potential associations are difficult to determine and were outside the scope of this study . It is clear , however , that these responses were not exclusive to volunteer 533 , as P1F12 binding can be blocked by the serum of most ZIKV-infected individuals ( Fig 5 ) . Recently described ZIKV-specific mAbs derived from Epstein-Barr virus–immortalized memory B cells are highly polyclonal and have undergone SHM [42] . However , SHM levels in these human anti-ZIKV mAbs were lower than SHM levels in mAbs isolated in response to primary infections or vaccination ( SARS- CoV , H5N1 , rabies vaccine ) , recurrent or chronic infections ( RSV , PIV , Staphylococcus aureus , Klebsiella pneumoniae , HCMV , HCV ) or autoimmune diseases [42] . Wang et al . have recently reported the isolation of 13 new ZIKV-specific mAbs from memory B cells , three of which had very little SHM [45] . These mAbs were isolated from memory cells sorted with soluble and monomeric ZIKV E proteins and , in contrast to P1F12 , bind to the recombinant protein [45] . In contrast , we isolated ZIKV-specific mAbs from circulating plasmablasts at D12 . The peak recall of memory B-cell derived plasmablasts is thought to occur within the first week post-secondary infection [63 , 64] . Thus , it is probable that most of the isolated mAbs did not have a memory-B cell origin , and it remains possible that some of the plasmablasts were sorted from the basal population that circulate in low frequencies in the blood . In conclusion , the isolation of mAbs using different B cell methods suggest that anti-ZIKV mAbs with germline characteristics are not limited to specific B cell subtypes [42 , 45] . Notably , the anti-ZIKV mAbs isolated to date are less mutated than the mAbs isolated after related DENV infections [42–45] . Together , these findings suggest possible differences in the development of Ab responses against ZIKV . Unfortunately , despite our efforts , we were unable to map P1F12’s binding site . We first employed an in vitro escape assay [65] , which did not result in a single mutated consensus sequence . Also , P1F12 did not bind to the prM/E proteins expressed in cells , precluding our ability to map this interaction using an Ala-mutated envelope panel [66 , 67] . Characterizing this interaction will , thus , require a significant effort that is beyond the scope of the current manuscript . Because the P1F12 mAb retains the ability to bind virions , our conclusion is that it binds to a conformational epitope . Based on the cohort of human plasma samples tested in this study , it appears that most ZIKV-infected individuals mount Ab responses against the epitope recognized by P1F12 . This epitope is recognized by Abs in individuals previously infected by ZIKV , thereby preventing the binding of P1F12 . By contrast , Abs in the plasma from individuals previously infected by any of the DENV serotypes , do not prevent binding of P1F12 . P1F12 may , therefore have potential as a diagnostic . Several diagnostic options for testing for ZIKV exposure exist , including RT-PCR , IgM ELISA , and PRNT methods [22 , 68] . While it is relatively straightforward to detect ZIKV nucleic acid during the acute phase in blood , urine , saliva , and semen , it has proven more difficult to design rapid and effective diagnostics for ZIKV exposure in the chronic phase . For samples collected after the first week of symptoms , the initial test is an anti-ZIKV , anti-DENV , anti-CHIKV virus IgM ELISA [68] . However , in patients who have received a flaviviral vaccine ( DENV , YFV , or JEV ) and/or have been infected with any Flaviviruses in the past , these assays may be difficult to interpret due to the cross-reactivity of the Abs [68–73] . Thus , a positive IgM test needs to be confirmed with a laborious PRNT assay . IgM antibodies persist for 2–12 weeks in serum , and sera from individuals previously infected for more than 12 weeks would also have to be confirmed with a virus neutralization-based method [68] . Our plasma inhibition assay may , perhaps , provide an alternative to these other techniques . In this study , we isolated plasmablast-derived Abs from a ZIKV-infected individual with unusual characteristics . The human IgG P1F12 has no or limited SHM yet binds to an immunodominant ZIKV epitope that is not present on any of the four DENV serotypes . Furthermore , this mAb can neutralize the virus with a Neut50 of approximately 2 μg / ml . Our results suggest that SHM-independent pathways may generate neutralizing Abs in the responses against ZIKV . | Antibody affinity maturation through somatic hypermutation ( SHM ) is thought to be critical for the development of antibodies with virus-neutralizing activity . Contrary to this notion , we describe novel human anti-Zika virus ( ZIKV ) antibodies with very low mutation levels , isolated from plasmablasts early after the onset of symptoms . Surprisingly , one IgG monoclonal antibody , P1F12 , bound to ZIKV and neutralized the virus , despite having no detectable mutations . This antibody is specific for ZIKV and did not cross-react with DENV . Furthermore , plasma from ZIKV-positive individuals blocked the interaction of P1F12 with ZIKV , whereas plasma from DENV-positive patients did not have this inhibitory ability . P1F12 targets an immunodominant site , as ZIKV-positive samples blocked P1F12-ZIKV binding . Our study shows that isotype-switched virus-specific neutralizing Abs can develop in humans directly from germline sequences . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"enzyme-linked",
"immunoassays",
"pathology",
"and",
"laboratory",
"medicine",
"immune",
"cells",
"body",
"fluids",
"pathogens",
"immunology",
"microbiology",
"cloning",
"viruses",
"rna",
"viruses",
"molecular",
"biology",
"techniques",
"antibodies",
"immunologic",
"techniques",
"research",
"and",
"analysis",
"methods",
"immune",
"system",
"proteins",
"white",
"blood",
"cells",
"artificial",
"gene",
"amplification",
"and",
"extension",
"animal",
"cells",
"proteins",
"medical",
"microbiology",
"microbial",
"pathogens",
"immunoassays",
"recombinant",
"proteins",
"molecular",
"biology",
"biochemistry",
"antibody-producing",
"cells",
"blood",
"cell",
"biology",
"flaviviruses",
"b",
"cells",
"polymerase",
"chain",
"reaction",
"anatomy",
"viral",
"pathogens",
"physiology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"organisms",
"zika",
"virus"
] | 2017 | A human inferred germline antibody binds to an immunodominant epitope and neutralizes Zika virus |
Tetherin is a membrane protein of unusual topology expressed from rodents to humans that accumulates enveloped virus particles on the surface of infected cells . However , whether this ‘tethering’ activity promotes or restricts retroviral spread during acute retrovirus infection in vivo is controversial . We report here the identification of a single nucleotide polymorphism in the Tetherin gene of NZW/LacJ ( NZW ) mice that mutated the canonical ATG start site to GTG . Translation of NZW Tetherin from downstream ATGs deleted a conserved dual-tyrosine endosomal sorting motif , resulting in higher cell surface expression and more potent inhibition of Friend retrovirus release compared to C57BL/6 ( B6 ) Tetherin in vitro . Analysis of ( B6×NZW ) F1 hybrid mice revealed that increased Tetherin cell surface expression in NZW mice is a recessive trait in vivo . Using a classical genetic backcrossing approach , NZW Tetherin expression strongly correlated with decreased Friend retrovirus replication and pathogenesis . However , the protective effect of NZW Tetherin was not observed in the context of B6 Apobec3/Rfv3 resistance . These findings identify the first functional Tetherin polymorphism within a mammalian host , demonstrate that Tetherin cell surface expression is a key parameter for retroviral restriction , and suggest the existence of a restriction factor hierarchy to counteract pathogenic retrovirus infections in vivo .
In order to infect and persist in the host , retroviruses such as HIV-1 encode proteins that counteract innate resistance genes that are also referred to as “restriction factors” . Host restriction factors have the potential to directly interfere with specific steps of the retrovirus life cycle and have been the subject of intense study in the last decade . In this regard , mechanistic studies on how the HIV-1 Vpu protein promotes virion release in vitro resulted in the discovery of the long-sought ‘Tetherin’ molecule [1]–[2] . Tetherin ( also known as BST-2 , CD317 , HM1 . 24 and PDCA-1 ) is a homodimeric protein containing an N-terminal transmembrane and C-terminal glycophosphatidyl inositol anchor [3] that ‘tethers’ virions on the surface of the infected cells , resulting in extensive virion aggregation [1]–[2] , [4]–[5] . The impact of Tetherin-mediated virion aggregation on retroviral spread is controversial . In pandemic HIV-1 strains lacking the human Tetherin antagonist Vpu , surface-tethered virions associate with the virological synapse , but this interaction has been reported to both inhibit [6]–[7] and promote [8] cell-to-cell virus spread . Human T-Lymphotropic Virus type I [9] and Feline Leukemia virus [10] were also suggested to utilize human and feline Tetherin , respectively , for cell-to-cell spread in vitro . In contrast , Tetherin-driven retrovirus evolution in the SIV or SHIV lentivirus infection models [11]–[12] suggested that Tetherin was an antiretroviral factor . Resolving these opposing views may require pathogenic retrovirus infection studies that isolate the Tetherin gene in vivo . Recently , a study in Tetherin deficient mice demonstrated that Tetherin restricts Moloney Murine Leukemia Virus ( MLV ) and a pathogenic MLV complex known as LP-BM5 in vivo [13] . In that study , Interferon-α ( IFN-α ) induction through poly ( I:C ) treatment was required to unmask the activity of Tetherin in newborn mice infected with Moloney MLV . Moreover , the effect of Tetherin on LP-BM5 infection of adult mice was not observed until after 8 weeks post-infection , during the chronic stage and when adaptive immune responses had already developed . In contrast , studies on mice deficient with another restriction factor , Apobec3 , revealed a significant impact on retrovirus replication and pathogenesis during the first week of retrovirus infection in both newborn [14]–[15] and adult immunocompetent mice [16]–[18] . Thus , in contrast to Apobec3 , the impact of Tetherin during acute retrovirus infection appeared to be minimal , suggesting that Tetherin may not be a potent innate restriction factor in vivo . Comparison of Tetherin gene sequences from various mammalian hosts revealed high levels of positive selection in Tetherin , likely reflecting the long-standing genetic conflict between retroviruses and mammalian hosts [19]–[20] . We therefore analyzed Tetherin gene sequences from catalogued inbred mouse strain genomes , and report here the identification of a single nucleotide polymorphism ( SNP ) in Tetherin that significantly increased its ability to inhibit retroviral replication and pathogenesis in vivo . Due to the nature of this unique SNP , the results revealed that Tetherin cell surface expression is critical for retroviral control . Finally , the classical backcrossing approach highlighted a specific genetic context that revealed potent Tetherin activity during acute retroviral infection in vivo .
To investigate if mouse Tetherin harbored polymorphisms in putative functional domains , we took advantage of catalogued polymorphisms from multiple inbred mouse strains archived in the Ensembl database ( Figure S1 in Text S1 ) . A mutation from ATG ( Methionine ) to GTG ( Valine ) ( dbSNP ID: rs51822354 ) was found in the Tetherin start site of NZW/LacJ ( NZW ) mice . This was confirmed by sequencing the region from NZW genomic DNA , but was not detected in C57BL/6 ( B6 ) ( Figure 1A ) or the closely related NZB strain ( Figure S1C in Text S1 ) . Since the Tetherin single nucleotide polymorphism ( SNP ) maps to the canonical start site , we first determined the translational efficiencies of B6 versus NZW Tetherin in a cell-free in vitro translation assay . B6 and NZW Tetherin were amplified from primary spleen samples and linked to a C-terminal 3×FLAG tag . We hypothesized that since GUG is a highly inefficient translational initiation codon in mammalian cells [21] , downstream Methionines at positions 13 and 16 may be used as alternative start sites . To investigate this possibility , Methionines at positions 1 , 13 and 16 were mutated singly or in combination to Alanine ( GCC ) ( Figure 1B ) . T7-promoter containing PCR amplicons were translated in rabbit reticulocyte lysates and the resulting Tetherin translation products were evaluated by Western blot ( Figure 1C ) . As shown in Figure 1D , wild-type B6 and NZW Tetherin were translated to equivalent levels , demonstrating that the Tetherin SNP did not affect Tetherin translation levels . In contrast , Alanine substitutions of NZW Tetherin at Methionine positions 13 and 16 ( NZW M13 , 16A mutant ) completely abrogated expression ( Figure 1D ) , demonstrating that Valine at position 1 could not be used for translational initiation . Downstream Methionines at positions 13 and 16 likely initiated Tetherin translation since Alanine substitutions of B6 Tetherin at Methionine position 1 ( B6 M1A mutant ) and Methionine positions 1 and 13 ( B6 M1 , 13A mutant ) still resulted in translation ( Figure 1D ) . Translation from downstream Methionines would result in an approximate 1 . 4 kDa decrease in molecular weight . We observed slight shifts in molecular weight between B6 WT , M1A and M1 , 13A Tetherin , as well as between NZW WT , M13A and M16A Tetherin ( Figure 1D ) that corresponded to translation products from remaining start sites . Overall , these findings indicated that NZW Tetherin is translated from downstream Methionines and lacked the N-terminal 12 amino acids ( Figure 1E ) . The N-terminal cytoplasmic domain of mammalian Tetherins encodes a conserved dual-Tyrosine motif at amino acid positions 6 and 8 that is critical for clathrin-mediated endocytosis [22]–[23] . Substituting Tyrosines at positions 6 and 8 of Tetherin with Alanines ( Y6 , 8A mutant ) decreased the internalization of Tetherin thereby increasing Tetherin cell surface expression [22]–[24] . Thus , deletion of the N-terminal 12 amino acids of B6 Tetherin as predicted for NZW Tetherin ( Figure 1D ) should increase cell surface expression . To test this hypothesis , untagged B6 and NZW Tetherin constructs were transfected into 293T cells and cell surface expression was analyzed . Using immunofluorescence microscopy , we observed brighter and more defined signals for NZW Tetherin on the plasma membrane compared to B6 Tetherin ( Figure S2 in Text S1 ) . To quantify cell surface expression , we performed flow cytometry ( Figure 2A ) , measuring both median fluorescence intensity ( MFI ) and percentage of Tetherin+ cells . The percentage of Tetherin+ cells was gated based on a FI cut-off that yielded <1% positivity from cells transfected with empty vector . As expected , NZW Tetherin was expressed to a significantly higher level on the surface of transfected 293T cells compared to B6 Tetherin ( Figure 2A ) . In fact , NZW Tetherin was expressed to a similar extent as the B6 Tetherin endocytosis mutant Y6 , 8A and a B6 Tetherin mutant with the canonical start site mutated to Alanine ( M1A ) ( Figure 2B ) . We next compared the total and cell surface levels of B6 and NZW Tetherin by staining transfected 293T cells in an intact ( cell surface only ) versus permeabilized ( cell surface plus intracellular ) state . For the same transfected cells , the percentage of Tetherin+ cells was higher in permeabilized versus intact cells ( Figure 2C ) , suggesting that a proportion of Tetherin is found inside the cells . Using the ratio of surface versus total Tetherin expression to compute intracellular levels , a significantly higher percentage of B6 Tetherin was found intracellularly compared to NZW Tetherin ( Figure 2D ) and the B6 M1A and B6 Y6 , 8A mutants ( Figure S3 in Text S1 ) , consistent with more efficient endocytosis of B6 Tetherin . Since the total and cell surface Tetherin expression was higher in NZW Tetherin compared to B6 Tetherin ( Figure 2C ) despite no difference in translation efficiencies ( Figure 1D ) , we hypothesized that B6 Tetherin may be more efficiently shuttled into endosomal compartments for degradation , as suggested by other reports [24]–[25] . We therefore treated B6 Tetherin-transfected cells with increasing doses of a dynamin-dependent endocytosis inhibitor , Dynasore [26] . Dynasore treatment restored B6 Tetherin cell surface expression to similar levels as NZW Tetherin ( Figure 2E; Figure S4 in Text S1 ) . To complement these results , we also utilized a dominant-negative dynamin mutant , K44A , that was previously shown to block an intermediate stage in coated vesicle formation likely due to decreased guanine-nucleotide binding affinity [27] . Co-expression of B6 Tetherin with Dynamin K44A [28] increased B6 Tetherin cell surface expression to the same level as NZW Tetherin ( Figure S5 in Text S1 ) . Thus , NZW Tetherin was expressed at higher levels on the cell surface consistent with deletion of the N-terminal 12 amino acid domain and the consequent defect in endosomal recycling . In cell culture , Tetherin activity results in aggregation of retrovirus particles on the infected cell surface , resulting in decreased virion release in the surrounding media [1]–[2] , [4]–[5] . Tethering of murine retroviruses by mouse Tetherin has yet to be visualized , but this is highly likely based on the ability of mouse Tetherin to tether HIV-1 as observed by electron microscopy [5] and inhibit the release of Moloney Murine Leukemia Virus [29] . We therefore compared the ability of B6 versus NZW Tetherin to inhibit the release of Friend Murine Leukemia Virus ( F-MuLV ) . An F-MuLV molecular clone was co-transfected with untagged mouse Tetherin constructs into 293T cells and the levels of total and infectious virions released after 2 days were measured by quantitative PCR and focal infectivity assays , respectively [30]–[31] . Between 23 to 34% of co-transfected 293T cells expressed FV envelope gp70 based on FACS staining following permeabilization ( Figure 3A ) . Relative to vector control , NZW Tetherin inhibited total virion release 44-fold , while B6 Tetherin inhibited 11-fold ( Figure 3B ) . Infectious virion release was inhibited 50-fold for NZW Tetherin and 17-fold for B6 Tetherin ( Figure 3C ) . The 3 to 4-fold difference in B6 versus NZW Tetherin virion release were statistically-significant ( Figure 3B–C ) . Moreover , the B6 M1A and B6 Y6 , 8A mutants , which exhibited higher cell surface expression ( Figure 2B ) , also inhibited F-MuLV virion release better than wild type B6 Tetherin ( Figure 3B–C ) . Since Tetherin has been reported to decrease HIV-1 virion infectivity [32] , we estimated the infectivity of released F-MuLV virions by calculating the ratio of non-log transformed infectious titers and viral loads , and normalized to vector controls [31] . A moderate reduction in virion infectivity was observed with Tetherin co-transfection but did not reach statistical significance ( Figure 3D ) . Overall , our results demonstrated that NZW Tetherin , which is retained on the cell surface to a greater degree than B6 Tetherin , was a more potent inhibitor of F-MuLV virion release than B6 Tetherin in vitro . Previous studies have shown that Tetherin dimers are the functional units critical for restricting virion release in vitro [4] , [33] . Thus , if B6 and NZW Tetherin heterodimers are formed , the endocytosis motif in B6 Tetherin could lead to decreased cell surface expression of the heterodimer . In support of this notion , co-transfection of B6 and NZW Tetherin at an equimolar ratio led to cell-surface expression that was similar to B6 Tetherin ( Figure 4A; Figure S6 in Text S1 ) . To extend this finding in vivo , Tetherin cell surface expression in splenocytes of B6 , NZW and ( B6×NZW ) F1 mice was evaluated ( Figure 4B ) . Tetherin was expressed to significantly higher cell surface levels in NZW mice compared to B6 mice on dendritic cells and B cells ( Figure 4C and Figure S7A in Text S1 ) which normally express Tetherin [34]–[35] . In addition , higher cell surface expression levels of Tetherin were observed in erythroblasts of NZW compared to B6 mice ( Figure 4D and Figure S7B in Text S1 ) . Importantly , Tetherin cell surface expression in hybrid ( B6×NZW ) F1 mice was similar to that of B6 mice ( Figure 4B–D; Figure S7A–B in Text S1 ) , indicating that reduced B6 Tetherin cell surface expression was dominant over NZW Tetherin in vivo . The finding that low cell surface expression of B6 Tetherin was dominant over the high cell surface expression of NZW Tetherin provided the opportunity to investigate the effects of Tetherin cell surface expression levels on retroviral pathogenesis . Mice were infected with Friend retrovirus ( FV ) complex , one of the few retroviruses that cause disease in adult immunocompetent mice , and for which extensive information on the genetics of host resistance and susceptibility has been generated [36] ( Table 1 ) . [ ( B6×NZW ) F1×NZW]B1 offspring were generated , of which half were expected to be TetherinVal/Val ( high-surface expressors ) and half TetherinMet/Val ( low-surface expressors ) ( Figure 5A ) . The impacts of Fv1 [37] and Fv2 [38] host restriction factors ( Table 1 ) were nullified by infecting with a dual ( NB ) -tropic FV strain and by backcrossing to Fv2 dominant susceptible NZW mice , respectively . In addition , individual mice were genotyped for H-2 , which controls T and B cell immunity [36] ( Figure S8A in Text S1 ) , and for Rfv3 , which restricts FV during acute infection , promotes recovery from viremia , influences neutralizing antibody responses and is encoded by Apobec3 [17] , [30] ( Figure S8B in Text S1 ) . B1 mice ( n = 58 ) were classified into 4 groups based on H-2 and Rfv3 genotypes , of which 3 had sufficient sample sizes ( ≥5 mice per cohort ) for FV infection studies ( Table S1 in Text S1 ) . To control for the antiviral effects of H-2b and Rfv3r alleles from the B6 genetic background [17] , [39] , we initially focused on B1 progeny that lacked these resistance alleles ( Figure 5A; Table S1 in Text S1 ) . Susceptible Rfv3s/s H-2z/z mice were infected with NB-tropic FV and at 7 days post-infection ( dpi ) , plasma , spleen and bone marrow samples were analyzed . Similar to other viral infections [33] , [40] , Tetherin expression was highly upregulated following FV infection . Splenocyte Tetherin expression increased from 5-11% Tetherin+ splenocytes in uninfected NZW and ( B6×NZW ) F1 mice ( Figure 5B , left panel ) to 58–87% at 7 dpi ( Figure 5B , right panel ) . In fact , in mice with acute FV infection , FV positive cells showed significantly higher Tetherin cell surface expression compared to FV negative cells ( Figure 5C ) . Interestingly , FV infected TetherinMet/Val mice still exhibited significantly lower Tetherin cell surface expression in splenic erythroblasts , dendritic cells and B cells as compared to TetherinVal/Val mice , consistent with their genotypes ( Figure 5D ) . We next evaluated FV infection levels and pathogenesis determinants at 7 dpi to determine if pathogenesis phenotypes correlated with the Tetherin genotype . TetherinMet/Val low-expressor mice exhibited 5-fold higher infectious plasma viremia ( Figure 6A ) and 7-fold higher plasma viral loads ( Figure 6B ) compared to TetherinVal/Val high-expressor mice . There was a significant inverse correlation between plasma viral load and Tetherin cell surface expression in splenocytes ( Figure 6C ) , but most significantly with erythroblast Tetherin MFI levels ( Figure 6D ) . Virion infectivity , computed from the ratio of infectious viremia and plasma viral load [31] , was not significantly different between the two cohorts ( Figure 6E ) . Importantly , TetherinMet/Val mice had 2-fold higher levels of FV infection in bone marrow cells ( Figure 7A ) that included dendritic cells , erythroblasts and B cells ( Figure 7B ) . FV-induced proliferation of Ter119+ erythroblasts , which eventually leads to splenomegaly and erythroleukemia [37] , was also increased in both the bone marrows and spleens of TetherinMet/Val mice ( Figure 7C ) . In fact , TetherinMet/Val mice had 2-fold higher levels of splenomegaly compared to TetherinVal/Val mice ( Figure 7D ) . These findings demonstrate that the recessive NZW Tetherin allele significantly inhibited acute FV replication and pathogenesis in vivo . Several innate retrovirus restriction factors have been identified in the last decade , but whether these factors operate in a synergistic manner to inhibit retroviruses in vivo remain unknown . The classical backcrossing approach provided an opportunity to investigate Tetherin restriction in the context of Apobec3/Rfv3 , a deoxycytidine deaminase that inhibits infectious viremia during acute FV infection [17] , [31] . B6 mice are Rfv3 resistant and express higher levels of Apobec3 compared to Rfv3 susceptible NZW mice [41] . Consistent with the dominance of the Rfv3 resistance allele , B1 mice genotyped as Apobec3/Rfv3r/s had significantly lower infectious viremia at 7 dpi compared to Apobec3/Rfv3s/s mice ( Figure 8A ) . Apobec3/Rfv3r/s mice were further genotyped as TetherinMet/Val or TetherinVal/Val , and infectious viremia ( Figure 8B ) , splenomegaly ( Figure 8C ) and bone marrow erythroblast levels ( Figure 8D ) were compared . None of these parameters were significantly different between TetherinMet/Val versus TetherinVal/Val mice ( Figure 8B–D ) . The Tetherin SNP also did not influence infectious viremia , splenomegaly and bone marrow erythroblast levels when Apobec3/Rfv3r/s mice were analyzed according to their H-2 haplotypes ( Table S1 in Text S1 ) . Thus , the protective effect of NZW Tetherin during acute FV infection was not observed in the context of Apobec3/Rfv3 resistance .
Despite major progress in elucidating how Tetherin interacts with retroviral pathogens in vitro , the demonstration of Tetherin as a potent innate restriction factor in vivo remains controversial . In fact , several studies suggested that Tetherin could facilitate cell-to-cell virus spread [8]–[10] . In this study , we identified a Tetherin SNP in NZW mice that significantly enhanced Tetherin cell-surface expression , providing the first significant association between Tetherin genomic variation and innate retrovirus restriction in vivo . Interrogating the role of the Tetherin SNP is not feasible with a straightforward knock-out mouse approach . We therefore utilized a classical genetic backcrossing approach that takes into account the major resistance and susceptibility genes mapped in the FV model [36] and the dominance of the B6 Tetherin cell surface expression phenotype . Our results revealed a direct correlation between Tetherin genotype , phenotype , and retroviral restriction . These backcross results excluded contributions by any gene not closely linked to Tetherin . While genes tightly linked to Tetherin may still contribute to the resistance phenotype , we argue that this is unlikely based on several lines of evidence . First of all , the in vivo restriction phenotype was consistent with the in vitro phenotype of NZW Tetherin more potently inhibiting FV virion release compared to B6 Tetherin . Furthermore , there was a strong correlation between Tetherin cell surface expression levels mediated by the Tetherin SNP and FV replication in vivo . Finally , none of the genes flanking Tetherin have previously been identified as retrovirus restriction factors ( Figure S1A in Text S1 ) . It is most likely that the phenotypes observed were mediated primarily by the Tetherin SNP . While this manuscript was being prepared , a study using Tetherin deficient B6 mice revealed that Tetherin acts as a resistance gene to counteract Moloney MLV and a pathogenic MLV strain known as LP-BM5 in vivo [13] . In Moloney MLV infection of newborn mice , treatment with poly ( I:C ) , a potent inducer of IFN-α , was required to unmask the antiretroviral activity of Tetherin . On the other hand , the impact of Tetherin was not observed until after 8 weeks following pathogenic LP-BM5 infection , a timepoint when IFN-α was induced , but fits within the chronic stage of infection and adaptive immune responses . Our results in the FV model were obtained during acute FV infection ( 7 dpi ) and did not require exogenous administration of poly ( I:C ) . Interestingly , the FV stock used in this study , as well as studies that identified the Fv1 , Fv2 , Rfv1 and Rfv3 restriction genes and their molecular counterparts [36] , contained Lactate Dehydrogenase-elevating virus ( LDV ) [42] , a potent inducer of Type I IFNs [43]–[44] . Thus , our findings are consistent with the notion that Type I IFN induction is necessary for Tetherin antiretroviral activity in vivo . However , our findings differ from [13] in that Type I IFN induction was not sufficient to reveal the antiviral effect of Tetherin . Even with LDV co-infection ( and by inference , Type I IFN induction [43]–[44] ) , the protective effect of NZW Tetherin was not observed in the context of B6 Apobec3/Rfv3 resistance . Thus , Apobec3 is dominant and not additive with Tetherin restriction in vivo . We hypothesize that in addition to the lack of Type I IFN induction , the presence of a potent B6 Apobec3 gene could explain the weak antiviral effect of B6 Tetherin observed in acute Moloney MLV and LP-BM5 infections [13] . Follow-up studies on Tetherin retrovirus restriction will likely be more informative in Rfv3 susceptible genetic backgrounds . We speculate that the lack of synergy between B6 Apobec3 and NZW Tetherin could be due to mechanistic incompatibility: Apobec3 activity results in the release of non-infectious virions [31] , while Tetherin activity prevents virion release [1]–[2] . This model mirrors restriction factor hierarchies that have been observed in the FV model . For example , Fv1 is dominant over Fv2 ( a B-tropic virus will not efficiently infect Fv2 susceptible mice that are Fv1n/n ) , while Fv2 is dominant over Rfv3 ( Fv2 resistant B6 mice lacking Apobec3 are expected to recover ) . The current study suggests the existence of pathways that cross-regulate Apobec3 and Tetherin during acute retrovirus infection in vivo . The finding that NZW Tetherin is a more potent inhibitor of FV infection than B6 Tetherin suggests that high cell surface expression is a key parameter for Tetherin retroviral restriction in vivo . This result may be unexpected if Tetherin is viewed primarily as a restriction factor since NZW Tetherin has lost the YxY endocytosis motif that is present in most mammalian Tetherins [20] . Thus , from an evolutionary standpoint , Tetherins with a functional YxY motif should be a more effective configuration for retroviral restriction in vivo if this restriction is its primary function . However , Tetherin ( or PDCA-1 ) is a marker of plasmacytoid dendritic cells ( pDC ) and may have biological functions independent of retroviral restriction , such as the transport of cytokines [34] and regulating the Type I interferon response [45] . Thus , the Tetherin endocytosis motif may have been conserved for these physiological functions and retroviral restriction is part of a dual function . Surprisingly , Tetherin deficient mice did not harbor perturbations in pDC function [13] . Further analysis of congenic mice with a canonical ( B6 ) , NZW and null Tetherin may yet uncover critical physiological roles for Tetherin in vivo . Studies implicating Tetherin as a modulator of cell-to-cell spread [8]–[10] predict that the higher cell-surface expression of NZW Tetherin should enhance virological synapse formation , retroviral cellular spread and pathogenesis . Moreover , a study demonstrating that Tetherin shuttles virions into endosomal compartments for degradation [25] predict that NZW Tetherin should have weaker antiretroviral activity . Reconciling these studies with our seemingly contradictory results will require tracking the fate of Tetherin-restricted FV virions in congenic mice encoding B6 versus NZW Tetherin . However , we note that most studies on Tetherin were performed in vitro , under conditions that lack immune mediators that are present in vivo . Aggregation of protein antigens has been known to improve vaccine immunogenicity [46] , and Tetherin mediated aggregation of viral particles may confer a similar effect . The ability of a retrovirus restriction factor to promote immune responses is not unprecedented: noninfectious particle release due to B6 Apobec3 activity primed a more effective IgG response directed against intact virions [31] . Future investigations should therefore determine whether cells expressing surface-tethered virion aggregates are more efficiently targeted for innate immune killing and whether NZW Tetherin activity could prime a more effective adaptive immune response .
C57BL/6J ( B6 ) , BALB/cJ and NZW/LacJ mice were purchased from the Jackson Laboratory . All mice were handled in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . The protocol was approved by the University of Colorado Health Sciences Center Animal Care and Use Committee [Permit Number B-89709 ( 10 ) 1E] . All infections were performed under isoflurane anaesthesia , and all efforts were made to minimize suffering . 293T and Mus dunni cells were cultured in DMEM ( Mediatech ) containing 10% Fetal Bovine Serum ( Gemini ) and penicillin/streptomycin/glutamine ( Mediatech ) . Total RNA from B6 and NZW spleens were extracted using the RNAEasy kit ( Qiagen ) and cDNAs were synthesized using random hexamers in the RT2 EZ First Strand synthesis kit ( SA Biosciences ) . Mouse Tetherin was amplified using a forward primer 52 bp upstream of the canonical start codon ( 5′- AAGCTTGCGGCCGCTAAGGGCGTGGCCTGGAAAGGGT ) and a reverse primer that excludes the stop codon ( 5′-GAATCCTCTAGAAAAGAGCAGGAACAGTGACACT ) . Cloning sites for NotI and XbaI ( italicized ) were included in the primer , allowing for direct subcloning of the PCR amplicon into the p3×FLAG-CMV-14 vector ( Sigma ) . Various Alanine ( GCC ) mutations were introduced using the Quikchange XL Mutagenesis Kit ( Stratagene ) . A stop codon ( TAG ) between the last amino acid in Tetherin and the 3×FLAG tag was introduced to prepare untagged versions of these constructs . All constructs were verified by DNA sequencing . T7 expression cassettes were generated from the Tetherin constructs by PCR amplification with a forward primer containing the underlined T7 promoter ( 5′- TAATACGACTCACTATAGGGTAAGGGCGTGGCCTGGAAAGGGT ) and a reverse primer that flanks the 3×FLAG tag . Purified PCR products ( 400 ng ) were added into 50 µl of rabbit reticulocyte lysate from the TNT T7 coupled in vitro transcription/translation kit ( Promega ) and incubated at 30°C for 90 min . 5 µl of lysate was solubilized in 50 µl of Laemmli buffer , and 5 µl of solubilized lysate was loaded in a 4–20% gradient Criterion Gel ( Biorad ) and transferred into PVDF membranes ( Invitrogen ) . High levels of reticulocyte proteins below 20 kDa were cut out of the membrane prior to incubation with antibodies . Membranes were blocked with 5% skim milk for 2 h , probed with 1∶500 dilution of anti-FLAG M2 monoclonal antibody ( Sigma ) overnight , washed and incubated with 1∶5000 anti-mouse horseradish peroxidase conjugate ( Amersham ) for 1 h . The blots were developed using the Western Lightning reagent ( Perkin-Elmer ) and visualized in a Chemidoc XRS+ system ( Biorad ) . Commercial anti-Tetherin antibodies from Abnova ( catalogue # PAB13047 ) and Abcam ( catalogue # ab14694 ) did not detect mouse Tetherin by Western blot analysis ( data not shown ) . Untagged Tetherin expression plasmids ( 500 ng ) were transfected in triplicate into individual wells of a 6-well plate containing 80 , 000 293T cells using the Fugene 6 reagent ( Promega ) . After 2 days , the cells were washed twice and harvested with 1 ml of FACS buffer ( PBS+2% FBS ) . Different doses of Dynasore ( Sigma ) were added 4 hr before harvest in some experiments . To quantify total Tetherin expression ( cell surface plus intracellular ) , 293T cells were permeabilized and stained using the Cytofix/Cytoperm Kit ( BD Biosciences ) . As an alternative to Dynasore treatment to block endocytosis , 1 µg of wild-type or K44A mutant dynamin in an eGFP-expression vector ( generously provided by P . De Camilli through Addgene , plasmids 22163 and 22197; [28] ) and 250 ng of B6 or NZW Tetherin were co-transfected into 293T cells using Fugene 6 . Tetherin expression was analyzed in gated dynamin-eGFP+ cells . In all cases , 60 µl of the cell suspension was stained with 0 . 5 µl of PE-conjugated anti-PDCA-1 ( eBioscience; clone eBio927 ) for 30 min on ice , washed twice with FACS buffer then fixed with 1% paraformaldehyde . Samples were analyzed in a FACSCalibur machine ( BD Biosciences ) , collecting at least 80 , 000 total events . Untagged Tetherin constructs as well as vector control ( 100 ng ) were transfected into 293T cells using Fugene 6 in poly-D-Lysine coated chamber slides ( BD Biocoat; BD Biosciences ) . After 2 days , the cells were washed with PBS , fixed with 4% paraformaldehyde then blocked for 1 hr at room temperature with PBS containing 8% normal goat serum and 0 . 02% Triton X-100 ( Sigma ) . The samples were stained with 1∶100 anti-PDCA1 Alexa-488 ( eBioscience ) overnight at 4°C . The following day , the cells were stained with diamidino2-phenylindole ( DAPI; Sigma ) for 10 min at room temperature , and then washed 3× with PBS . The samples were visualized in a Zeiss Axioplan2 microscope . The F-MuLV molecular clone pLRB302 [47] ( 1 µg ) was co-transfected in triplicate with 50 ng Tetherin expression constructs into 6-well plates containing 80 , 000 293T cells plated the previous day with Fugene 6 ( Promega ) . After 2 days , MAb 720 supernatant was used to detect FV envelope gp70 expression by incubating permeabilized cells for 30 min , followed by 2 washes with FACS buffer and incubation for 30 min with goat anti-mouse IgG1 conjugated to APC ( Columbia Biosciences ) . Infectious titers in day 2 supernatants were quantified using a focal infectivity assay in Mus dunni cells with MAb 720 [17] . Total virion titers were determined using a quantitative real-time PCR based viral load assay [30]–[31] from 100 µl of culture supernatant treated with 2 . 5U DNAseI ( Benzonase; EMD Millipore ) for 30 min at room temperature . The same assay was also used to measure plasma viral load from 50 µl of infected mouse plasma . Quantitative PCR was performed at least in triplicate and had PCR efficiencies >90% . H-2 was genotyped by directly sequencing a 1 . 0-kb H2-Q1 PCR fragment amplified using primers H2 . F ( AACCTGGGTCAGGTCCTTCT ) and H2 . R ( CATGGCTGACAGAGGCTACA ) ( Figure S8A in Text S1 ) . To genotype Apobec3/Rfv3 , we used a multiplex primer set consisting of primers spanning ( mA3 . F and mA3 . R ) and positioned ( LTR . F ) within a 530 bp retroviral xenotropic murine leukemia virus ( X-MLV ) insertion in B6 mice [30] , [41] ( Figure S8B in Text S1 ) . The primer sequences are: mA3 . F ( TTCACAACCCCCATACTTGG ) ; mA3 . R ( CAGGCTGGTCTCAAACGATA ) ; and LTR . F ( TTGGGGAACCTGAAACTGAG ) . To genotype the start codon mutation in Tetherin , we amplified a 582 bp fragment spanning the start codon using primers Bst2 . F ( AAACCTTGGCCTTTGGTCTT ) and Bst2 . R ( TGTGACGGCGAAGTAGATTG ) and directly sequenced the PCR products using an internal primer Bst2 . seq ( GCGGACAGCCACTGTTAAGT ) ( Figure S1C in Text S1 ) . Tail DNA samples ( 10 ng ) were added into a 20 µl PCR reaction consisting of 1× Phusion HF Buffer , 10 mM dNTP , 20 pmol of primers and 1 . 25 U of Hot Start Phusion polymerase ( New England Biolabs ) . Cycling conditions in a PE 9700 machine included a 30 s hot-start at 98°C , followed by 30 cycles of 98°C 10 s denaturation , 60°C 30 s annealing and 72°C 1 min elongation . The FV stock used in this study contains an NB-tropic F-MuLV helper virus , a replication-defective Spleen-Focus Forming Virus and LDV . LDV is a natural mouse virus endemic in wild mouse populations , and its presence in the FV stocks has been traced to the 1960s from naturally infected mice [48]–[49] . Major findings that include the identification of Fv1 , Fv2 , Rfv1 and Rfv3 and their molecular counterparts utilized LDV+ FV stocks [36] , [42] . These 4 genes were used to classify and genotype the B1 cohort . Thus , an LDV+ FV inoculum was essential for this study . LDV was shown to stimulate Type I IFN production and modulates adaptive immune responses , but does not significantly influence acute FV infection levels [42] . FV inoculum stocks were prepared in BALB/c mice and had equivalent titers in BALB/c and NZW mice ( data not shown ) . NB-tropic FV ( 500 SFFU ) was infected intravenously in 300 µl RPMI and at 7 dpi , plasma samples , bone marrows and spleens were harvested . Due to severe male aggression , infections were performed in female B1 mice , without prior knowledge of the Tetherin genotype . Bone marrow and spleen cells ( 106 cells ) were stained with MAb 720 for 30 min , then co-stained with: Ter119-FITC ( clone TER-119 ) , CD3-Alexa700 ( 17A2 ) , ( BD Biosciences ) ; PDCA-1-PE ( eBio927 ) , CD11c-PE-Cy7 ( N418 ) , ( eBioscience ) ; CD19-PerCP-Cy5 . 5 ( 6D5 ) ( Biolegend ) and anti-mouse IgG1-APC ( Columbia Biosciences ) . Isotype controls and cells from uninfected mice were used for gating . The multicolor FACS panel yielded similar data from mice spleen and bone marrow with or without the addition of a mouse anti-CD16 Fc blocker ( BD Biosciences ) ( data not shown ) . Cells were processed in an LSR-II flow cytometer ( BD Biosciences ) , collecting up to 250 , 000 events per sample . Datasets were analyzed using the Flowjo software ( Treestar ) . | Significant portions of the human and mouse genomes are comprised of retroviral sequences , revealing the long history of conflict between mammalian hosts and retroviruses that led to the evolution of host restriction factors . Nucleotide mutations in restriction factor genes provide a glimpse of this ongoing evolutionary process , but studies that directly probe the impact of restriction factor mutations during retrovirus infection are limited . In this study , we identified a single nucleotide mutation in the Tetherin host restriction gene that resulted in retention of Tetherin on the cell surface . In cell culture , Tetherin accumulates virions on the infected cell surface and prevents virion release , but some studies suggested that Tetherin might facilitate cell-to-cell virus spread . Our studies reveal that the Tetherin polymorphism inhibits retrovirus replication and disease . Thus , increased Tetherin cell surface expression enhanced the antiretroviral function of Tetherin . These results could have important implications in harnessing the biology of Tetherin for controlling pathogenic retroviruses such as HIV-1 . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genetics",
"biology",
"microbiology",
"genetics",
"and",
"genomics"
] | 2012 | A Single Nucleotide Polymorphism in Tetherin Promotes Retrovirus Restriction In Vivo |
Polyparasitism can lead to severe disability in endemic populations . Yet , the association between soil-transmitted helminth ( STH ) and the cumulative incidence of Schistosoma japonicum infection has not been described . The aim of this work was to quantify the effect of misclassification error , which occurs when less than 100% accurate tests are used , in STH and S . japonicum infection status on the estimation of this association . Longitudinal data from 2276 participants in 50 villages in Samar province , Philippines treated at baseline for S . japonicum infection and followed for one year , served as the basis for this analysis . Participants provided 1–3 stool samples at baseline and 12 months later ( 2004–2005 ) to detect infections with STH and S . japonicum using the Kato-Katz technique . Variation from day-to-day in the excretion of eggs in feces introduces individual variations in the sensitivity and specificity of the Kato-Katz to detect infection . Bayesian logit models were used to take this variation into account and to investigate the impact of misclassification error on the association between these infections . Uniform priors for sensitivity and specificity of the diagnostic test to detect the three STH and S . japonicum were used . All results were adjusted for age , sex , occupation , and village-level clustering . Without correction for misclassification error , the odds ratios ( ORs ) between hookworm , Ascaris lumbricoides , and Trichuris trichiura , and S . japonicum infections were 1 . 28 ( 95% Bayesian credible intervals: 0 . 93 , 1 . 76 ) , 0 . 91 ( 95% BCI: 0 . 66 , 1 . 26 ) , and 1 . 11 ( 95% BCI: 0 . 80 , 1 . 55 ) , respectively , and 2 . 13 ( 95% BCI: 1 . 16 , 4 . 08 ) , 0 . 74 ( 95% BCI: 0 . 43 , 1 . 25 ) , and 1 . 32 ( 95% BCI: 0 . 80 , 2 . 27 ) , respectively , after correction for misclassification error for both exposure and outcome . The misclassification bias increased with decreasing test accuracy . Hookworm infection was found to be associated with increased 12-month cumulative incidence of S . japonicum infection after correction for misclassification error . Such important associations might be missed in analyses which do not adjust for misclassification errors .
Polyparasitism is a common feature in parasite endemic regions , which includes most developing countries [1] , [2] . High prevalence of co-infection with soil-transmitted helminths ( STHs ) , which include roundworm ( Ascaris lumbricoides ) , whipworm ( Trichuris trichiura ) , and hookworm ( Ancylostoma duodenale and Necator americanus ) , and Schistosoma spp . has been reported [3] , [4] . Together , these infections correspond to an estimated 43 . 5 million disability-adjusted life years ( DALYs ) lost annually [5] , [6] . Schistosomiasis and STH infections are associated with conditions of poverty , such as poor hygiene , lack of safe water , inadequate sanitation and factors such as water management systems , age , gender , and farming related activities [4] , [5] , [7]–[14] . Laboratory studies suggest that infection with one helminth may influence the outcome of infection with another helminth [15] . Positive cross-sectional correlation and synergism between schistosome and STH infections have been reported [2] , [3] , [6] , [16]–[18] . Immunosuppressive effect of STH has been reported , particularly with hookworm infections [19] , [20] . The influence of STH infection on risk of infection with schistosomes has not been epidemiologically investigated . One challenge faced by investigators is the use of a less than perfect diagnostic test . The outcome , exposures , confounding variables , or any combination of these can contain errors [21]–[23] . Error in identification of infection status occurs when the test used to identify the infection is not 100% accurate , or not a ‘gold’ standard test [21] , [24] , [25] . Schistosoma japonicum and STH infections are most commonly detected by examining a stool sample under the microscope for the presence of parasitic eggs . Variation from day-to-day in the excretion of S . japonicum and STH eggs in human feces has been reported [26]–[29] . Collecting stool samples over consecutive days has been shown to improve the sensitivity of coprological tests like Kato-Katz [29] , [30] . However , in practice , an unequal number of stool specimens per subject are collected as it is difficult to collect the desired number of stool samples from each subject . This produces potential complications in diagnosing S . japonicum and STH infections as the sensitivity and specificity of the diagnostic tests vary according to the number of stool samples examined [31] , [32] . The purpose of this study was to show the impact of adjusting for misclassification error in estimating the effect of STH infections on the 12-months cumulative incidence of S . japonicum infection . Measuring such impact will contribute to a better understanding of the association between STH and schistosomiasis .
The research was approved by the institutional review board ( IRB ) of the Brown University in the United States and the IRB of the Research Institute for Tropical Medicine in the Philippines . The data analysis component of the study was reviewed and approved by the University of Oklahoma Health Sciences Center IRB . The chiefs of all villages were asked permission for the village to be included in the study . In addition , all eligible participants were asked for their consent to participate . Only those individuals who provided written informed consent were included . Written informed consent for individuals below 18 years old was obtained and provided by parents or legal guardians . We used data from a longitudinal study conducted between January 2004 and December 2005 in the province of Samar , the Philippines . The main purpose of the original study was to assess the effect of water and animal management systems on the transmission of S . japonicum infection . The design of the baseline study was described elsewhere [33] . A brief summary is given below . Seventy-five out of 134 villages endemic for S . japonicum in Samar in 2002 were eligible for participation [33] . The inclusion criteria were safety and accessibility of the field team , location and number of households in each village . Twenty-five primarily rain-fed villages and 25 villages with some form of man-made irrigation system were selected . Eligible households were those of at least five members and where at least one member was working full time in a rain-fed farm in “rain-fed” villages and at least 50% of the time in a man-made irrigated farm in “irrigated” villages . A maximum of 35 eligible households were randomly selected from each village using the following procedure . A list of 50 random numbers was created ( one list per village ) . Eligible households were allocated consecutive numbers and visited in the order chosen at random . If a household refused to participate , the next available household was asked to participate . When 35 or fewer households were eligible in a village , they were all invited to participate in the study . At most six individuals including at least one full-time rice farmer were selected at random from each household . An individual-level interview included questions on age , gender , and occupation . Participants were asked to provide one stool sample ( morning or first ) per day for three consecutive days . Each participant provided between one and three stool samples . If a participant provided a stool sample on one of the three days but was unable for any reason to provide stool samples on other days , that person was still considered as a stool sample provider . Stool envelopes ( of wax paper and book paper ) with popsicle sticks were distributed to participants a day before the actual stool collection . At least thumb-size stool samples were submitted . Portions from different parts of the stool were taken to fill up the template . Although consistency of the stool sample was not recorded , only pasty to formed stool could be accommodated in the stool envelopes . Stool samples were processed 2–3 h after collection . Two slides were prepared from each stool sample . All slides were placed in a styrofoam box with cold packs inside at the end of each collection day . At the end of each collection week all slides were brought to a designated laboratory and transferred to a refrigerator . The time delay between stool sample processing and microscopic reading associated with day one stool collection ( provided by 99 . 45% of participants ) ranged from less than 24 hours to as long as 20 days with a median of 4 days ( inter-quartile range: 2–6 days ) . Stool samples were examined for the presence of eggs of S . japonicum and the three STHs . No distinction between N . americanus and A . duodenale eggs was made , although prior reports from the Philippines found exclusively Necator spp . infections [34] . The Kato-Katz technique was used to detect the helminth eggs in stool samples [35] . The number of eggs per gram of stool ( epg ) was counted for S . japonicum . Although the eggs of each of the STHs were originally documented qualitatively in five response categories ( 0 , + through ++++ ) , STHs were considered as dichotomous variables ( observed infected or uninfected ) since the researchers were particularly interested in this association . Also , since the infection of interest of the original study was schistosomiasis , the semi-quantitative ascertainment of STH infection may not have been as accurate as that for schistosomiasis . Laboratory technicians were blinded to the identity of the provider of the stool sample they were preparing and reading and did not know if two stool samples were from the same participant ( two consecutive day's sample ) . Details about the mass treatment have been published elsewhere [36] . Briefly , following the baseline data and stool collection , all residents who were ≥5 years of age at the time and living in the 50 study villages were offered praziquantel . Praziquantel was administered in two equal split doses to give each individual a total of 60 mg/kg . The split doses were administered 4 hours apart with the first dose usually between 9 am and noon . All participants who provided baseline stool samples had been notified of their test results before treatment was offered . Before mass treatment , community preparation was implemented and an effort was made to ensure all cases found to be positive for S . japonicum were treated . Despite these efforts , the village-level participation proportion varied from 16% to 81% [36] . The parasitological test results were shared with the local ministry of health and the national schistosomiasis control team and it was decided to treat villagers positive to STH at the end of the whole study , that is , after the 12-months follow-up . This approach was approved by both IRBs . All of the study participants were asked to provide three stool samples over three consecutive days 12 months after the mass treatment . All individuals who provided at least one stool sample were considered as follow-up stool sample providers . Stool samples were processed and examined in the same manner and by the same people as at baseline . Some of the participants who provided the baseline stool samples did not participate in the mass treatment program . Moreover , not all participants provided stool samples during the follow-up survey . The 12-month cumulative incidence of S . japonicum infection/reinfection following mass treatment can only be calculated among the “at-risk” participants who provided at least one stool sample at baseline and follow-up and received treatment . For the purpose of this study , we assumed 100% efficacy of praziquantel for the treatment of schistosomiasis . As mentioned earlier , we obtained between one and three stool samples on consecutive days from each participant at baseline and follow-up . This introduces individual variations in the sensitivity and specificity of the Kato-Katz to detect infection . To take this variation into account , and to adjust for the village-level clustering of infection , we used a Bayesian latent class hierarchical cumulative-logit regression model based on a method described by Joseph and others ( 1995 ) and adapted to our problem ( 1 , 2 , or 3 days of sampling ) for S . japonicum in animals and in humans in the Philippines [25] , [33] , [37] , [38] . The probability of any single test being positive is the sum of the probability of a true positive result and the probability of a false positive result . If P is the total probability of a positive test , then , from the properties of diagnostic tests , we have When there is more than one test per person , the properties of multiple tests can be modeled using probability P as the probability parameter of a binomial distribution , assuming that the tests are independent from each other [37] . In the absence of a ‘gold’ standard test , the true status of each subject is unknown , and hence can be considered as ‘latent data’ . According to Bayes' theorem , the joint posterior distribution is proportional to the product of the likelihood function and prior distribution , from which all inferences can be obtained . The posterior distribution is not directly available , but inferences about each parameter are available using a Gibbs sampler algorithm , as has become standard in Bayesian analysis . The unknown true infection status for each subject can be estimated once the sensitivity and specificity have been estimated . The main outcome of interest here is the probability distribution of the true S . japonicum infection category at follow-up . S . japonicum epg counts were grouped into three categories namely: uninfected ( 0 epg ) , light infection ( 1 to 100 epg ) and moderate to heavy infection ( over 100 epg ) [33] . With a three-category outcome variable , classification errors must be further subdivided . For example , when a participant who is truly negative tests positive , there are two possible errors and 1-specificity or the false positive rate must be divided into light or moderate/heavy misclassification errors . The exposure of interest is the probability distribution of true STH infection status ( for a particular STH ) classified as positive or negative . Separate models were carried out for each of the three STHs . Each hierarchical model consists of three levels , as follows: the first level includes one intercept parameter for each village and independent variables for age , sex , occupation , and one of the STHs under study . At the second level of the hierarchical model , the intercept parameters from each of the 50 villages are modeled as a linear regression to account for the clustering of infection within village . At the third level , prior distributions were specified for all parameters . Uniform ( uninformative ) prior distributions on the range from 0 to 1 ( parameters of the beta distribution: α = 1 , β = 1 ) were used for sensitivity and specificity of all three STH infections . For S . japonicum , prior specificity mean ( SD ) for one stool sample was based on our previous work and set to 94 . 7% ( 4 . 0% ) , and prior sensitivities ( SD ) for detecting light infection and moderately to heavy infection were set to 54 . 1% ( 10 . 1% ) and 75 . 3% ( 15% ) , respectively [33] . The above model was modified to construct three additional models: one model accounted for misclassification error in outcome but not in exposure , one accounted for misclassification error in exposure but not in outcome , and another one did not account for any misclassification error . For models where misclassification error was not accounted for , an individual with any stool sample positive for a particular STH was considered as infection positive for that STH . For S . japonicum , epg per participant ( intensity of infection ) was obtained by averaging the epg of all stool samples collected from a participant , which is the most commonly used method for calculating overall epg per participant [1] , [39] , [40] . We assumed conditional independence between subsequent tests in our model , meaning in practice that when more than one sample was available from a subject , the test results are independent from each other , conditional on the person's true infection status . In other words , the probability of a positive ( or negative ) test depends only on the true status , and once this true staus is known , does not depend on any test results from other days . This assumption seemed reasonable , and simplifies the statistical model compared to a model that might account for any between-day dependencies . WinBUGS software ( version 1 . 4 . 3 , MRC Biostatistics Unit , Cambridge , UK ) was used to implement the Gibbs sampler algorithm . Posterior medians of random samples derived from marginal posterior densities were used as point estimates , reported with 95% Bayesian credible intervals ( BCI ) . The programs written in WinBUGS are available upon request to the authors .
Of the 5624 individuals who agreed to participate in the study at baseline , 2276 ( 40 . 5% ) constitute the group “at-risk” . The “at-risk” group and those who were not treated with praziquantel or did not provided any stool sample during the follow-up ( “not at-risk” group ) are compared in Table 1 . A higher proportion of people in the “at-risk” group had a positive schistosomiasis test at baseline ( 23 . 5% ) as compared to those in the “not at-risk” group ( 10 . 5% ) . Because of this discrepancy , there were more rice farmers in the “at-risk” group than in the ‘not at-risk’ group ( 50 . 2% vs . 40 . 9% ) , since rice farming is associated with S . japonicum infection . Having been positive at baseline , however , did not have an impact on the probability of providing a stool sample at follow-up among those people who did receive treatment ( 75 . 9% vs . 76 . 1% ) . Figure 1 displays the OR estimates for the exposure variable ( STH infection ) from models with and without correction for misclassification error . The OR estimates ( 95% BCI ) for hookworm infection changed from 1 . 28 ( 0 . 93 , 1 . 76 ) without any adjustment for misclassification error to 2 . 13 ( 1 . 16 , 4 . 08 ) when both exposure ( hookworm infection ) and outcome ( S . japonicum infection ) were corrected for misclassification error . For A . lumbricoides and T . trichiura , the OR changed from 0 . 91 ( 0 . 66 , 1 . 26 ) to 0 . 74 ( 0 . 43 , 1 . 25 ) and 1 . 11 ( 0 . 80 , 1 . 55 ) to 1 . 32 ( 0 . 80 , 2 . 27 ) , respectively . Correction for misclassification error in either exposure or outcome gave intermediate estimates . However , only adjusting for misclassification error in S . japonicum had a larger impact on the OR estimates and their 95% BCI than only adjusting for the misclassification error in the STH . In general , misclassification error-adjusted estimates were further away from the null value and had wider confidence intervals than non-adjusted estimates . In addition , the impact of adjusting for misclassification error on OR estimates and their 95% BCI was larger for hookworm which had the lowest sensitivity and specificity values . Table 2 provides OR estimates for covariates from respective STH models , with and without adjustment for misclassification error . For all three STH models , misclassification error-unadjusted OR estimate for >40 year-old individuals ( reference: ≤10 years ) was approximately 1 . 5 times that found in the exposure and outcome misclassification-adjusted model . Also , for all three STH models , OR estimates for males ( reference: females ) from the misclassification error-adjusted model were considerably different from OR estimates found in the unadjusted model . In general , both exposure and outcome misclassification error-adjusted ORs , and only outcome-adjusted ORs were similar whereas misclassification error-unadjusted ORs and only exposure-adjusted ORs were similar . The estimated 95% BCI from models adjusting for misclassification error in the outcome variables , with or without adjustment for misclassification error in the STH , were wider than those from models without adjustment of the outcome variable . Adjusting for misclassification error of STH only did not impact the width of the 95% BCI of the ORs of other variables in the model .
To our knowledge , this is the first longitudinal study to estimate the effect of STH infection on the 12-month risk of S . japonicum infection in a population where both of these infections are endemic . In addition , this study minimizes several potential biases by including adjustment for misclassification error in both dependent and independent variables , varying sensitivity and specificity of both tests depending on the numbers of samples available , accounting for clustering between individuals within villages , and taking care of other possible confounders . The adjusted model suggests that hookworm infection is associated with increased 12-month risk of S . japonicum infection following treatment with praziquantel . The two other STH studied did not have an important effect on the risk of infection with schistosomiasis . Although our analysis included only about one third of the baseline participants from 50 villages , the longitudinal sample size was large enough for this analysis . When comparing individuals included in and excluded from the analysis , we found more rice farmers in the ‘at-risk’ group than in the ‘not at-risk’ group . This is because more males were treated than females ( 56 . 4% vs . 43 . 6% ) , and because more rice farmers were infected with S . japonicum at baseline . A larger proportion of individuals infected with S . japonicum at baseline received treatment [36] . However , this did not have an impact on the probability of providing a stool sample at follow-up among those people who did receive treatment . So , the use of the “at-risk” group of participants is unlikely to introduce selection bias and to affect the validity of our estimates . Our results show that OR estimates for all three STHs are pulled away from the null value when the OR estimates are adjusted for misclassification error . This effect of non-differential misclassification has long been recognized , although this is not always the case when exposure and outcome variables are dependent , a discrete variable assumes more than two values , or there is misclassification error in the confounding variable [21] , [23] , [41] , [42] . The effect of misclassification on the OR estimates of the association between STH and the risk of S . japonicum infection differed for the three STHs under study . The magnitude of impact of misclassification error depends on the sensitivity , specificity , and true prevalence of the variable ( s ) of interest . The relative change in the OR estimates between the unadjusted model and the model adjusting for misclassification error of STH and S . japonicum was larger for hookworm than the other STHs . This is likely to be due to the considerably lower sensitivity ( single stool sample ) of the Kato-Katz for hookworm as compared to that for A . lumbricoides and T . trichiura [43] . Two studies have reported estimates of cross-sectional association between hookworm infection and infection by another schistosome species ( S . mansoni ) . Keiser and others ( 2002 ) reported an OR of 2 . 25 ( 95% CI: 1 . 31 , 3 . 85 ) from their study conducted among 325 school children in Côte d'Ivoire [2] . Fleming and others ( 2006 ) reported an OR of 2 . 95 ( 95% CI: 2 . 19 , 3 . 98 ) from a study conducted among 1332 individuals in Brazil [17] . Their results , which did not adjust for misclassification error , could be due to the cross-sectional nature of their study , which could increase the association between the prevalences of hookworm and schistomiasis . It is also possible that the association between hookworm and schistosomiasis is larger for S . mansoni than for S . japonicum or that the Kato-Katz performs better for the diagnosis of S . mansoni , thus reducing the effect of misclassifaction error . Moreover , temporality of the association could not be ascertained because of the cross-sectional design of these studies . Longitudinal design of our study allowed us to assess the impact of hookworm infection on the incidence of schistosomiasis japonica , after adequate adjustment for misclassification error . Even though the OR may overestimate somewhat the relative risk , these measures are likely to be reasonably close in our study since the risk of re-infection was in the order of 13% . Important changes in OR estimates for other covariates were also observed . The OR estimates for covariates when only S . japonicum data ( outcome ) were adjusted for misclassification error were very close to the OR when both S . japonicum and STH data were adjusted . In contrast , the OR estimates for covariates when only STH data ( exposure ) were adjusted for misclassification error were very close to unadjusted OR estimates . This is because the strength of the association between the covariates and S . japonicum infection was considerably larger than the confounding effect of STH infections . Nevertheless , even correction for misclassification error in the outcome variable only was capable of changing estimate of effect of some of the covariates on the risk of S . japonicum infection . This has important implications for the assessment of the confounding effect of these variables and their association with the risk of S . japonicum infection . We also observed wider confidence intervals for all misclassification error-adjusted ORs . This results directly from incorporating uncertainty in estimating infection status [21] , [44] . The largest impact of misclassification error was observed for the association between hookworm and S . japonicum , which was negligible in the unadjusted model and important on the adjusted one . Several authors have provided numerical examples in their publications showing larger effects of joint misclassification of both exposure and outcome [22] , [41] , [45] . For A . lumbricoides and T . trichiura , OR point estimates indicate a negative and a positive relationship , respectively , but of a smaller magnitude . The efficacy of praziquantel for the treatment of schistosomiasis has been reported to range between 71% and 99% in published literature [46] , [47] , [48] . However , more recent papers have reported an efficacy of praziquantel for the treatment of schistosomiasis around 96% [46] , [47] . The “at-risk” group size is likely to be affected by a lower efficacy as treatment with praziquantel does not completely cure everyone who has the infection . In our study , we assumed 100% efficacy of praziquantel for the treatment of schistosomiasis and decided not to adjust for a lower efficacy of praziquantal . This would have required yet another level of uncertainty for only a small proportion of the population ( the efficacy is very high ) , and is unlikely to have changed our conclusions . Another limitation of this study is that our model assumes conditional independence of test results within each individual given the latent true infection status which is always uncertain . To assess conditional dependence we first have to build a more complex model assuming that there is at least some dependence . This allows examination of the size of the dependence parameter and whether or not its use is meaningful [49] . Exploring such a complex model is beyond the scope of this paper . However , several authors have noted that overlooking conditional dependence does not substantially change parameter estimates [49]–[51] . Our results were adjusted for risk factors most often reported to be associated with schistosomiasis , and often shared with hookworm , such as age , gender , occupation , and the village where people live . Although some additional unmeasured confounding factors may explain the observed association , such factors would need to have a very strong relationship with both hookworm and schistosomiasis to modify our conclusion . Our data suggest that hookworm infection is associated with increased 12-month cumulative incidence of S . japonicum infection . Such important associations might be missed in analyses which do not adjust for misclassification errors . Our findings have important implications for control of these infections in regions where these worms are co-endemic . Effective control of one helminth can lead to reduction in incidence of another and help to reduce the overall burden of helminthic infection in affected regions . | Hookworm , roundworm , and whipworm are collectively known as soil-transmitted helminths . These worms are prevalent in most of the developing countries along with another parasitic infection called schistosomiasis . The tests commonly used to detect infection with these worms are less than 100% accurate . This leads to misclassification of infection status since these tests cannot always correctly indentify infection . We conducted an epidemiological study where such a test , the Kato-Katz technique , was used . In our study we tried to show how misclassification error can influence the association between soil-transmitted helminth infection and schistosomiasis in humans . We used a statistical technique to calculate epidemiological measures of association after correcting for the inaccuracy of the test . Our results show that there is a major difference between epidemiological measures of association before and after the correction of the inaccuracy of the test . After correction of the inaccuracy of the test , soil-transmitted helminth infection was found to be associated with increased risk of acquiring schistosomiasis . This has major public health implications since effective control of one worm can lead to reduction in the occurrence of another and help to reduce the overall burden of worm infection in affected regions . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"infectious",
"diseases/neglected",
"tropical",
"diseases",
"public",
"health",
"and",
"epidemiology/epidemiology",
"infectious",
"diseases/helminth",
"infections",
"public",
"health",
"and",
"epidemiology/global",
"health",
"public",
"health",
"and",
"epidemiology/infectious",
"diseases",
"infectious",
"diseases/epidemiology",
"and",
"control",
"of",
"infectious",
"diseases"
] | 2011 | Assessing the Impact of Misclassification Error on an Epidemiological Association between Two Helminthic Infections |
Identifying human genes relevant for the processing of pain requires difficult-to-conduct and expensive large-scale clinical trials . Here , we examine a novel integrative paradigm for data-driven discovery of pain gene candidates , taking advantage of the vast amount of existing disease-related clinical literature and gene expression microarray data stored in large international repositories . First , thousands of diseases were ranked according to a disease-specific pain index ( DSPI ) , derived from Medical Subject Heading ( MESH ) annotations in MEDLINE . Second , gene expression profiles of 121 of these human diseases were obtained from public sources . Third , genes with expression variation significantly correlated with DSPI across diseases were selected as candidate pain genes . Finally , selected candidate pain genes were genotyped in an independent human cohort and prospectively evaluated for significant association between variants and measures of pain sensitivity . The strongest signal was with rs4512126 ( 5q32 , ABLIM3 , P = 1 . 3×10−10 ) for the sensitivity to cold pressor pain in males , but not in females . Significant associations were also observed with rs12548828 , rs7826700 and rs1075791 on 8q22 . 2 within NCALD ( P = 1 . 7×10−4 , 1 . 8×10−4 , and 2 . 2×10−4 respectively ) . Our results demonstrate the utility of a novel paradigm that integrates publicly available disease-specific gene expression data with clinical data curated from MEDLINE to facilitate the discovery of pain-relevant genes . This data-derived list of pain gene candidates enables additional focused and efficient biological studies validating additional candidates .
A significant number of diseases are associated with pain , thereby affecting the quality of life of many individuals . The Institute of Medicine's recent report titled , “Relieving Pain in America” presented pain as a public health challenge , and emphasized the need for an integrative approach to understand mechanisms underlying pain [1] . Such understanding is critical for developing more effective and individualized strategies targeting the prevention and treatment of pain . Studies in rodents and humans have established the importance of genetic factors in the processing of pain [2] , [3] , [4] . However , identifying genes important to complex phenotypes such as pain using genome-wide association studies has been challenging [5] . Candidate gene studies have identified many gene variants associated with susceptibility to pain [6] , [7] . Despite these advances , genetic discoveries in the domain of pain have been slow in forthcoming compared to other fields [8] . Pain is among the most difficult phenotypes to study due to its complex and subjective nature . The perception of pain is influenced by a multitude of variables including gender , age , mood , ethnicity and genetic factors [9] , and a recent meta-analysis highlighted the overall small effect size attributable to any gene variant associated with the processing of pain [10] . The polygenetic nature of pain and the small effect size of gene variants pose significant challenges for pain gene discovery . Candidate gene studies have proven successful in the identification of pain genes . A particularly promising approach used gene expression microarray analysis to select candidate genes [11] , [12] . More recently a meta-analysis of publicly available microarray data from rodents exposed to neuropathic or inflammatory pain was able to efficiently prioritize pain-related genes [13] . A similar approach using human gene expression data could be highly beneficial in generating data-driven hypotheses for pain genetics . However , there is currently a paucity of public gene expression data related to specific human pain conditions . In this study , we describe an integrative approach exploiting publically available gene expression data for a large set of disease conditions to develop a disease-specific pain index ( DSPI ) . This approach is based on the hypothesis that differences at the gene expression level correlating with pain indices would allow identifying novel pain gene candidates [14] . We validated this approach through a targeted genetic association study in an independent human cohort , where variants of selected pain gene candidates were evaluated for associations with experimental pain sensitivity measures in humans .
We built the disease-specific pain index ( DSPI ) using our literature-based approach , in which 2962 diseases were ranked according to their disease-pain ratio . Table 1 displays the 20 diseases with the highest pain indices . Diseases included Prinzmetal's angina , neuralgia , causalgia , chronic plantar fasciitis and polyarthralgia; all conditions associated with severe pain . Among the diseases ranked at the bottom of the pain index list were fetal alcohol syndrome , cretinism , hermaphroditism and fetal erythroblastosis; all conditions not primarily associated with pain ( Table S1 ) . Inspection of the DSPI indicates that diseases with a high pain index are typically associated with significant clinical pain , while pain is not a hallmark of diseases with a low pain index . As such , the pain index captures relevant aspects of disease-related pain . However , the DSPI relies on the fraction of disease-related publications in PubMed that are associated with the Medical Subject Heading ( MESH ) term “pain” and therefore , is subject to some bias . For example , a disease associated with significant pain but hardly studied in the context of pain would rank inappropriately low on the DSPI list . A practical example is the pain index of cholera with a rank of 2936 , which is at the bottom of the list . Cholera is clearly associated with painful symptoms . Inspection of the DSPI generally revealed relatively low ranks for infectious diseases , likely indicating that the research community predominantly focuses on the most relevant aspects of the condition under study . This suggests that the DSPI also captures to some extend the relevance of pain across multiple diseases . As previously described , the raw microarray data for 311 diseases were extracted from public gene expression databases [15] , [16] , [17] . A list of 3812 differentially expressed ( DE ) genes was then compiled ( see Materials and Methods ) . Pain indices were available for 121 of the 311 diseases with suitable microarray data . The 121 disease-related gene expression changes were ordered according to the DSPI . For each of the 3812 differentially expressed genes , the gene expression fold change across every disease was correlated with the DSPI . This allowed identifying genes whose expression changes were significantly correlated with pain . The sensitivity and accuracy of this strategy for capturing genes implicated in the processing of pain was first evaluated with the aid of the Pain Gene Database ( PGD ) [18] . The PGD catalogs genes whose transgenic or knockout mouse counterparts have exhibited changes in pain-related phenotypes . The PGD is actively maintained and , to our knowledge , is the only pain-related gene database . Figure 1 shows the receiver operating characteristic ( ROC ) curve with confidence intervals . The area under the curve ( AUC ) was 60 . 5% indicating a prioritization of known pain genes from the PGD by our method . We evaluated the significance of the association of the 3812 genes with the DSPI using a threshold-based estimated false discovery rate . Forty-seven genes were significantly associated with the DSPI ( pFDR<0 . 01; Table 2 ) . Among the 47 genes , two genes , DLG4 ( PSD-95 ) and CHRNA4 , were referenced in the PGD [19] , [20] . DLG4 and CHRNA4 were both found to have expression changes in 13 of 121 diseases that were positively correlated with pain indices ( Figure 2A–B ) . In light of this significant but modest prioritization of pain related genes , we applied our pipeline to another medically relevant concept: “Inflammation” . As described above for pain we extracted from MEDLINE a Disease-Specific Inflammation Index ( DSII ) and retrieved gene significantly associated with this index . Using genes belonging to the Gene Ontology category “Inflammatory Response” ( GO:0006954 ) as gold standard we computed the area under the ROC curve . Figure S1 shows a clear prioritization of known “Inflammatory response” genes through our pipeline with an AUC of 73 . 2% . Selected genes from the candidate list were prospectively tested for variants that may be associated with differential pain sensitivity in an independent human cohort . These genes were chosen based on their high correlation with the DSPI and plausible biology as assessed by the available literature and human expression profile across tissue using The Scripps Research Institute BioGPS database [21] . The selected genes were: ( i ) ABLIM3 ( actin binding LIM protein family , member 3 ) , PDE2A ( phosphodiesterase 2A , cGMP-stimulated ) , CREB1 ( cAMP responsive element binding protein 1 ) , NAALAD2 ( N-acetylated alpha-linked acidic dipeptidase 2 ) , and NCALD ( neurocalcin delta ) ( Figure 2C–G ) . ABLIM3 was selected as our top candidate as it showed the highest correlation with the DSPI . The cGMP-sensitive phosphodiesterase PDE2A localizes at the neuronal membrane in synapses and has been described as being regulated by TNFα , a known proinflammatory cytokine shown to sensitize primary nociceptors [22] . Additionally , a recent study on Grueneberg ganglion neurons , that are proposed thermosensors , revealed a key role of cGMP enzyme in cold temperature sensing [23] . Interestingly , PDE2A provides a mechanism for nitric oxide-mediated cGMP synthesis to control intracellular concentrations of cAMP [24] . cAMP is a key second messenger that activate numerous downstream protein , notably cyclic-AMP-response element ( CRE ) -binding protein ( CREB ) that activate classical immediate-early genes such as c-Fos , which are associated with nociceptive afferent activation [25] , [26] . NAALAD2 is highly similar in sequence to NAALAD1 and both hydrolyze N-acetyl-L-aspartate-L-glutamate ( NAAG ) to N-acetyl-aspartate and glutamate , a neuropeptide that activates and antagonizes neuronal N-methyl-D-aspartate ( NMDA ) receptors [27] . Based on nociceptive tests in rats , NAALAD1 was found to plays a role in maintaining mechanical allodynia after carrageenan injection [28] . Of note , NAALAD1 was also positively correlated with the DSPI but not to the same level of significance as NAALAD2 . Finally , NCALD ( neurocalcin delta ) is a calcium-binding protein abundantly and almost exclusively expressed in the central nervous system that has not previously been associated with pain [29] . The genotyping study was conducted in samples obtained from twins enrolled in an ongoing independent IRB-approved pharmacogenomic study testing subjects' sensitivity to experimental heat and cold pressor pain among other outcomes ( see Materials and Methods for details ) . The association study was performed using a generalized least square ( GLS ) test . GLS allowed us to model different variances between monozygotic twin pairs ( MZ ) , dizygotic twin pairs ( DZ ) , and the sexes , as each of these factors has previously been shown to influence pain measures [4] , [30] , [31] , [32] , [33] . Within the five selected genes , 251 tag SNPs were tested . Polymorphisms in ABLIM3 ( rs4512126 ) and NCALD ( rs12548828 , rs7826700 , and rs1075791 ) showed significant association with the cold pressor pain threshold after Bonferroni correction ( Figure 3A–B ) . Linkage disequilibrium ( LD ) analysis of the genotyped SNPs revealed a relatively weak LD structure around these polymorphisms . The LD structure in both genes was similar between the study cohort and the HapMap CEU population for the same region ( Figure S2 and S3 ) . Interestingly , the influence of the rs4512126 loci on the cold pressor pain threshold was tested , which revealed a male specific effect for individuals with the T/T allele ( Figure 4A ) . Males with homozygous T/T alleles exhibited a significantly higher mean pain cold threshold than all other groups ( p = 0 . 005 , 4×10−4 , 0 . 02 , 0 . 005 , 0 . 01 , for A/A Males , A/A Females , A/T Male , A/T Females and T/T Females , respectively ) . The largest effect sizes ( Cohen's d ) were observed between T/T Males and A/A Males and Females ( 0 . 38 and 0 . 39 , respectively ) . Effect sizes between T/T Males and the other groups were below the small effect size threshold ( < = 0 . 2 ) with 0 . 16 , 0 . 11 and 0 . 17 for A/T Males , A/T Females and TT Females respectively .
The primary objective of this study was to demonstrate the utility and validity of a novel , data-driven approach for generating a list of pain gene candidates . Such a list could facilitate the discovery of pain genes . We first validated our approach by demonstrating a statistically significant sensitivity and specificity prioritization of known pain-related genes contained in the Pain Gene Database ( PGD ) . In addition , further genotyping of a human cohort revealed a significant association between variants of the newly discovered pain gene candidates ABLIM3 and NCALD with measures of pain sensitivity in an independent human cohort . A major emphasis of this study was to document the utility of the principal approach and highlight its future potential . The ever growing amount of publically available molecular and clinical data should allow for expanding and refining this approach to generate more comprehensive and specific lists . For example , as more data becomes available , it may be possible to link gene expression of diseases to specific types of pain , such as neuropathic pain . Similarly , the outlined approach can be expanded to include proteomic data sets , which should provide additional insight into signaling pathways relevant to the processing of pain . Finally , the pain associated with a specific disease can be construed differently . For example , disease-specific pain ratings could be retrieved from databases of large health care organizations [34] . There are a few limitations in our approach and study . First , among the 47 candidate pain genes significantly correlating with the DSPI , only two are referenced in the PGD . While the PGD is a valuable resource of curated information and likely represents the best available reference , it is not yet a globally accepted master repository containing all pain genes , especially those resulting from human studies . The database is constrained by the fact that it only catalogs genes revealed by studies examining nociception in mechanistic – but not disease-related – models in knock-out mice . It should also be noted that gene expression data for diseases and matched controls were only available for 121 diseases . As a result only 130 of the 300 genes listed in the PGD could be explored in the current study . The presented paradigm did not capture genes such as KCNS1 , GCH1 , COMT or OPRM1 , each of which has been implicated in the processing of pain [9] . This may partially be due to the fact that the current algorithm favored the discovery of genes exhibiting gradual gene expression change across different diseases . Additionally , our approach relied on gene expression changes in diseased tissue , which may not always capture important changes in secondary tissues relevant for the processing of pain , such as neuronal tissues or blood vessels . Additionally , some of these genes , such as KCNS1 , are thought to be important in specific types of pain like neuropathic pain , but might not participate genetically in determining pain of other etiologies represented in the 121 diseases . There is considerable potential for more refined approaches in the near future to resolve some of these limitations , as there are a constantly growing number of publicly available repositories containing molecular and phenotypic data sets . ABLIM3 is a newly characterized protein-coding gene belonging to the actin binding LIM protein family , which is composed of 3 members ( ABLIM1-3 ) and shows a high degree of conservation throughout evolution in vertebrates . ABLIM3 is expressed in various tissues , most prominently in muscle and neuronal tissue [35] , [36] . While relatively little is known about the biological function of the ABLIM protein family , conservation of key structural features suggests comparable biological function as linkers between the actin cytoskeleton , cell signaling pathways and transcription events [36] . For example , the ortholog of ABLIM1 in C . elegans ( UNC-115 ) has been implicated in axonal guidance during outgrowth through interaction with Receptor for Activated C Kinase ( RACK-1 ) [37] . Presently , a potential functional role for ABLIM3 in the perception or processing of pain is not apparent . ABLIM3 could potentially affect nociceptive signaling by regulating synaptic strength through actin rearrangement and modulation of synaptic spine density [38] . Neuroplasticity has been shown to play a role in pathological pain and to happen both at the molecular and cellular levels [39] . However , the association of ABLIM3 with pain is a novel finding that is based on a data-driven approach but is not anchored in our current understanding of pain biology . While this approach may offer the advantage of making unexpected and important discoveries , it requires establishing the biological relevance of such discoveries in subsequent experimental steps . The SNP rs4512126 ( 5q32 ) is located in the second and largest intron of ABLIM3 . This variant was found in weak linkage disequilibrium ( >0 . 6 ) with five other SNPs and in perfect linkage disequilibrium with rs4546368 located in the same intronic region of ABLIM3 . All were non-coding SNPs . Similar to ABLIM3 , NCALD has never been reported to be associated with pain . However , several polymorphisms in the 3′ UTR have been associated with mRNA instability and diabetic nephropathy [40] . Individuals carrying the A/A allele possessed a higher cold pain threshold . Nevertheless , we acknowledge that our discovered association between NCALD and pain cold was modest , demonstrated in only a single cohort , and barely above the Bonferroni corrected threshold . The parallel approach using quantile normalized phenotypical pain measures did not sustain the association for NCALD ( Figure S4 ) . Further genotyping in alternative cohorts and deep sequencing of these regions would be needed to reveal a potentially causal SNP . Interestingly , only males with homozygous ABLIM3 T/T showed a significant association with cold pressor pain sensitivity in our study . The sex-specific association of a gene variant with the cold pressor pain threshold is not surprising . Genetic polymorphisms associated with pain in humans and animals have identified a striking number of sexual dimorphisms with either male- or female-specific genetic effects , or a significant difference between the sexes [7] , [9] , [30] , [34] , [41] . We present a novel paradigm linking publically available molecular data to clinically relevant phenotypic data for generating a list of candidate genes relevant to the processing of pain . Algorithms for accessing and integrating such data to examine disease-relevant mechanisms are of growing interest as publically available data sets grow at an ever-increasing rate . The outlined approach can complement existing research efforts by assisting the formulation of data-driven hypotheses , and may serve as a template to discover genetic components of other clinically important phenotypes .
The twin study was approved by the Institutional Review Boards of Stanford University and SRI International . All subjects gave written informed consent prior to participation . MeSH is a comprehensive vocabulary thesaurus organized in a hierarchical structure allowing the indexing of publications with various levels of specificity . MeSH terms are used by trained human curators to annotate publications referenced in MEDLINE . We first built a thesaurus of 3743 disease-related MeSH terms using the Unified Medical Language System ( UMLS ) ; restricting ourselves to terms belonging to the following semantic type: Pathologic Function ( T046 ) , Disease and Syndrome ( T047 ) and Neoplastic Process ( T191 ) [42] . Searching MEDLINE for each of these MeSH terms gave us the number of publications published on each disease . For each disease returning a result , we conducted a second search in MEDLINE to count the number of papers published that were also annotated with the MeSH heading term “pain”[mh] . Searching MEDLINE for “pain”[mh] includes publications annotated with any terms hierarchically below “pain” in MeSH , such as Aches , Burning pain , and others . MEDLINE searches were automated using the EUtils programming tools available from the NCBI ( http://eutils . ncbi . nlm . nih . gov ) . The ratio of these two counts formed the disease-pain ratio , as shown in equation ( 1 ) . ( 1 ) The comprehensive disease-specific pain index ( DSPI ) was established by ranking all 2962 diseases by their respective disease-pain ratio ( Table S1 ) . Implicit to our algorithm is the assumption that each disease provides unique qualitative information that may be diluted if weighting results by publication frequency . Thus , the only criteria for inclusion of the disease in the DSPI was to have at least one co-citation with a “pain” related MeSH term ( Figure 5A–B ) . A similar approach was followed to establish the Disease-Specific Inflammation Index ( DSII ) by searching for publication annotated with the MeSH term “Inflammation”[mh] . We annotated all datasets from the National Center for Biotechnology Information ( NCBI ) Gene Expression Omnibus ( GEO ) and the European Bioinformatics Institute ( EBI ) ArrayExpress ( AE ) public databases with UMLS identifiers for diseases as previously published [43] , [44] , [45] , [46] , [47] . We further evaluated these data sets to determine whether or not the submitted biological experiments measured a normal control state ( disease free tissues ) complimentary to the annotated disease state . This was done to ensure that differentially expressed ( DE ) genes could be extracted for each disease . Drug treated samples were excluded from the study . Disease , tissue and substance annotations were manually reviewed in a post-processing step to ensure accuracy . Extraction of data from GEO and AE according to outlined steps revealed 311 diseases explored across 456 publicly available data sets and comprising 14 , 457 individual microarrays from 169 different tissues . In this study , microarrays were pre-processed and DE genes lists were generated using Rank Product ( Figure 5C ) [48] . We kept only genes with a q-value ( gene-specific false-discovery rate ) level ≤0 . 05 [49] . One-hundred twenty-one of the 311 diseases retrieved from GEO and AE were also present in the DSPI list , and thus could be associated with disease-specific pain indices ( see Table S1 for the DSPI and Table S2 for the overlapping list of 121 diseases ) . A small number of animal diseases were present in our DSPI since MEDLINE also covers veterinary and animal diseases . These were automatically excluded from the rest of the analysis , as retrieved gene expression data were limited to humans . Fold change values of the DE genes for each of 121 diseases were organized into a matrix with diseases as columns and genes as rows ( Figure 5D ) . Because the lists of DE genes varied from one disease to another , we defined an arbitrary threshold of minimum gene representation . Information on a gene had to be present in at least 10% of the listed diseases . The final matrix contained 3812 genes as rows and 121 diseases arranged in columns . Finally , disease-columns in the matrix were ordered according to their DSPI rank , i . e . , from the lowest to the highest pain index . The Spearman rank correlation was then computed for each gene using the fold-change values against the DSPI ranking . We computed the positive false discovery rate ( pFDR ) values for each gene by permuting the diseases rank and re-computing the Spearman correlation for each gene . The operation was repeated 1000 times to obtain a null distribution of the correlation coefficients . The pFDR values were calculated as the ratio of the expected proportion of false positive V over the total number of hypothesis rejected R ( 2 ) [50] . ( 2 ) We evaluated the sensitivity and accuracy of our method in prioritizing pain genes using Receiver Operating Characteristic ( ROC ) curves . We compared the DSPI-based pain-gene list against a list of known pain genes from the PGD ( www . jbldesign . com/jmogil/; accessed October 2010 ) [18] . We acknowledge that the PGD only contains data from studies in knockout mice . However , the PGD is to our knowledge the only available repository of pain genes . All 308 mouse genes in the PGD were retrieved manually and translated to 300 human homologs using the NCBI Homologene database ( www . ncbi . nlm . nih . gov/homologene ) . All 300 mouse genes were translated to unique human genes except 8 genes lacking homologs . These eight genes were not further considered . We established the confidence intervals for the ROC curve using a leave-one-out resampling method by repeatedly recalculating the pain-gene rankings with nine-tenths of the 121 diseases . These alternative pain-gene lists were then used to compute the confidence interval of the standard error of our pain-gene ranking . We used the ROCR package from R to compute the ROC curves [51] . Of note , the gene expression measurements on the 121 diseases only included 130 of the 300 known pain genes from the PGD . Gene extracted using the DSII were compared to a gold standard made from the genes belonging to the “Inflammatory response” Gene Ontology category ( GO:0006954 ) . The gene list was retrieved from the Molecular Signature Database [52] . Our study used samples and data from a pre-existing large pharmacogenomic study in twins examining the heritability of various opioid effects [32] . More specifically , data on subjects' sensitivity to heat and cold pressor pain before drug exposure were retrieved . The twin study was approved by the Institutional Review Boards of Stanford University and SRI International . All subjects gave written informed consent prior to participation . In total , 228 healthy pain-free twins ( 114 twin pairs ) of diverse ethnicities were genotyped and phenotyped . We considered only individuals with self-declared European ancestry , which represented 179 individuals with an age range of 18–68 years . Zygosity status of identical ( MZ ) and fraternal ( DZ ) twins was assessed by genotyping concordance using a panel of 47 SNPs [53] . Relevant covariates known to potentially confound measures of pain were also assessed and included demographic factors , age , sex , education , depressed mood , anxiety , sleep , and blood pressure . A detailed description of methods has been published elsewhere [32] . Overall , the cohort used for this analysis was not balanced for sex ( 107 females and 72 males ) and consisted of 54 dizygotic and 125 monozygotic twins . Experimental pain measurements were performed in the Human Pain Laboratory of the Department of Anesthesia at Stanford University School of Medicine . Heat pain was induced with a thermal sensory analyzer ( TSA-II , Medoc Advance Medical System , Durham , North Carolina ) . A 3×3 cm thermode was placed in contact with skin at the volar forearm . Starting at 35°C , the thermode temperature was increased at a rate of 1°C/s . Study participants pushed a button of a hand-held device at the onset of pain . This procedure was repeated 4 times with an inter-stimulus interval of 30 seconds . The average temperature ( C° ) eliciting pain was recorded as the pain threshold . Cold pressor pain is thought to mimic important qualities of clinical pain , since verbal descriptors for both types of pain are strikingly similar [54] . Cold pressor pain is more sustained than heat pain and is associated with a much stronger affective response [55] . The cold-pressor pain model can be viewed as a tool examining an integrated pain response with a strong affective component , while the heat pain model is better suited to explore sensory-discriminative aspects of pain . Sensitivity to cold pressor pain was tested by having subjects immerse their hand to mid-forearm in ice-water ( 1–2°C ) continuously recirculated within a 12-liter container . The palm of the hand was in full contact with the bottom of the container . Subjects were asked to indicate the onset of dull , aching pain typically perceived in the wrist and to withdraw the hand once pain became intolerable . The time ( seconds ) to the onset of pain was recorded as the cold pain threshold and the time to withdrawing the hand was recorded as the cold pain tolerance . The twin cohort was genotyped for 251 SNPs across 5 genes selected from the list of candidate pain genes ( Table 2 ) . We selected LD tag SNPs ( r2 = 1 ) based on the HapMap CEU population using the Tagger software from the Broad Institute ( http://www . broadinstitute . org/mpg/tagger/ ) [56] . Twin's DNA was extracted from peripheral blood lymphocytes and genotyped using a custom-designed Oligo Pool for Methylation Assay ( Golden Gate Genotyping Assay , Illumina , Inc , San Diego , CA ) and BeadXpress ( iGenix Inc . , Bainbridge Island , WA ) . We filtered out all SNPs with a call rate <90% . Out of 251 SNPs assayed only 216 yielded successful results following Illumina quality controls . Some successfully genotyped loci had missing genotypes for a small number of twins and were imputed using the homozygous wildtype allele from the population . We filtered out all SNPs with a minor allele frequency <5% or those whose genotype frequency departed from Hardy-Weinberg equilibrium at p<0 . 01 . In summary , 207 SNPs were tested against heat pain threshold , cold pain threshold and cold pain tolerance . While twin individuals were not required to test association of our candidate genes with pain , twins allowed us to control for environmental variability in pain measurements . Rather than utilize methods to correct for the relatedness of observations coming from the two members of the same twin pair , we used a model in which each pair was treated as a single observation by using within-pair genotype and phenotype averages . Genotypes were transformed to numbers according to the allele frequency , 0 , 1 , 2 for homozygous wildtype , heterozygous and homozygous rare , respectively . Then , the genotypes of twins were averaged within each pair . The genotype of single twin was not altered . Categorical covariate data such as sexes were discretized to −1 and 1 for male and female respectively and then averaged . DZ twin pairs of different sex were coded 0 . To test association of measures of pain sensitivity with each SNP , we used a generalized least square ( GLS ) regression model [57] examining the null hypothesis that pain and genotype are not associated . We regressed on the pain score measured , y , against the genotype for the SNP considered while controlling for depression of the individual ( Beck Depression Index ) and sex ( 3 ) . ( 3 ) GLS allowed us to model different variances of measured traits in MZ and DZ twins and in males and females . We related genotype to the heat pain threshold ( degree C° ) and the log of both the cold pressor pain threshold and cold pressor pain tolerance . P-values were corrected for multiple hypotheses testing using the Bonferroni correction . One twin with the T/T allele for rs4512126 reached the maximum allowed time in the cold threshold test ( 3 min ) . This result was considered an outlier and was removed from the analysis . In parallel , we also evaluated the effect of the pain phenotype normally distributed through quantile normalization . This analysis revealed similar results . The rs4512126 ( ABLIM3 ) polymorphism remained significant for cold pain threshold ( Figure S4A ) . Additionally , rs4512126 and rs7715362 ( ABLIM3 ) showed a significant association with cold pain tolerance ( Figure S4B ) and rs7720260 ( ABLIM3 ) showed a significant association with heat pain threshold ( Figure S4C ) . However , previously found significant polymorphisms in NCALD did not pass the Bonferroni threshold in this analysis . We computed the Cohen's d effect sizes for the difference observed between men and women homozygous for the minor allele of the rs4512126 variant as follows d = [xm−xf]/pooled standard deviation , where xm and xf are the average cold pressor pain threshold in males and females , respectively [58] . Positive values indicate a higher male average pain threshold , and negative values indicate a higher female average pain threshold . The result is unit free and Cohen proposed that benchmark values for what should be considered a ‘small’ , ‘medium’ and ‘large’ effect ( d> = 0 . 2 , 0 . 5 , 0 . 8 , respectively ) [59] . We analyzed differences in mean cold threshold for rs4512126 by performing two-sample t-tests with unequal sample size and unequal variances ( Welch two sample t-test ) . | The mechanisms underlying pain are incompletely understood , and are hard to study due to the subjective and complex nature of pain . From a genetics perspective , the discovery of genes relevant for the processing of pain in humans has been slow and genome-wide association studies have not been successful in yielding significantly associated variants . Targeted approaches examining specific candidate genes may be more promising . We present a novel integrative approach that combines publicly available molecular data and automatically extracted knowledge regarding pain contained in the literature to assist the discovery of novel pain genes . We prospectively validated this approach by demonstrating a significant association between several newly identified pain gene candidates and sensitivity to cold pressor pain . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"text",
"mining",
"public",
"health",
"and",
"epidemiology",
"pain",
"management",
"genetic",
"polymorphism",
"epidemiology",
"anesthesiology",
"population",
"genetics",
"biology",
"computational",
"biology",
"population",
"biology",
"anesthesiology",
"and",
"pain",
"management",
"disease",
"informatics"
] | 2012 | Integrative Approach to Pain Genetics Identifies Pain Sensitivity Loci across Diseases |
We present a computer simulation and associated experimental validation of assembly of glial-like support cells into the interweaving hexagonal lattice that spans the Drosophila pupal eye . This process of cell movements organizes the ommatidial array into a functional pattern . Unlike earlier simulations that focused on the arrangements of cells within individual ommatidia , here we examine the local movements that lead to large-scale organization of the emerging eye field . Simulations based on our experimental observations of cell adhesion , cell death , and cell movement successfully patterned a tracing of an emerging wild-type pupal eye . Surprisingly , altering cell adhesion had only a mild effect on patterning , contradicting our previous hypothesis that the patterning was primarily the result of preferential adhesion between IRM-class surface proteins . Instead , our simulations highlighted the importance of programmed cell death ( PCD ) as well as a previously unappreciated variable: the expansion of cells' apical surface areas , which promoted rearrangement of neighboring cells . We tested this prediction experimentally by preventing expansion in the apical area of individual cells: patterning was disrupted in a manner predicted by our simulations . Our work demonstrates the value of combining computer simulation with in vivo experiments to uncover novel mechanisms that are perpetuated throughout the eye field . It also demonstrates the utility of the Glazier–Graner–Hogeweg model ( GGH ) for modeling the links between local cellular interactions and emergent properties of developing epithelia as well as predicting unanticipated results in vivo .
Epithelial patterning , in which cells assume required positions within emerging epithelia , is essential to the development of all animals . Such patterning results from local interactions that correctly localize each cell using limited molecular information . Simple patterns can employ a single surface factor , often adhesion molecules such as cadherins [1] , [2] . Mathematical models and computer simulations of these processes based on local reduction of free-energy can replicate experimentally observed cell shapes within epithelia as diverse as embryonic germ layers and Drosophila ommatidial patterns [3] , [4] , [5] , [6] , [7] , [8] . However , these models do not address cell placement , which commonly plays a key role in producing functional tissues for example in the mammalian and insect retinas [9] , [10] . Recently , we proposed that multiple adhesion molecules expressed in precise spatial patterns can generate more complex patterns via local energy minimization [11] . Such models self-organize based on a small number of cell and global properties . However , we did not verify that such forces could control the arrangement of cells in a complex pattern [11] . The Drosophila eye is a striking example of cell placement , in which every cell has a stereotyped position . It is a mosaic of approximately 750 precisely organized ommatidia ( Figure 1A ) . Each ommatidial core ( OC ) contains eight photoreceptor neurons and six glial-like support cells ( two primary pigment cells ( 1°s ) and four cone cells ( c ) ; Figure 1B ) that aggregate in the larval and early pupal eye anlage . To explore assembly of the cone cell quartet Kafer et al . used Glazier-Graner-Hogeweg ( GGH ) -based simulations to model experimental evidence that , within each OC , cone cells assemble into precise quartets through homophilic Cadherin-based adhesion and tension in the cells' actin cortices [3] . Cone cell assembly , similar to the packing of soap bubbles , is therefore an example of short-range , surface-energy-driven aggregation of like cells into a cluster . This local phenomenon demonstrates the ability to utilize local cell-cell angles as an indication of the local processes that assemble discrete cell clusters [3] , [6] . Considering local cell shape changes leaves open the broader question , however , of how multiple processes— including the dynamic aspects of cell adhesion , cell death , cell movement , and cell shape changes— act in concert to achieve progressive , coordinated patterning across an epithelium . Here we focus on the emergence of a complex , interweaving array that organizes the visual field . After OC assembly , an interweaving hexagonal lattice of secondary ( 2° ) and tertiary ( 3° ) pigment cells ( collectively termed inter-ommatidial pigment cells , IPCs ) and sensory bristles ( Figure 1F ) develops that re-organizes the retinal field . Cell re-arrangements begin at the tissue's surface [12] and the lattice eventually extends the length of the eye field to optically isolate neighboring OCs . The emergence of this IPC lattice during pupal development requires regulated cell adhesion , directed cell movement , PCD , and changes in cell shapes ( reviewed in [13] , [14] , [15] , [16] ) . IPC patterning provides a useful model for exploring the patterning mechanisms required to assemble emerging epithelial tissues . This paper simulates the known parameters that organize cells into a locally repetitive pattern that connects across an entire tissue , an aspect of development that has not been previously modeled . Ordered assembly of the IPC lattice requires members of the Nephrin superfamily of proteins , which include the heterophilic type I transmembrane proteins Hibris ( Hbs ) and Roughest ( Rst; [11] , [17] ) . Experimentally , altering the levels of Hbs and Rst in the developing pupal eye disrupts patterning of the IPC lattice [11] . By analogy with Steinberg's Differential Adhesion Model of homophilic adhesion [1] , we proposed a Preferential Adhesion Model of the assembly of the hexagonal lattice based on heterophilic adhesion between cells expressing Hibris ( 1°s , cone cells ) and Roughest ( neighboring IPCs; [11] ) . In this model , IPCs reduce their contacts with each other and enhance their contacts to neighboring 1°s to promote Hbs/Rst binding , reducing their local free energy . We tested this model using simulations based on the Glazier-Graner-Hogeweg model ( GGH , also known as the Cellular Potts Model , CPM ) , a multi-cell methodology that allows for non-uniform cell shapes . The GGH successfully reproduces much of the phenomenology of differential adhesion-based cell sorting [5] , [18] , [19] . We show that simulations using this methodology can replicate the step-wise patterning of ommatidia within the pupal eye field . Surprisingly , our simulations suggest that preferential adhesion is not sufficient to position IPCs . As anticipated , the simulations demonstrated that selective cell death provides robustness to patterning , in particular the movement of 3°s into their final niche . More surprising , robust simulated patterning required a previously unappreciated mechanism: steady expansion of the OCs' surface profiles that effectively ‘crowd’ neighboring IPCs into a proper hexagonal array . Following this suggestion , we used over-expression of the ubiquitin ligase Smurf to prevent normal surface expansion of the OCs; neighboring IPCs failed to pattern properly as our simulations predicted . Thus , our in silico predictions suggested the importance of specific mechanisms in development and led to new experiments that confirmed the importance of these mechanisms , demonstrating that developmental simulations can predict novel mechanisms and indicate specific experiments required to assess them in vivo .
The Drosophila pupal eye is a post-mitotic , pseudo-stratified epithelium within which every cell eventually achieves a stereotyped position . OCs assemble fully by the young pupal stage: photoreceptor neurons and cone cells emerge in the larva , and the 1°s complete each core by enwrapping the cone cells at the surface by 21–22 hours after puparium formation ( APF; Figure 1D ) . Ablation studies have demonstrated that the photoreceptors do not contribute to the surface hexagonal patterning [20] . Between 21 and 24 hours APF , cell rearrangements and programmed cell death refine the remaining interommatidial precursor cells ( IPCs ) to a single row ( Figure 1E; [12] , [16] , [21] ) . By 30 hours APF , further cell rearrangements and deaths pare the pattern down to a single cell for each side of the hexagon and a single cell at each vertex ( Figure 1F ) . Roughly one-third [22] of the cells present at the beginning of pupal development die via PCD before the eye fully patterns . Cell division is essentially complete by this stage , so the hexagonal pattern primarily results from cell rearrangements and deaths . Our simple two-dimensional simulation of eye development implements the free-energy mechanisms proposed previously [11] using CompuCell3D [23] , [24] , an open-source implementation of the GGH . We therefore focused on patterning at the apical surface of the epithelium for two reasons . First , as the pupal eye develops , cell-cell contacts begin at apical surfaces then extend basally through the epithelium [12] . Second , most factors known to initiate patterning are localized to the apical surface including Notch , EGFR , Hibris , Roughest , Cindr , Pyd , etc . [11] , [25] , [26] , [27] , [28] . The GGH is a multi-cell model that has accurately reproduced cell sorting based on differential cell adhesion [5] , [18] , [19] . It represents each ‘cell’ as a collection of points in a two-dimensional grid ( [29] , [30]; see Methods ) . The spatial configuration and physical properties of these cells determine the ‘energy’ landscape of the pattern . Cells move by extending and retracting apical extensions , favoring changes that reduce the local pattern energy . The rate of these extensions determines the timescale of the simulation , measured in Monte Carlo Steps ( MCS ) . In our simulations we used MCS to represent developmental time; we define the relationship between the two below . Laser ablation studies demonstrated that the key IPC patterning interaction is between ( i ) the cone cells and 1°s of the OCs and ( ii ) IPCs [20] . In defining IPCs we noted that previous experiments demonstrated ( i ) bristles are nonessential for patterning and ( ii ) 2°s and 3°s are molecularly indistinguishable , differing only in their positions within the hexagonal pattern ( Figure 1B; [20][31] , [32] ) . We therefore simulated two cell types: OCs and IPCs . We assessed patterning accuracy by monitoring the emergence of 2°s and , separately , 3°s . In the GGH , adhesion takes the form of a boundary energy ( J; see Methods ) . A higher boundary energy corresponds to weaker cell-cell adhesion . While absolute values for experimental cell-cell adhesion strengths are unknown , previous experiments have found stronger adhesion at OC:IPC contacts than at IPC:IPC contacts [11] . Stronger OC:IPC adhesion prevents contact between neighboring OCs . A ring of IPCs bounds the eye field . Based on these observations , we initially assumed a hierarchy of contact energies: JOC , OC>>JOC , Medium>>JIPC , Medium>JIPC , IPC≥JIPC , OC , where the ‘Medium’ cell type represented the intracellular space surrounding the edge of the developing imaginal disc , allowing tissue expansion . We initially assumed OC:IPC adhesion to be stronger than IPC:IPC adhesion to prevent OC fusions; we later tested this assumption as well . Finally , a broad array of values for OC:OC adhesion were tested and yielded similar results . Pupal epithelial cells have an intrinsic cytoskeleton-driven motility that appears in continuous live imaging as jostling— short range undirected movements— within the pupal eye field [33] . This motility likely depends on interactions between surface junction factors and the actin cytoskeleton [34] , [35] . The GGH represents this intrinsic motility by the parameter T ( see Methods ) . A larger T corresponds to higher cell motility in the simulation [36] . Though the mechanisms that direct PCD of specific IPCs are not fully understood , PCD depends on levels of Rst [14] , [32] , [37] , [38] . In the GGH , we simulate PCD by changing a cell's target area ( AT ) to 0 [39] , causing the cell to shrink and disappear . We implemented a simple PCD rule based on two experimental observations: ( i ) 1°s are necessary for cell survival ( ablation of 1°s leads to PCD of all adjacent IPCs [20] ) and ( ii ) reduction of a cell's apical surface area predicts PCD [32] . Regarding the latter , reduction of the apical profile of emerging IPCs was sufficient to increase the likelihood of an emerging IPC's death [11] . In our simulations , the probability that an IPC ‘died’ via PCD increased when its contact with OCs became smaller than a threshold length ( L; see Methods ) , thus mimicking the observed biological behavior . Experimentally , IPCs in the periphery of the pupal eye do not undergo PCD in the young pupa [22] , [40]; we therefore prevented the IPCs at the edges of simulated eyes from dying regardless of the extent of their contact with the OCs . As the developing pupal eye matures , the apical cell-surface profile ( that is , the cross-sectional area ) of the OCs increases roughly linearly in time due to apical migration of cone and 1° cell nuclei ( Figure 2A , compare Figure 1D , E and F; [41] ) . In our experiments , the average area of each OC approximately doubled between 22:00 and 27:30 hours APF , while the apical area of the IPCs remained approximately unchanged ( Figure 2B ) . To simulate this steady increase in apical IPC profiles we observed in situ we doubled the target area and target perimeter of simulated OCs in a linear progression ( Table S1 ) . We traced micrographs of wild-type eyes at 23:00–24:30 hours APF ( Figure 2C , 2D; see Methods ) . By this stage 1°s have fully enveloped each ommatidium , the pattern of 2°s has advanced but is not complete , and most 3°s have yet to move to the vertices . We used a representative tracing as the initial condition of our simulations , then ran for 50 , 000 MCS . We initially tested several values for adhesion energies and temperature to establish the range that yielded patterns closest to experiment ( Table S2 ) . The number of ( i ) IPCs that established themselves correctly in vertices as 3°s and ( ii ) single cells that occupied a 2° cell locus provided straightforward measures of pattern precision . Importantly , all wild-type patterning was essentially complete by 20 , 000–30 , 000 MCS . Thus 10 , 000 MCS is equivalent to approximately 5 . 5 hours . We ran simulations to 50 , 000 MCS to ensure that we were observing a pattern in equilibrium and to provide additional steps for simulations employing ‘mutant’ parameters . To determine values for adhesion , we assumed the hierarchy described above and tested various values ( Table S2 ) to yield cell shapes and placement similar to those observed in wild-type retinae ( Table S3 ) . Altering the cell mobility parameter over a broad range did not affect the positions of cells at the end of our simulation: 3°s formed between 25<T<150 , so we selected an intermediate value of T = 60 for further simulations . By varying the threshold contact length for induction of PCD over a range of motilities , we found that L = 16 generated ommatidia with the proper number of cells with occasional surplus IPCs that were confined to 2° niches , an observation consistent with the occasional ectopic 2°s observed in mature wild type pupal eyes ( Figure 1F ) . We delayed PCD until the simulation had run for 10 , 000 MCS to equilibrate cell sizes and remove any bias in cell death due to the initial configuration . However , delaying the onset of PCD did not affect the outcome of the simulations ( data not shown ) . Using an adhesion hierarchy in which IPCs adhered more to OCs ( JOC , IPC = 35 ) than to each other ( JIPC , IPC = 55; Table S3 ) produced a striking phenocopy of a wild type eye with a single 3° in most vertices ( Figure 3 and Supplemental Video S1 ) . This final pattern was consistent over multiple runs and reached a steady-state equilibrium after 50 , 000 MCS ( equivalent to approximately 41 hours APF in the developing pupa ) . Most patterning was completed by 30 , 000 MCS , which corresponds to approximately 30 hours APF , mirroring the time frame of development in vivo . Having determined parameters that replicated wild-type development , we next tested the contributions to patterning from different mechanisms by varying parameters from the wild-type values . To reduce PCD , we decreased the threshold length of OC:IPC contact required for survival , resulting in ectopic 2°-like cells consistently positioned end-to-end ( Figure 4A , 4C ) . Further , we observed a direct correlation between the number of cells present in the simulation and ( i ) the number of single cells occupying a 2° cell locus ( Figure 4C ) and ( ii ) the number of 3°s eventually established ( Figure 4D ) . As cell death was increased toward normal levels in the simulation the number of successfully established 2°s and 3°s increased as well ( Figure 4E ) . To test these predictions experimentally , we blocked PCD during pupal eye patterning with the caspase inhibitor DIAP1 . Despite the greater number of IPCs , interommatidial cells still assembled into an interweaving hexagonal lattice as previously shown [37] , [42] . Also as observed in our simulations , ectopic 2°s were found primarily end-to-end in experimental tissue ( Figure 5A–D ) . PCD inhibition that produced three excess IPCs within a standardized region ( see Methods , Figure 5 ) led to occasional misplacement of a single 2° cell and 3° vertex in pupal eyes . However when the number of IPCs exceeded 22 within this standardized region—an excess of 10 or more IPCs—we rarely ( 0 . 6% ) observed the full complement of three correct 3°s and never observed six correct 2°s around an ommatidium . The number of 2°s and 3°s in vivo was progressively reduced as the number of IPCs increased: for example , nearly five correctly positioned 2°/3°s were observed in tissue with 15 IPCs per ommatidial region; this number decreased to less than one 2°/3° in regions with more than 24 IPCs ( Figure 5E ) . The ability of our model to correctly predict this relationship provides further validation of our simulation parameters . Our recent work has suggested that the preferential adhesion of IPCs ( expressing Rst ) to 1°s ( expressing Hbs ) is a major driving force in IPC patterning [11] . We assessed this ‘Preferential Adhesion’ model by simulating altered adhesion . We simulated three types of adhesion between OCs and IPCs: ‘preferential’ ( JOC , IPC<JIPC , IPC ) , ‘flat’ ( JOC , IPC = JIPC , IPC ) , or ‘anti-preferential’ ( JIPC , IPC<JOC , IPC ) . Remarkably , all three adhesion hierarchies led to correct or nearly correct assembly of 2°s and 3°s , though with decreasing levels of robustness ( Figure 6 ) . The reduced adhesion between IPCs and OCs in the ‘flat’ and ‘anti-preferential’ adhesion simulations led to decreased average IPC:OC contact length , increased number of cells dying by PCD , and increased frequency of defects due to missing 2° cells in ( Figure 6B and C ) . To check whether cell death masked the patterning effects of differential adhesion , we repeated the different adhesion simulations with a reduced rate of PCD that produced an intermediate number of 3°s . Under these conditions , ‘preferential’ adhesion resulted in only slightly better patterning than either ‘flat’ or ‘anti-preferential’ ( Supplemental Video S2 ) , as assessed by the number of correctly located 2°s and 3°s ( Figure 6D , 6E ) and the accuracy of the overall hexagonal patterning . Our surprising conclusion is that in silico ‘preferential adhesion’ contributes to patterning robustness but that it is not sufficient to create the hexagonal pattern . We were unable to recapitulate our different in silico adhesion conditions in vivo , perhaps because manipulating the levels of Rst and Hbs led to confounding non-adhesion-related effects potentially due to aberrant signaling [11] , [27] , [28] . The failure of PCD plus preferential adhesion to fully account for patterning within our simulations suggested the importance of additional mechanisms . In re-assessing our experiments , we observed that ommatidial apical profiles expanded significantly as OC nuclei migrated apically throughout the patterning period , while IPC profiles remained roughly constant ( Figures 1C–F , 2B ) . We therefore explored the contribution of OC surface expansion to 2° and 3° formation . Blocking OC expansion in silico led to indiscriminant death of IPCs ( data not shown ) , since IPCs had very little available OC surface to bind to and thus died according to our PCD rules . As we discussed above , blocking cell death but retaining OC expansion permitted the emergence of at least some 2° and 3°s ( Figure 7A , 7B ) . Eliminating both OC expansion and cell death , however , blocked emergence of nearly all 2° and 3° cells ( Figure 4 , 7A , 7B and 7D ) . This result indicates that expanding OCs play an obligate role in simulated IPC patterning . If the expanding apical surface area of OCs relative to IPCs promotes 3° formation in silico , then increasing both OC and IPC sizes concurrently should inhibit lattice patterning . Having IPCs slowly double in size over the course of the simulation as OCs grew ( Supplemental Video S3 , Table S1 ) significantly reduced the number of correctly specified 2° and 3°s ( Figure 7A , 7B ) . This reduction did not appear to result from the slight decrease in cell death we observed with expanding IPCs ( compare Figure 4B to Figure 7C ) . As schematized in Figure 7E , the size of a cell's apical surface profile is closely tied to how close its nucleus is to the surface: the nucleus makes up the large cross-sectional area of the cell ( Figure 2A; [43] , [44] ) . To test whether the changing relative sizes of OCs and IPCs contributes to eye patterning in vivo , we identified a mutant in which the surface profiles of 1°s failed to properly expand . The smurf/lack locus encodes an ubiquitin ligase with several functions including degradation of the cytoskeleton regulator Rho1 [43] , [44] . Expressing ectopic smurf in isolated 1°s led to an autonomous reduction in their apical surface areas ( Figure 7F , 7G , 7H ) . Attempts to manipulate cell size by modulating insulin signaling or nuclear positioning by perturbing marbles failed to consistently alter apical surface areas ( data not shown ) . As our simulations predicted , reducing the apical surface area of one or more 1°s led to local mis-patterning of the neighboring , genotypically normal IPC lattice . In 69% of ommatidia with isolated smurf-expressing 1°s , neighboring IPCs patterned incorrectly ( N = 87 ) . In control GFP-expressing 1°s , by comparison , IPC errors were observed in 3 . 5% of ommatidia ( N = 502 , not shown ) . Patterning defects included loss of IPCs , mis-patterned 2°s , and a reduction in proper 3°s ( Figure 7G , 7H ) . We observed analogous defects in GGH simulations in which half of the OC was designed to not expand ( Figure 7J , Supplemental Figure S1 ) . Interestingly , symmetric OC reduction led to milder IPC patterning defects in our simulations , suggesting that balanced expansion across the local field is necessary for proper patterning ( Figure 7I ) . Again consistent with the GGH prediction , expression of smurf in both 1°s within an ommatidium had a less severe effect on patterning ( Figure 7F ) . In addition to further validating GGH predictions , this result also indicates that the in vivo IPC defects were not due to unanticipated effects of reduced smurf in 1°s . Together , our data support the view that properly expanding OCs are a central component of IPC patterning . While our simulations suggest that preferential adhesion contributes to the formation of 2° and 3° cells within the eye , we also observed that these cells can pattern without preferential adhesion . This result raises the question as to why rst mutant phenotypes are stronger than the mild effects generated in our ‘flat’ ( JOC , IPC = JIPC , IPC ) adhesive paradigm . If the model's prediction is correct regarding adhesion , the rst locus may have activities in addition to adhesion as proposed for Neph1 ( e . g . , [45] , [46] , [47] ) and/or it may also be required for earlier stages of patterning . To test the latter prediction , we traced a 20 hr APF genotypically rstCT mutant eye and used it as the starting point for our standard wild type simulation . Interestingly and unlike wild type tracings , simulations using tracings of rstCT eyes as their initial conditions consistently yielded uniformly incorrect final patterns even after 50 , 000 MCS ( Figure 7K ) . Thus , modeling of rst mutant eyes predicts that rst acts prior to 20 hour APF , during stages when adhesion-mediated IPC movement is not thought to occur . Indeed , visual inspection of 20 hr APF rstCT mutant eyes indicated consistently abnormal OC spacing ( e . g . , Figure 7K; data not shown ) .
Generation of a hexagonal lattice of 2°s and 3°s requires interactions between two different cell types . Using a relatively small number of physical mechanisms demonstrated to regulate patterning in vivo , we produced a cellular configuration that replicates the precision of the Drosophila IPC lattice in silico . Most gene mutations alter multiple mechanisms simultaneously , such as adhesion plus cell death ( e . g . , rst , pyd; [11] , [27] , [28] ) ; our simulations allowed us to isolate and examine each mechanism separately to explore its role and importance in patterning . In addition to supporting the central importance of selective programmed cell death and its relationship to 1°:IPC contacts , our simulations highlighted the importance of a previously overlooked mechanism , challenging our current understanding of how cells organize within the emerging eye epithelium . It has also caused us to re-evaluate the role played by rst in IPC patterning . Table S4 compares our simulations with in vivo results . Although the parameters utilized for proto-2°s and proto-3°s were identical our simulations correctly introduced a single cell into each 3° niche , suggesting that the 3° fate results from cell sorting within the eye and does not necessarily reflect a distinct differentiation state . The lack of experimental molecular markers that distinguish 2°s from 3°s is consistent with this simulation result . Further , while blocking PCD experimentally had little effect on the overall pattern of the lattice [14] , [32] , [37] , [38] , our simulations suggested that PCD is a major driving force behind establishment of 3°s . In our simulations , the number of 3° errors was directly proportional to the number of ectopic IPCs . We confirmed the accuracy of these predictions by reducing PCD experimentally during patterning of the pupal eye ( Figure 5 ) . Our simulations also indicated the importance of a novel patterning component: the progressive increase in the OC:IPC apical surface profile ratio . A role for this process in eye patterning has been neither previously suggested nor explored . Our results suggest that expansion of the ommatidial array helps generate a hexagonal pattern by ‘crowding’ IPCs into elongate shapes that encourage proper cell death and correct 2° and 3° formation ( Figure 7E ) . Our model predicts that asymmetric changes in 1°s will alter this pattern while symmetric changes will exhibit milder effects ( Figure 7I , 7J ) . We have validated this prediction experimentally ( Figure 7F , 7G , 7H ) . Nuclear movements are commonly tightly coordinated during tissue maturation . For example , during emergence of the developing brain , nuclei move in a coordinated fashion ( ‘nucleokinesis’ ) that both indicates and is perhaps necessary for progressive cell specification , cell cycle , and stratification ( e . g . , [48] , [49] , [50] , [51] , [52] ) . Work in cultured neurons suggests this process requires dynamic interactions between components of the cytoskeleton , the nucleus , and cell adhesion during neuronal movement ( e . g . , [52] , [53] , [54] , [55] ) . These dynamic nuclear movements can lead to dramatic changes in cell shape: in many columnar epithelia including the developing Drosophila eye , the nucleus occupies the cell's region of greatest cross-sectional area and its movement can strongly distort the local environment . When harnessed in the fly pupal eye , this distortion provides necessary patterning information . While the role of nuclear movements in cell fate specification and movements is only beginning to be appreciated , our results suggest that the resulting changes in cell packing—for example apical surface area—can strongly influence precise cell placement . We speculate that this mechanism is commonly employed in other emerging epithelia as well .
In the Glazier-Graner-Hogeweg ( GGH ) model used for our simulations [5] , [18] , [19] , each cell exists as a group of points on an underlying lattice ( for our simulation code and configuration files see Protocol S1 ) . Exchanging lattice sites between adjacent cells at their boundaries randomly perturbs the shape and location of the cells . The probability that a proposed exchange occurs depends on its effect on the energy of the lattice , as determined by an energy function H . If a copy reduces the energy , we accept it with probability 1 . Otherwise , we accept it with probability e−ΔH/T , where T , represents the intrinsic motility of the cells [5] , [18] , [19] . Each potential exchange is analyzed in a random order , with one exchange attempt for each lattice site defining the simulation's unit of time , a Monte Carlo Step ( MCS ) [5] , [18] , [19] , here equivalent to about 2 seconds . Each cell possesses a cell type , τ , which determines its physical properties and the contribution the cell makes to the overall energy of the lattice . Our simulations include energies due to cell adhesion and cell-area and cell-perimeter constraints . Adhesion provides a mechanism for building complex structures , as well as for holding them together once they have formed . To represent variations in energy due to adhesion between cells of different types , we define a boundary energy that depends on , the boundary energy per unit length between two cells ( ) of given types ( ) at a link ( the interface between two neighboring lattice sites ) : ( 1 ) where the sum is over all neighboring pairs of lattice sites and ( the neighbor range may be greater than one lattice site ) , and the boundary-energy coefficients are symmetric , ( 2 ) Table S3 lists the contact energies in our simulations . To restrict variation of cell area we use an area constraint , of the following form: ( 3 ) where for cell , denotes the inverse compressibility of the cell , is the number of lattice sites in the cell ( its area ) , and is the cell's target area . This constraint defines as the pressure inside the cell , that is , the susceptibility to grow or shrink in subsequent steps . A cell with has a positive internal pressure and thus grows , while a cell with has a negative internal pressure and thus shrinks . Since the experimental cells have nearly fixed amounts of cell membrane , we use a surface-area constraint of form: ( 4 ) where is the surface area of cell , is its target perimeter . is its inverse membrane compressibility . Adding the boundary energy and area constraint terms together ( Equations ( 1 ) , ( 3 ) and ( 4 ) , we obtain the basic GGH effective energy: ( 5 ) These equations are the 2D form of the canonical GGH equations [24] . The implementation of the GGH that we used for our simulations supports both 3D and 2D forms and detects which to use appropriately [24] . Table S5 lists the area and perimeter constraints used for our simulations . These constraints were selected to ensure that OCs maintained a roughly circular shape as observed in vivo [33] and allow minimal constraints on the shape of the IPCs . We simulate cell death by setting to 0 [39] . We used the open-source program , CompuCell3D , to implement the GGH ( www . compucell3d . org ) . CompuCell3D leaves implementation of dynamic properties of cells ( e . g . cell growth rate , cell death rate , etc . ) to the user . Our simulations implemented apical surface expansion as a CompuCell steppable , a callable software module executed a fixed number of times per MCS . In our implementation , we incremented the target area and/or target perimeter of cells of specific cell types by a defined amount until the target area reached a maximum . Table S1 lists the expansion parameters in our simulations . Our goal was to determine the parameters necessary to obtain stable 3°s . In vivo , beginning at 23:00–24:00 hours APF , the IPCs of the eye reorganize so that each IPC touches at least two 1°s [11] , [12] , [16] . Since we were unsure if this event created a configuration necessary and sufficient for 3° formation or merely increased the number of 1°s contacted by each IPC , we created our initial simulation configuration by tracing micrographs of dissected pupal eyes staged between 23:00–24:30 APF and stained with antibodies recognizing an adherens-junction marker ( Figure 2B and C ) . A custom program was used to convert tracings into a CompuCell3D-readable format . While we traced several eyes , we used a single representative tracing as the initialization file for each simulation in this study . IPC cells die with a probability determined by their contact length ( in lattice units ) with neighboring OCs . If the contact area is greater than a threshold length ( L ) , P ( Death ) = 0 , otherwise: ( 6 ) where CSOC denotes the contact length between an IPC cell and its OC neighbors , λDeath scales this dependency and PMAX ( Death ) is the maximum probability an IPC cell will die per MCS . We evaluated each IPC cell for PCD once after each MCS and cells which died had their target area set to 0 . All simulations used λDeath = 1 . 2 and PMAX ( Death ) = 0 . 01 . We measured average OC and IPC sizes from images of live wild type retinae ( GMR-Gal4/UAS-αCatenin-GFP ) acquired as described in [33] . We outlined cells by hand and measured and recorded the area encompassing the cell using ImageJ ( NIH ) . For each relevant case we measured at least 22 different OCs and 48 different IPCs . To keep track of patterning in our simulations , we plotted the total number of cells , the number of 2° and 3° cells and the average area and perimeter for each cell type at 500 MCS intervals . We scored as 3°s all cells of type IPC that contacted exactly three OCs . We scored as 2°s all cells of type IPC that contacted exactly two OCs and two 3°s . We assembled videos from individual PNG files using the Mencoder program , part of the MPlayer software package ( www . mplayerhq . hu ) and converted them to MPEG format using ffmpeg ( ffmpeg . mplayerhq . hu/ ) . To count the number of IPCs in the lattice surrounding a single ommatidium , we connected the centers of the surrounding six ommatidia to form an hexagonal outline on micrographs of dissected pupal eyes ( Figure 5 ) staged to 41:00 hours APF at 25°C and stained with antibodies to DE-Cadherin ( DSHB ) as described in [27] . We counted cells within each hexagonal outline , arbitrarily scoring any cell that traversed the outline boundary as half a cell . To exclude any potential affects that bristle groups have on PCD [32] we analyzed only ommatidia with 3 correctly-positioned bristle groupings . We scored as 3°s all IPCs that contacted exactly three 1°s . We scored a total of 479 ommatidia and plotted the average number of 3°s per ommatidium . We heat shocked hs-Flp; act>y>Gal4 , UAS-GFP/+; UAS-smurf/+ pupae at 37°C for 20–30 min at 18:00–20:00 hr APF and dissected them at 40:00–42:00 hr APF . We used antibodies to Armadillo ( N2 7A1 , DSHB ) to visualize adherens junctions as previously described [56] . We dissected wild type Canton S pupal eyes at 18:00 hr APF , incubated them in a glutaraldehyde/potassium permanganate fix and processed and imaged them as described previously [12] . | During development , organs are assembled through a complex combination of cell proliferation , programmed cell death , cell movements , etc . These aspects of tissue maturation must be achieved with a limited gene set—to achieve complexity , tissues utilize patterning mechanisms . That is , “rules” are used to create heterogeneity in initially homogeneous cell populations . A large number of genes and cell biology mechanisms have been uncovered that mediate this process but we have a limited understanding of how these factors act together to generate the large-scale patterns necessary to create a useful organ . Here , we combine computational modeling with in situ experiments in the developing Drosophila eye to explore these issues . Computer modeling is often criticized for describing known outcomes . We demonstrate how the Glazier–Graner–Hogeweg model can successfully predict surprising outcomes contradictory to models that emerged from our previous studies . We then validated these predictions in the developing eye . These mechanisms , which include the importance of dynamic nuclear movements , may prove generally important in directing cells into their proper niches as developing epithelia mature . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"developmental",
"biology/morphogenesis",
"and",
"cell",
"biology",
"cell",
"biology/morphogenesis",
"and",
"cell",
"biology",
"developmental",
"biology/pattern",
"formation",
"cell",
"biology/developmental",
"molecular",
"mechanisms",
"computational",
"biology"
] | 2010 | Computer Simulation of Cellular Patterning Within the Drosophila Pupal Eye |
Upon viral infection , the mitochondrial antiviral signaling ( MAVS ) -IKKβ pathway is activated to restrict viral replication . Manipulation of immune signaling events by pathogens has been an outstanding theme of host-pathogen interaction . Here we report that the loss of MAVS or IKKβ impaired the lytic replication of gamma-herpesvirus 68 ( γHV68 ) , a model herpesvirus for human Kaposi's sarcoma-associated herpesvirus and Epstein-Barr virus . γHV68 infection activated IKKβ in a MAVS-dependent manner; however , IKKβ phosphorylated and promoted the transcriptional activation of the γHV68 replication and transcription activator ( RTA ) . Mutational analyses identified IKKβ phosphorylation sites , through which RTA-mediated transcription was increased by IKKβ , within the transactivation domain of RTA . Moreover , the lytic replication of recombinant γHV68 carrying mutations within the IKKβ phosphorylation sites was greatly impaired . These findings support the conclusion that γHV68 hijacks the antiviral MAVS-IKKβ pathway to promote viral transcription and lytic infection , representing an example whereby viral replication is coupled to host immune activation .
Host cells activate innate immune signaling pathways to defend against invading pathogens . Pattern recognition receptors , including Toll-like receptors and cytosolic sensors ( such as NOD-like receptors and RIG-I-like receptors ) , recognize pathogen-associated structural components and initiate signal transduction that leads to the biosynthesis and secretion of pro-inflammatory cytokines and interferons , thereby mounting a potent host immune response [1] , [2] . To survive within an infected host , viruses have evolved intricate strategies to counteract host immune responses . Herpesviruses and poxviruses have large genomes and therefore have the capacity to encode numerous proteins that modulate host immune responses . Mitochonrial antiviral signaling ( MAVS , also known as IPS-1 , VISA , and CARDIF ) protein serves as an adaptor to activate both the NFκB and interferon regulatory factor ( IRF ) pathways [3] , [4] , [5] , [6] . MAVS relays signals from RIG-I and MDA-5 , cytosolic sensors that recognize viral dsRNA or ssRNA bearing 5′-triphosphate [7] , [8] , to the IKKα/β/γ and TBK-1/IKKε ( also known as IKKi ) kinase complexes [4] , [6] . IKKα/β , together with the scaffold protein IKKγ , phosphorylates the inhibitor of NFκB ( IκB ) and promotes its subsequent ubiquitination and degradation by the proteasome , thereby unleashing NFκB that translocates into the nucleus to activate gene expression of pro-inflammatory cytokines [9] , [10] . By contrast , TBK-1 and IKKε directly phosphorylate a serine/threonine-rich sequence within the carboxyl termini of IRF3 and IRF7 , leading to the dimerization and nuclear translocation of these transcription factors [11] , [12] . Together with NFκB and c-Jun/ATF-2 , IRF3 and IRF7 bind to the interferon ( IFN ) -β enhancer and initiate the transcription of IFN-β [13] , [14] . Ultimately , these signaling events promote cytokine and interferon production , establishing an antiviral state in infected cells . Although it is not clear how MAVS activates these immune kinases , recent findings have established the vital roles of MAVS in host antiviral innate immunity [15] . Interestingly , the mitochondrial localization of MAVS is critical for its ability to activate downstream signaling events . As such , various RNA viruses , exemplified by human hepatitis C virus ( HCV ) , encode proteases that cleave MAVS from the outer membrane of the mitochondrion , thereby disarming MAVS-dependent signaling cascades and the host antiviral innate immunity [6] , [16] , [17] , [18] . Murine gamma-herpesvirus 68 ( γHV68 or MHV-68 ) is closely related to human Kaposi's sarcoma-associated herpesvirus ( KSHV ) and Epstein-Barr virus ( EBV ) [19] . KSHV and EBV are lymphotropic DNA viruses that are causally linked to malignancies of lymphoid or endothelial/epithelial origin , including lymphoma , nasopharyngeal carcinoma , and Kaposi's sarcoma [20] , [21] . Persisting within host immune cells , KSHV and EBV are known to evade , manipulate , and exploit host immune pathways [22] , [23] . Emerging studies suggest that γ-herpesviruses may usurp host innate immune responses for their infection [24] , [25] , [26] . However , it is not known how human KSHV and EBV manipulate innate immune pathways in vivo . Such investigations are greatly hampered by the lack of permissive cell lines and animal models for both KSHV and EBV . By contrast , γHV68 infection in laboratory mice leads to a robust acute infection in the lung and a long-term latent infection in the spleen . For murine γHV68 and human KSHV , the replication and transcription activator ( RTA , encoded by ORF50 ) is necessary and sufficient to initiate lytic replication from latently-infected cells , supporting the notion that RTA integrates diverse signaling pathways to initiate lytic replication [27] , [28] , [29] . Using γHV68 as a surrogate for human KSHV and EBV , we have unexpectedly discovered that γHV68 activated IKKβ to phosphorylate RTA and promote RTA transcriptional activation , thereby increasing viral gene transcription and lytic replication . As such , RTA phosphorylation by IKKβ couples γHV68 gene expression and lytic replication to host innate immune activation , representing the first example whereby a virus hijacks the antiviral MAVS-IKKβ pathway to promote its lytic replication .
To investigate the roles of MAVS in γHV68 infection , wild-type ( MAVS+/+ ) , heterozygous ( MAVS+/− ) , and knockout ( MAVS−/− ) mice were intranasally ( i . n . ) infected with 40 plaque-forming unit ( PFU ) γHV68 . γHV68 acute infection in the lung was measured by plaque assays at 4 , 7 , 10 , 13 , and 16 days post-infection ( d . p . i . ) . In MAVS+/+ mice , γHV68 titers peaked at 7 d . p . i . with approximately 500 PFU/lung and declined to 100 PFU/lung at 10 d . p . i . Viral load was undetectable by 13 d . p . i . , indicating that γHV68 acute infection in the lung had been cleared ( Figure 1A ) . Similar viral loads in the lungs of heterozygous mice ( MAVS+/− ) were observed ( data not shown ) . Surprisingly , although viral loads at 7 d . p . i . in the lungs of MAVS−/− mice were comparable to those in the lungs of MAVS+/+ mice , γHV68 was nearly undetectable at 10 d . p . i . ( Figure 1A ) . By contrast , γHV68 latent infection as characterized by viral genome frequency , persistent lytic replication , and reactivation was similar in splenocytes of MAVS+/+ and MAVS−/− mice at 16 and 45 d . p . i . ( Figure S1 ) . These observations suggest that MAVS plays a specific role ( s ) in γHV68 acute infection . To determine whether γHV68 infection altered MAVS expression , we infected BL/6 mice intranasally with a high dose ( 1×105 PFU ) of γHV68 , presumably permitting synchronized and maximal infection of lung epithelial cells . MAVS mRNA levels were determined by quantitative real-time PCR ( qRT-PCR ) . The levels of MAVS mRNA were transiently increased at 2 . 5 and 5 d . p . i . in the lung and spleen , respectively ( Figure S2A ) . Interestingly , the up-regulation of MAVS mRNA preceded that of viral RTA mRNA ( Figure S2A and S2B ) , and that higher viral RTA mRNA levels tightly correlated with higher MAVS mRNA levels at 2 . 5 and 5 d . p . i . , when MAVS mRNA levels peaked in the lung and spleen ( Figure S2C ) . Together with the reduced viral load in the lungs of MAVS−/− mice ( Figure 1A ) , these results suggest that MAVS is necessary for efficient lytic replication in mice and that the transiently induced MAVS expression by γHV68 infection may facilitate viral lytic replication in vivo . To investigate the roles of MAVS in γHV68 infection , we then assessed the effects of MAVS-deficiency on γHV68 lytic replication ex vivo . Mouse embryonic fibroblasts ( MEFs ) were infected with a GFP-marked recombinant γHV68 ( γHV68 K3/GFP ) and viral replication was examined by fluorescence microscopy and plaque assays . Surprisingly , γHV68 displayed delayed replication kinetics in MAVS−/− MEFs compared to MAVS+/+ MEFs at multiplicities of infection ( MOI ) of 0 . 01 and 0 . 1 ( Figure 1B , 1C and S3 ) . To quantitatively determine the effect of MAVS on γHV68 lytic infection , we examined γHV68 lytic replication in MAVS+/+ and MAVS−/− MEFs by plaque assays . In fact , γHV68 formed approximately four-fold more plaques in MAVS+/+ MEFs than those in MAVS−/− MEFs , indicative of reduced initiation of lytic replication in MAVS-deficient MEFs ( Figure 1D , S4A , and S4B ) . Interestingly , the plaque size of γHV68 was equivalent in MAVS+/+ and MAVS−/− MEFs ( Figure S4C and S4D ) . To test whether MAVS−/− MEFs are defective in supporting viral lytic replication in general , we examined the lytic replication of vesicular stomatitis virus ( VSV ) , a prototype RNA virus , with a plaque assay . Consistent with an antiviral activity of MAVS against RNA viruses , VSV formed 10-fold more plaques in MAVS−/− MEFs than those in MAVS+/+ MEFs ( Figure 1D ) . The diminished lytic replication of γHV68 in MAVS-deficient MEFs is consistent with the reduced acute infection observed in the lung . To test whether exogenously expressed MAVS is able to restore γHV68 lytic replication , we generated lentivirus in 293T cells and MEFs stably expressing human MAVS ( hMAVS ) was established with puromycin selection ( Figure 1E ) . As shown in Figure 1F and 1G , exogenous hMAVS restored γHV68 lytic replication by a plaque assay and multi-step growth curves . Nevertheless , these results together support the conclusion that MAVS is necessary for efficient γHV68 lytic replication in vivo and ex vivo . Two known pathways , the IKKα/β/γ-NFκB and TBK-1/IKKε-IRF3/7 pathway , have been characterized downstream of MAVS ( Figure 2A ) [3] , [4] . We therefore used MEFs deficient in key components of aforementioned pathways to identify downstream effectors of MAVS that are critical for γHV68 lytic infection . Plaque assays and multi-step growth curves of γHV68 lytic infection showed that deficiency in TRAF6 , IKKγ , and IKKβ , but not deficiency in the closely related IKKα , recapitulated phenotypes of MAVS deficiency ( Figure 2B and 2C ) . Notably , TRAF6 is necessary for MAVS to activate IKKβ that requires IKKγ , a scaffold protein for both IKKα and IKKβ [5] . By contrast , deficiency in type I IFN receptor ( IFNAR ) and double deficiency in IRF3 and IRF7 had no discernable effect on the plaque numbers of γHV68 in MEFs , indicating that the IRF-IFN signaling pathway is not critical for the initiation of γHV68 lytic replication ( Figure 2B ) . Furthermore , the exogenous IKKβ expression reconstituted by lentivirus restored the lytic replication of γHV68 as determined by a plaque assay and multi-step growth curves ( Figure 2D , 2E , and 2F ) . Interestingly , the expression of IKKβ in MAVS−/− did not increase γHV68 lytic replication by a plaque assay ( Figure 2E ) , suggesting that the MAVS-dependent activation of IKKβ , rather than the absolute expression level of IKKβ , is crucial for efficient γHV68 lytic replication . Additionally , exogenous IKKβ did not increase γHV68 plaque numbers in MAVS+/+ MEFs ( Figure 2E ) , implying that endogenous IKKβ is sufficient to support efficient γHV68 lytic replication . Of note , lentivirus infection reduces the difference of γHV68 plaque forming capacity in wild-type MEFs and in MEFs deficient in MAVS and IKKβ ( Figure 1F and 2D ) . Collectively , these data indicate that the MAVS-dependent IKKβ activation is critical for efficient γHV68 lytic replication . To assess whether the kinase activity of IKKβ is important for γHV68 lytic infection , we performed plaque assays with or without the specific IKKβ inhibitor , Bay11-7082 ( Bay11 ) . This experiment revealed that Bay11 reduced the plaque number of γHV68 in a dose-dependent manner ( Figure 3A ) . Whereas treatment with 1 µM of Bay11 at 0 . 5 h before infection reduced γHV68 plaque number by 52% , the same treatment at 7 h post-infection ( h . p . i . ) reduced the plaque number by 29% , emphasizing the important roles of IKKβ during early γHV68 infection ( Figure 3A ) . We further examined IKKβ activity by an in vitro kinase assay with IKKβ precipitated from MAVS+/+ and MAVS−/− MEFs infected with γHV68 . The IKKβ kinase activity was transiently and moderately increased in MAVS+/+ MEFs , however , it was drastically diminished in MAVS−/− MEFs after γHV68 infection ( Figure 3B ) . The activation of IKKβ was further supported by the rapid degradation of IκBα concurrent to IKKβ activation by γHV68 infection in MAVS+/+ MEFs , but not in MAVS−/− MEFs ( Figure 3C ) . To test whether UV-inactivated virus is able to trigger IKKβ activation , we examined the levels of IKKβ kinase activity and IκBα in MAVS+/+ MEFs by in vitro kinase and immunoblot assays , respectively . Interestingly , UV-inactivated γHV68 activated IKKβ and reduced IκBα protein levels , although less efficiently than live γHV68 ( Figure 3D and 3E ) . This observation suggests that γHV68 lytic replication is necessary to activate the MAVS-IKKβ pathway . Alternatively , UV treatment may damage or disrupt viral structural components whose integrity is necessary to activate the MAVS-IKKβ pathway . MAVS activation by RNA viruses is known to increase the expression of pro-inflammatory cytokines and interferons . However , γHV68 appears to be a poor inducer for these antiviral molecules , suggesting that γHV68 evades signaling events downstream of the MAVS adaptor . Indeed , γHV68 infection failed to up-regulate the expression of IFN-β ( Figure S5A ) . In agreement with this observation , γHV68 RTA , similar to KSHV RTA [30] , is sufficient to reduce IRF3 expression ( Figure S5B ) . Meanwhile , it was previously shown that γHV68 infection did not significantly activate NFκB during early infection [31] , suggesting that γHV68 uncouples NFκB activation from activated IKKβ . Taken together , these results support the conclusion that γHV68 infection selectively activates IKKβ to promote viral lytic replication . To discern the molecular mechanisms underlying the requirement of the MAVS-IKKβ pathway in γHV68 lytic infection , levels of γHV68 genomic DNA and mRNA were assessed by PCR or reverse transcription followed by real-time PCR analyses , respectively . At a low MOI ( 0 . 01 ) , analyses by PCR ( Figure 4A ) and real-time PCR ( Figure 4B ) revealed comparable levels of viral genomes in MAVS+/+ and MAVS−/− MEFs early after de novo infection , suggesting comparable viral entry into MAVS+/+ and MAVS−/− MEFs . Interestingly , levels of viral mRNA transcripts representing immediate early ( RTA , ORF73 , and ORF57 ) and early ( ORF60 and ORF9 ) gene products in MAVS+/+ MEFs were higher than those in MAVS−/− MEFs as determined by reverse-transcriptase PCR ( Figure 4C ) . Real-time PCR analyses with cDNA showed approximately 4- to 16-fold higher levels of γHV68 mRNA transcripts in MAVS+/+ MEFs compared to those in MAVS−/− MEFs at 2 and 3 d . p . i . ( Figure 4D ) . It has been shown that TRAF6 is necessary for MAVS to activate IKKβ [5] and exogenous TRAF6 is sufficient to activate IKKβ . To further examine the effects of the MAVS-IKKβ pathway on levels of γHV68 mRNA transcripts , a bacterial artificial chromosome ( BAC ) containing the γHV68 genome and a plasmid expressing TRAF6 were transfected into 293T cells . The effects of exogenous TRAF6 ( that activates IKKβ ) on viral transcription were determined by reverse transcription and real-time PCR . At 28 h post-transfection , a time point when immediate early and early genes are transcribed , exogenous TRAF6 efficiently increased the mRNA levels of γHV68 RTA , ORF57 , ORF60 , and ORF73 , without discernable effect on levels of viral genomic DNA ( Figure 4E and 4F ) . These results , obtained under conditions of loss of function ( MAVS−/− MEFs ) and gain of function ( TRAF6 expression ) , indicate that the activated IKKβ increases the levels of γHV68 mRNA transcripts . MAVS is an adaptor that activates IKKβ and the MAVS-dependent IKKβ increases γHV68 mRNA levels . We thus postulated that MAVS influences γHV68 transcription via its downstream IKKβ on RTA , because RTA , the master transcription activator , is critical for γHV68 lytic replication . To test this hypothesis , we examined whether IKKβ phosphorylates γHV68 RTA . IKKβ was purified from 293T cells and bacterial GST fusion proteins containing the RTA internal region ( RTA-M , aa 335–466 ) or the RTA C-terminal transactivation domain ( RTA-C , aa 457–583 ) were purified from E . coli ( Figure 5A ) . In the presence of [32P]γATP , IKKβ efficiently transferred the phosphate group to GST-RTA-C . By contrast , GST was not phosphorylated and GST-RTA-M was weakly phosphorylated by IKKβ . Furthermore , the kinase domain deletion variant of IKKβ ( IKKβΔKD ) failed to phosphorylate GST-RTA-C and GST-RTA-M ( Figure 5A ) , and IKKα had only residual kinase activity toward RTA-C ( Figure S6 ) . To confirm the MAVS- and IKKβ-dependent phosphorylation of RTA , RTA phosphorylation in γHV68-infected cells was analyzed by autoradiography and immunoblot . We found that MAVS- and IKKβ deficiency reduced RTA phosphorylation by 50% and 85% , respectively , while reconstituted IKKβ expression restored RTA phosphorylation to that of RTA in MAVS+/+ MEFs ( Figure 5B ) . To assess the roles of phosphorylation of RTA in transcription regulation , luciferase reporter assays were carried out with plasmids containing RTA-responsive promoters of RTA , ORF57 , and M3 . As shown in Figure 5C , the transcription activity of RTA on all three promoters was significantly increased by exogenous TRAF6 and IKKβ , but not by the kinase dead variant IKKβΔKD , supporting the notion that IKKβ promotes RTA transcription activation via phosphorylation . When expressed to similar levels of IKKβ , IKKβΔKD had no significant effect on RTA transcriptional activation ( Figure S7 ) . Given that RTA is a substrate for IKKβ , we sought to examine whether RTA can physically associate with the IKKα/β/γ complex . However , we were unable to detect interaction between RTA and any of the three subunits of IKKα/β/γ by co-immunoprecipitation ( data not shown ) , suggesting that the RTA interaction with the IKKα/β/γ complex is transient or mediated via additional cellular proteins . To identify IKKβ phosphorylation sites , series of truncations from the C-terminus of RTA were constructed and purified as GST fusion proteins for in vitro kinase assays with IKKβ . These experiments demonstrated that the IKKβ phosphorylation sites were located within the region containing residues 540 through 567 ( Figure S8 ) . Given that IKKβ is a serine/threonine kinase , clusters of various serine/threonine residues were changed to alanines and RTA phosphorylation was assessed similarly . Two clusters of mutations , replacement of S550T552S556 ( STS/A ) and T561T562S564 ( TTS/A ) by alanines , reduced the phosphorylation levels of RTA-C by approximately 72% and 45% , respectively ( Figure 5D and S8 ) . These results indicate that the STS and TTS sequences represent two major IKKβ phosphorylation sites within the transactivation domain of RTA . To further examine the roles of IKKβ phosphorylation in regulating RTA transcription activity , reporter assays with plasmids containing wild-type RTA , the STS/A and TTS/A variants were carried out with exogenously expressed IKKβ . The STS/A and TTS/A variants had lower basal activity to activate promoters of RTA , ORF57 , and M3 . Moreover , exogenous IKKβ failed to further stimulate the transcription activities of the STS/A and TTS/A variants to activate promoters of RTA and ORF57 ( Figure 5E ) . Interestingly , the STS/A variant activated M3 promoter to the level of wild-type RTA with or without IKKβ , indicating that the STS site is dispensable for IKKβ to promote RTA transcriptional activity on the M3 promoter ( Figure 5E ) . It is noteworthy that the STS/A and TTS/A variants were expressed at higher levels than wild-type RTA , the transcription activities of the STS/A and TTS/A variants were approximately 50% and 20% of that of wild-type RTA , respectively , when luciferase activity was normalized against protein levels ( Figure 5F ) . Collectively , these results demonstrated that IKKβ promotes RTA transcriptional activation via phosphorylation of the TTS and STS sites within the transactivation domain . To further investigate the roles of RTA phosphorylation , we assessed the effects of the STS/A and TTS/A mutations on γHV68 lytic replication . Taking advantage of the γHV68-containing BAC with a transposon insertion that inactivates RTA ( ORF50 Null ) [32] , a recombination-based strategy [33] was employed to generate viruses carrying wild-type RTA ( Null Rescued , designated NR ) , the STS/A allele , or the TTS/A allele ( Figure 6A ) . Whereas we easily obtained recombinant γHV68 containing wild-type RTA ( γHV68 . NR ) or the TTS/A allele ( γHV68 . TTS/A ) , the STS/A variant failed to support γHV68 recombination in multiple independent experiments . This observation suggests an essential role for the phosphorylated STS sequence in γHV68 lytic replication . To confirm the integrity of viral genomic DNA , we performed restriction digestion with KpnI and EcoRI , and analyzed with agarose gel electrophoresis . As expected , the removal of the Kanamycin cassette within RTA alleles reduced the 9-kb fragment to 7 . 5-kb counterpart released by KpnI digestion ( Figure 6B ) , and abolished an EcoRI site within the Kanamycin cassette ( Figure 6C ) . To assess the transcriptional activity of RTA derived from BAC DNA , BAC DNA and the M3p luciferase reporter plasmid were transfected into 293T cells and RTA transcriptional activity was assessed by luciferase reporter assay . The activity of wild-type RTA to activate M3 promoter was approximately 6-fold higher than that of the TTS/A mutant ( Figure 6D ) . Using 293T cells transfected with the γHV68 BAC containing the TTS/A allele and a plasmid expressing TRAF6 , we assessed the effects of TRAF6 ( that activates IKKβ ) on γHV68 gene expression . In contrast to what was observed for the γHV68 BAC containing wild-type RTA ( Figure 4F ) , exogenous TRAF6 had marginal effects on the levels of viral mRNAs transcribed from γHV68 BAC containing the TTS/A allele ( Figure 6E ) . These findings are consistent with the observation that IKKβ failed to further promote the transcription of the TTS/A variant ( Figure 5E ) , supporting the conclusion that the TTS residues constitute an IKKβ phosphorylation sequence by which RTA-dependent transcription is positively regulated . Next , we examined whether recombinant γHV68 . TTS/A recapitulates the defects of wild-type γHV68 lytic replication in MEFs deficient in MAVS and IKKβ ( plaque assays and multi-step growth curves ) . To assess the effects of the TTS/A mutation on γHV68 transcription activation , we normalized viral genomes immediately after γHV68 de novo infection of MEFs by qRT-PCR . With equal number of viral genomes , γHV68 . NR displayed approximately 32-fold higher of RTA mRNA than recombinant γHV68 . TTS/A in MAVS+/+ MEFs at 30 h . p . i . ( Figure 7A ) . This is consistent with the observation that RTA activates its own promoter to facilitate viral lytic replication ( Figure 5C and 5E ) . Furthermore , multi-step growth curves ( at an MOI of 0 . 01 ) demonstrated that γHV68 . TTS/A had delayed replication kinetics and produced >3 orders of magnitude less virion progeny in MAVS+/+ MEFs ( Figure 7B ) . To test whether RTA phosphorylation and the MAVS-IKKβ pathway are functionally redundant , we examined the replication kinetics of recombinant γHV68 . NR and γHV68 . TTS/A in wild-type , MAVS−/− , and IKKβ−/− MEFs . Consistent with our previous observations ( Figure 1C , 2C , and S3B ) , γHV68 . NR showed delayed lytic replication in MAVS−/− and IKKβ−/− MEFs ( Figure 7B and 7C ) . Remarkably , γHV68 . TTS/A replicated with similar kinetics in wild-type , MAVS−/− , and IKKβ−/− MEFs , suggesting that the MAVS-IKKβ pathway functions on RTA to promote viral lytic replication ( Figure 7B and 7C ) . However , these replication defects of recombinant γHV68 carrying the TTS/A mutation are much more pronounced than the phenotypes of wild-type γHV68 in MAVS−/− and IKKβ−/− MEFs , implying that additional kinases may influence RTA transcriptional activation via phosphorylation of the TTS site . Taken together , we conclude that the TTS site of RTA is likely phosphorylated by IKKβ and is crucially important for γHV68 lytic replication .
Here we provide evidence that murine γHV68 hijacks the antiviral MAVS-IKKβ pathway to promote its lytic replication . The MAVS adaptor is important for host defense against invading pathogens , including various DNA and RNA viruses . For example , mice lacking MAVS were severely compromised in innate immune defense against VSV infection , leading to an elevated peak viral load and prolonged acute viral infection [34] . The antiviral effects of MAVS have been observed against the infection of a number of RNA and DNA pathogens [35] , [36] , [37] . To our surprise , γHV68 viral load in the lungs of MAVS−/− mice was significantly lower than that in the lungs of MAVS+/+ mice at 10 d . p . i . The reduced viral load of γHV68 in MAVS−/− mice is counter-intuitive to the presumed antiviral function of the MAVS adaptor in promoting innate immune responses . Although type I interferons in γHV68-infected mice were undetectable [38] , mice deficient in type I IFN receptor had higher viral loads and succumbed to γHV68 infection [39] . We surmise that the effects of MAVS deficiency on γHV68 acute infection is likely under-estimated , providing that MAVS is critical for interferon production in response to viral infection . Thus , the viral load of γHV68 acute infection in MAVS−/− mice likely represents a “neutralized” phenotype , in which reduced γHV68 lytic replication is compensated by the lack of type I interferon inhibition . Moreover , the observation that viral RTA mRNA levels correlates tightly with the MAVS mRNA levels during early γHV68 acute infection suggests that MAVS is necessary for γHV68 lytic replication ( Figure S2 ) . Although we have not formally excluded the contribution of host immune responses against γHV68 infection to the reduced viral load at 10 d . p . i . in MAVS−/− mice , our experiments with γHV68 replication ex vivo demonstrated critical roles of the MAVS-IKKβ pathway in facilitating γHV68 lytic infection . During early stages of viral infection , γHV68 activated IKKβ in a MAVS-dependent manner , a signaling event that is likely triggered by a variety of pathogens . The MAVS-dependent activation was supported by elevated IKKβ kinase activity and accelerated IκBα degradation , signature signaling events downstream of the MAVS adaptor . Although the up-regulation of IKKβ kianse activity appears modest , γHV68 may direct IKKβ kinase activity to efficiently modify cellular and viral components that are critical for γHV68 infection , such as RTA . Consequently , γHV68 can harness activated IKKβ without inducing NFκB activation that may be resulted from massive IKKβ activation . Indeed , it was reported that γHV68 infection does not induce NFκB activation during early infection [40] , suggesting that modest IKKβ activation is beneficial for γHV68 infection and that γHV68 may uncouple NFκB activation from IKKβ activation . Interestingly , γHV68 appears to block the interferon limb of the MAVS-dependent innate immune pathway . In fact , we found that γHV68 infection failed to induce the expression of IFN-β ( Figure S5A ) . Consistent with this observation , γHV68 RTA , similar to KSHV RTA [30] , is sufficient to reduce IRF3 protein ( Figure S5B ) , potentially abrogating the production of interferons that otherwise would potently thwart γHV68 replication . Moreover , ORF36 was reported to deregulate the phosphorylated form of IRF3 and inhibit interferon production [41] . These observations suggest that γHV68 selectively activates the MAVS-IKKβ pathway to promote viral lytic replication . Within this report , we have identified one requisite role of the MAVS-IKKβ pathway in γHV68 lytic replication with MEFs deficient in key components of this pathway . Phenotypically , γHV68 displayed similar replication defects in MEFs deficient in MAVS , IKKβ , and IKKγ , although the replication defects in IKKβ−/− and IKKγ−/− MEFs were more pronounced than those in MAVS−/− MEFs ( Figure 1C , 2B , and 2C ) . This result supports the corollary that IKKβ , with the scaffold protein IKKγ , functions downstream of MAVS and likely integrates additional signaling emanating from other innate immune pathways including Toll-like receptors . It is worthy to point out that our result does not exclude the antiviral activity of the IRF-IFN pathway in γHV68 lytic replication , although deficiency of IRF3 and IRF7 or IFNAR did not appear to impact the initiation of γHV68 lytic infection as assessed by plaque assays ( Figure 2B ) . It is possible that the IRF-IFN pathway may inhibit molecular events other than the initiation of lytic replication and reduce viral yield during γHV68 infection . Mechanistically , we identified γHV68 RTA , the master viral replication transactivator , as one of the IKKβ kinase substrates . Phosphorylation of RTA by IKKβ increases RTA transcriptional activity and consequently viral mRNA production . Indeed , γHV68 had lower levels of various mRNA transcripts that correlated with reduced lytic replication in MAVS−/− MEFs ( Figure 1 and 4 ) . Conversely , exogenous TRAF6 potentiated RTA transcriptional activity and substantially increased the levels of viral mRNA transcripts ( Figure 4F and 5C ) . Additionally , exogenously reconstituted expression of MAVS and IKKβ restored RTA phosphorylation ( Figure 5B ) and restored γHV68 lytic replication ( Figure 1 and 2 ) . Moreover , lytic replication of recombinant γHV68 viruses carrying mutations within the IKKβ phosphorylation sites was greatly impaired , displaying phenotypes that are more pronounced than those of wild-type γHV68 in MEFs deficient in components of the MAVS-IKKβ pathway . Conceivably , other kinases and signaling pathways may converge to modulate RTA transcriptional activation via phosphorylation within these identified IKKβ sites . For example , virus-encoded kinases , such as the functionally conserved ORF36 , may amplify the phosphorylation cascade that is initiated by the MAVS-IKKβ pathway [42] . Most importantly , RTA auto-activates its own promoter and increases RTA protein that , in turn , up-regulates the expression of numerous immediate early and early genes during γHV68 infection . Thus , the 50–80% reduction in RTA transcriptional activity of the STS/A and TTS/A variants ( Figure 5F ) likely translates into , through the aforementioned amplification cascades , the viral yields that are less than 0 . 1% of the recombinant γHV68 . NR ( Figure 7B ) . Finally , it is noteworthy that deficiency in MAVS and IKKβ and mutations within RTA exhibited distinct phenotypes ( such as peak viral titers of multi-step growth curves ) , in addition to the shared reduction of γHV68 lytic replication . These differing effects on γHV68 infection are likely due to their unique hierarchical position within the MAVS-IKKβ-RTA signaling axis . In essence , these experiments identified novel phosphorylation sites within RTA that couples γHV68 lytic replication to the antiviral IKKβ kinase . These findings collectively demonstrate that the MAVS-dependent IKKβ kinase activity is critical for RTA transcriptional activation and γHV68 lytic replication . Interestingly , Gwack et al . reported that phosphorylation of the internal serine/threonine-rich region of KSHV and γHV68 RTA inhibited RTA transcriptional activity and suppressed viral lytic replication [43] . Together with our findings , these results indicate that site-specific phosphorylation determines the transcriptional activity , and likely the promoter-specificity , of gamma-herpesvirus RTA . Although it is well accepted that the NFκB pathway is crucial for gamma-herpesvirus latent infection [44] , the roles of this pathway in gamma-herpesvirus lytic replication appear to be inconsistent . Particularly , Krug et al . reported that the recombinant γHV68 expressing the IκBα super suppressor replicated indistinguishably compared to wild type γHV68 [31] . Thus , the authors concluded that the NFκB pathway is dispensable for γHV68 lytic replication . By contrast , it was shown that RelA , the p65 subunit of an NFκB transcription dimer , inhibits γHV68 lytic replication through suppressing RTA transcription activity in 293T cells [45] . Finally , our current report indicates that the MAVS-IKKβ pathway is necessary for efficient γHV68 lytic replication . However , the seemingly paradox can be explained by the differential effects of three distinct components of the NFκB pathway on γHV68 lytic replication . Although the IκBα super suppressor is commonly employed to inhibit the activation of the NFκB transcription factors , it is important to note that no significant NFκB activation was observed during early γHV68 infection ( within the first 6 hours post-infection ) [40] , temporal phase in which the critical roles of IKKβ was indentified by our genetic and biochemical experiments . Conceivably , the unphosphorylatable IκBα super suppressor may not impact IKKβ kinase activity . By contrast , we have focused on the IKKβ kinase and our study indicated that the ability of IKKβ to promote viral lytic replication largely stems from IKKβ kinase activity to phosphorylate RTA and increase RTA transcriptional activation . Apparently , neither IκBα , nor RelA can do so in replace of IKKβ function . On the other hand , although RelA was shown to suppress γHV68 lytic replication [45] , the lack of NFκB activation during early γHV68 infection implies that γHV68 uncouples NFκB activation from IKKβ activation , which are otherwise tightly correlated . As such , γHV68 infection may selectively activate the IKKβ kinase , while sparing the inhibition by preventing NFκB activation . Therefore , a scenario that potentially accommodates all three reports is that nuclear activated RelA is necessary to inhibit γHV68 lytic replication and γHV68 is capable of preventing RelA activation in an IκBα-independent manner . Crucial to this hypothesis is the mechanisms that γHV68 has evolved to thwart NFκB activation and future experiments are necessary to address this possibility . It was previously reported that γHV68 was impaired for latency establishment and reactivation in MyD88-deficient mice , although the lytic replication of γHV68 appeared to be normal in these mice [26] . Moreover , agonists specific for TLR7/8 , which activate downstream signaling events through MyD88 , induced KSHV lytic gene expression and reactivated KSHV replication from latently-infected B cells [25] . The specific roles of MAVS in lytic replication and MyD88 in latent infection are consistent with their distinct functions in innate immune responses of epithelial cells and immune cells , respectively . Given that MyD88 also activates the IKKα/β kinase complex , it is possible that IKKβ-dependent activation of RTA may contribute to γHV68 and KSHV latent infection as well . Finally , reduced lytic replication of human KSHV and cytomegalovirus has been observed under experimental conditions in which IKKβ was inhibited by Bay11 , implying that human KSHV and cytomegalovirus have evolved similar molecular mechanisms to facilitate lytic replication [46] , [47] , [48] . Taken together , the mechanism whereby an antiviral innate immune signaling pathway is exploited to promote viral lytic replication may be applied to other herpesviruses and viral reactivation from latency . This study thus has uncovered an intricate interplay between the viral replication transactivator , RTA , and the MAVS-IKKβ pathway . To our best knowledge , this is the first example that illustrates how a virus hijacks an antiviral signaling pathway , downstream of cytosolic sensors , to initiate its lytic replication . Perhaps , co-evolution between the persistent herpesviruses and their hosts has selected viruses that exploited the inevitable innate immune activation by viral infection . Although our current study delineates the key signaling events downstream of MAVS and IKKβ , it remains unknown what viral components and cellular factors activate the MAVS-IKKβ pathway and whether these mechanisms are shared by the oncogenic KSHV and EBV to promote lytic replication or reactivation .
For protein expression in mammalian cells , all genes were cloned into pcDNA5/FRT/TO ( Invitrogen ) unless specified . For protein expression and purification in E . coli , the internal region ( RTA-M , aa 335–466 ) and C-terminal transactivation domain ( RTA-C , aa 457–583 ) of RTA were cloned into pGEX-4T-1 ( Promega ) with BamHI and XhoI sites . NIH3T3 cells , HEK293T ( 293T ) cells and mouse embryonic fibroblasts ( MEFs ) were maintained in DMEM ( Mediatech ) with 8% newborn calf serum ( NCS ) or fetal bovine serum ( FBS ) , respectively . MAVS+/+ , MAVS−/− , IKKβ−/− , IKKγ−/− and TRAF6−/− MEFs were described previously [4] , [34] . IKKα−/− MEFs were kindly provided by Dr . Amyn A . Habib ( Neurology , UT Southwestern ) . IFNAR+/+ and IFNAR−/− MEFs were kindly provided by Dr . Michael Gale ( Immunology , University of Washington ) . IRF3+/+IRF7+/+ and IRF3−/−IRF7−/− MEFs were kindly provided by Dr . Jae Jung ( Microbiology , University of Southern California ) . γHV68 K3/GFP was kindly provided by Dr . Philip Stevenson ( Cambridge University , UK ) . Wild-type γHV68 and γHV68 K3/GFP were amplified in NIH3T3 cells , and VSV-GFP virus was amplified in BHK-21 cells . Viral titer was determined by a plaque assay with NIH3T3 cells . All animal experiments were performed in accordance to NIH guidelines , the Animal Welfare Act , and US federal law . The experimental protocol ( entitled: Innate immune pathways in γHV68 infection ) were approved by the Institutional Animal Care and Use Committee ( IACUC ) . All animals were housed in a centralized research animal facility that is accredited by the Association of Assessment and Accreditation of Laboratory Animal Care International , and that is fully staffed with trained husbandry , technical , and veterinary personnel . Wild-type ( MAVS+/+ ) , heterozygous ( MAVS+/− ) , and knockout ( MAVS−/− ) mice were described previously [34] . Gender-matched , 6- to 8-week old littermate mice were intranasally ( i . n . ) inoculated with 40 plaque-forming unit ( PFU ) wild-type γHV68 . To assess MAVS expression in the lung and spleen , BL/6 mice were intranasally infected with 1×105 PFU γHV68 . The lungs and spleens were harvested and homogenized in DMEM . Viral titer of mice tissues or cell lysates was assessed by a plaque assay on NIH3T3 monolayers . After three rounds of freezing and thawing , 10-fold serially-diluted virus supernatants were added onto NIH3T3 cells and incubated for 2 hours at 37°C . Then , DMEM containing 2% NCS and 0 . 75% methylcellulose ( Sigma ) was added after removing the supernatant . Plaques were counted at day 6 post-infection . The detection limit for this assay is 5 PFU . To assess the infectivity of γHV68 on various MEFs , a similar plaque assay was carried out with the initial cell density of 5000 cells/cm2 . In Bay11-7082 treatment assay , 0 . 5µM or 1µM Bay11-7082 was added at 0 . 5 h before infection or 7 h post-infection . Supernatant was removed after 30 min incubation at 37°C , and cells were washed with medium and incubated for plaque formation . Glutathione-S-transferase ( GST ) and GST fusion proteins containing the internal region and the transactivation domain of RTA were expressed with IPTG induction and purified with glutathione-sepharose as previously described [33] . Eluted proteins were re-suspended in 25% glycerol and stored at −20°C for kinase assays . To purify IKKβ and IKKβΔKD , 293T cells were transfected with pcDNA3 containing Flag-IKKβ and Flag-IKKβΔKD . At 48 h post-transfection , cells were lysed with kinase purification buffer ( 150 mM NaCl , 20 mM Tris . HCl pH7 . 4 , 10% Glycerol , 0 . 5% Triton X-100 , 0 . 5 mM DTT ) and subject to one-step affinity purification with anti-Flag M2-conjugated agarose ( Sigma ) . Proteins were eluted with 0 . 2 mg/ml Flag peptide in kinase buffer ( 50 mM KCl , 2 mM MgCl2 , 2 mM MnCl2 , 1 mM DTT , 10 mM NaF , 25 mM HEPES , pH7 . 5 ) and stored in 25% glycerol at −80°C . Commercial antibodies used in this study include: anti-Flag ( Sigma ) , anti-GFP ( Covance ) , anti-IKKβ ( H4 ) , anti-IκBα ( C20 ) ( Santa Cruz Biotech . ) , anti-actin ( Abcam . ) . To generate antibody to γHV68 RTA , the mixture of GST fusion proteins containing the RTA-M and RTA-C was used to immunize a rabbit and polyclonal antibodies were tested for the specificity with pre-immune serum as control . Endogenous IKKβ or exogenously expressed IKKβ and IKKβΔKD were used for in vitro kinase assays . The kinase reaction includes 0 . 5 µg GST or GST fusion proteins , 100 µCi [32P]γATP , and approximately 250 ng kinase in 20 µl of kinase buffer . Reaction was incubated at room temperature for 25 min and denatured proteins were analyzed by SDS-PAGE and autoradiography . To determine the relative levels of viral transcripts , total RNA was extracted from MEFs or mice tissues using TRIzol reagent ( Invitrogen ) . To remove genomic DNA , total RNA was treated with RNase-free DNase I ( New England Biolab ) at 37°C for 1 hour . After heat inactivation , total RNA was re-purified with TRIzol reagent . cDNA was prepared with 1 . 5 µg total RNA and reverse transcriptase ( Invitrogen ) . RNA was then removed by incubation with RNase H ( Epicentre ) . Abundance of viral transcripts was assessed by qRT-PCR . Mouse β-actin was used as an internal control . Primers used in this study were summarized in Table S1 . Bulk splenocytes were re-suspended in DMEM , and plated onto primary MEF monolayers in 96-well plates in 2-fold serial dilutions ( from 105 to 48 cells/well ) as previously described [49] . Twelve wells were plated every dilution . Reactivation percentage was scored for cytopathic effects ( CPE ) positive wells on day 6 . In order to measure preformed infectious virus , disrupted cells were plated onto primary MEF monolayers . This procedure destroys over 99% of the cells , but has minimal effect on preformed infectious virus , thus allowing distinction between reactivation from latency and persistent infection . The frequency of splenocytes harboring wild-type γHV68 genome was assessed by a single-copy-sensitive nested PCR analysis of serial dilutions of splenocytes as previously described [49] . Briefly , mice spleens were homogenized and re-suspended in isotonic buffer and subjected to 3-fold serial dilutions ( from 104 to 41 cells/well ) in a background of uninfected RAW 264 . 7 cells , with a total of 104 cells per well . Twelve replicates were plated for each cell dilution . After being plated , cells were subjected to lysis by proteinase K at 56°C for 8 hours . After inactivating the enzyme for 30 minutes at 85°C , samples were subjected to nested PCR using primers specific for γHV68 ORF72 . Positive controls of 10 , 1 , and 0 . 1 copies of viral DNA and negative controls of uninfected RAW 264 . 7 cells alone were included on each plate . Reaction products were separated using 2 . 5% UltraPure agarose ( Invitrogen ) gels and visualized by ethidium bromide staining . Reactivation and LDPCR results were analyzed using GraphPad Prism software ( GraphPad Software , San Diego , CA ) . The frequencies of genome-positive cells were statistically analyzed using the paired Student's t-test . The frequencies of viral genome-positive cells were determined from a nonlinear regression analysis of sigmoidal dose-response best-fit curve data . Based on a Poisson distribution , the frequency at which at least one event is present in a given population occurs at the point at which the regression analysis line intersects 63 . 2% . Pooled data of at least three independent experiments were used to calculate P values with the two-tailed , unpaired Student's t-test . 293T cells ( 2×105 cells/well ) were seeded in 24-well plates 16 hours before transfection . A total of 377 ng of plasmid DNA per well was co-transfected by the calcium phosphate method ( Clontech ) . The plasmid cocktail includes 75 ng of luciferase plasmid ( RTAp_luc , ORF57p_Luc or M3p_luc ) , 200 ng of pCMV-β-galactosidase plasmid , 2 ng of pcDNA5_RTA and 100 ng of pcDNA5 containing TRAF6 , IKKβ or IKKβΔKD . At 21 hours post-transfection , whole cell lysates were used to measure the firefly luciferase activity and β-galactosidase activity . The bacterial artificial chromosome ( BAC ) system was used to generate recombinant γHV68 similarly to what was described previously [33] . Briefly , wild-type RTA or the STS/A and TTS/A alleles were PCR amplified with overlapping PCR primers . Purified PCR products , along with the BAC clone 5 . 15 [32] containing a transposon within the transactivation domain of RTA ( between nucleotide 69269 and 69270 , according to accession number U97553 ) , were transfected into NIH3T3 cells with Lipofectamine 2000 ( Invitrogen ) . Virus in the supernatant was further amplified with NIH3T3 cells . To isolate circular BAC DNA , NIH3T3 cells were infected with recombinant γHV68 and DNA was extracted according to Hirt's protocol [50] and electroporated into ElectroMAX DH10B cells ( Invitrogen ) . BAC DNA containing γHV68 genome was digested with EcoRI and KpnI to rule out chromosome rearrangement . Meanwhile , the RTA alleles were amplified by PCR and sequenced to confirm desired mutations . Selected clones were transfected into NIH3T3 cells and recombinant γHV68 was amplified for subsequent experiments . RIG-I , 230073; MDA-5 , 71586; MAVS , 228607; TBK-1 , 56480; IKKε , 56489; IRF3 , 54131; IRF7 , 54123; c-Jun , 16476; ATF-2 , 11909; IFNβ , 15977; IFNAR , 15975; TRAF3 , 22031; TRAF6 , 22034; IKKγ , 16151; IKKα , 12675; IKKβ , 16150; IκBα , 18035; NFκB1 , 18033; RelA , 19697; MyD88 , 35956; TLR7 , 170743; TLR8 , 170744 . | Innate immunity represents the first line of defense against pathogen infection . Recent studies uncovered an array of sensors that detect pathogen-associated molecular patterns and induce antiviral cytokine production via two closely related kinase complexes , i . e . , the IKKα/β/γ and TBK-1/IKKε . To counteract host immune defense , herpesviruses have evolved diverse strategies to evade , manipulate , and exploit host immune responses . Here we report that infection by murine gamma-herpesvirus 68 ( γHV68 ) , a model gamma-herpesvirus for human Kaposi's sarcoma-associated herpesvirus and Epstein-Barr virus , activated the IKKβ kinase and IKKβ was usurped to promote viral transcriptional activation . As such , uncoupling IKKβ from transcriptional activation by biochemical and genetic approaches impaired γHV68 lytic replication . Our study represents an example whereby viral lytic replication is coupled to host innate immune activation and sheds light on herpesvirus exploitation of immune responses . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"microbiology/innate",
"immunity"
] | 2010 | Murine Gamma-Herpesvirus 68 Hijacks MAVS and IKKβ to Initiate Lytic Replication |
Tree cover is a fundamental structural characteristic and driver of ecosystem processes in terrestrial ecosystems , and trees are a major global carbon ( C ) sink . Fire and herbivores have been hypothesized to play dominant roles in regulating trees in African savannas , but the evidence for this is conflicting . Moving up a trophic scale , the factors that regulate fire occurrence and herbivores , such as disease and predation , are poorly understood for any given ecosystem . We used a Bayesian state-space model to show that the wildebeest population irruption that followed disease ( rinderpest ) eradication in the Serengeti ecosystem of East Africa led to a widespread reduction in the extent of fire and an ongoing recovery of the tree population . This supports the hypothesis that disease has played a key role in the regulation of this ecosystem . We then link our state-space model with theoretical and empirical results quantifying the effects of grazing and fire on soil carbon to predict that this cascade may have led to important shifts in the size of pools of C stored in soil and biomass . Our results suggest that the dynamics of herbivores and fire are tightly coupled at landscape scales , that fire exerts clear top-down effects on tree density , and that disease outbreaks in dominant herbivores can lead to complex trophic cascades in savanna ecosystems . We propose that the long-term status of the Serengeti and other intensely grazed savannas as sources or sinks for C may be fundamentally linked to the control of disease outbreaks and poaching .
In addition to being a prominent structural feature of savanna and forest ecosystems , tree cover has far-reaching consequences for ecosystem function [1] , [2] . Trees are a key component of stored carbon ( C ) , and thus important in the potential for ecosystems to act as carbon dioxide ( CO2 ) sinks in the effort to curb global warming . Despite this , understanding the factors that influence tree cover , herbaceous production , and soil organic matter in savannas and other nonforest biomes remains a vexing and challenging problem in ecology [3] , [4] . It has been hypothesized that top-down limitation by fire and herbivores plays a dominant role in regulating tree cover within bounds determined by rainfall [5] . Although rainfall does indeed appear to impose an upper limit on tree cover in savanna ecosystems [5]–[7] , evidence to support the role of fire and herbivores as factors driving tree cover below this maximum is conflicting [4]–[6] , [8] . There has accordingly long been disagreement among ecologists about the relative importance of climate , fire , and herbivores ( especially elephants ) as determinants of tree-to-grass ratios and tree cover in African savannas [3] , [9] , [10] . Studies at the next trophic level do little to clarify the situation as the factors that regulate herbivores ( such as disease and predation ) and fire occurrence are poorly understood for any given ecosystem . We drew on a 44-y time series ( 1960–2003 ) to identify the direct and indirect links among disease , herbivores , fire , rainfall , and changes in tree density ( which we use here as a measure of tree cover ) in the 25 , 000 km2 Serengeti-Mara ecosystem of East Africa ( Figure 1 ) . Elephants ( the dominant browsers ) , fire , and wildebeest ( the dominant grazers ) have all been proposed as important drivers contributing to changes in tree cover [11]–[14] . It has been suggested that rinderpest eradication set in motion a far-reaching and ongoing regulatory trophic cascade throughout the ecosystem , with the resulting irruption of wildebeest leading to a reduction of grass biomass and fire frequency , and an increase in tree cover [15]–[17] . Here we use a rigorous statistical approach to examine the evidence for this cascade , as well as competing explanations for historic patterns of fire prevalence and fluctuations in tree density . We further examine how changes at various nodes in this cascade ( herbivores , fire , and trees ) may have shifted the carbon ( C ) balance of the Serengeti ecosystem over the past half-century . We compared ten competing models for the determinants of fire and tree density change in this ecosystem ( Table 1 ) . These models jointly investigated the effects of grazer abundance and rainfall on fire , and the influence of fire , elephants , grazers , rainfall , and atmospheric CO2 concentration on per capita changes in tree density inferred from photopanoramas .
The model with the strongest support , based on the deviance information criterion ( DIC ) ( Table 1 ) , identified wildebeest ( Figure 2A , presumably via their grazing impact on grass biomass ) and intra-annual variation in rainfall ( the ratio of wet∶dry rainfall ) as the best predictors of fire occurrence ( defined as the proportion of the ecosystem that burns per year ) . The differences in model DIC values ( Table 1 ) suggested that wildebeest grazing is a better predictor of fire than is intra-annual rainfall variation , but both of these variables contributed to the observed global patterns of fire occurrence in the Serengeti ( Figure 2C ) as inferred from the credible intervals of their coefficients ( β1 and β2 , respectively; Equation 4 and Table 2 ) . The inclusion of mean annual rainfall did not improve model fit ( Table 1 ) . The results also suggested that that fire alone—and not elephants ( Figure 2B ) , mean annual rainfall , or atmospheric CO2—has been the primary driver of observed changes in tree density ( Figure 2A–2E ) . Per capita tree density changes were negative from 1960 until the mid 1970s , becoming positive thereafter ( decelerating after 1990 ) ; our model closely tracked these trends ( Fig . 2D , 2E , and 2G ) . Furthermore , about a third of the variance in tree density change that was unexplained by the best-fitting model could be explained by variation in density ( Figure 2H ) : photopanorama sequences with low initial tree density had faster per capita growth than expected , suggesting that density dependence ( which we could not model explicitly , as we only had data on relative density changes within photopanorama sites ) has also played an important role in regulating tree dynamics . The DIC results were clear in teasing apart the drivers of fire occurrence over time , but less clear in terms of inferring the factors regulating tree density . On the one hand , model 3 performed better than ( or as well as ) more complex models , but on the other , models 2 ( fire effects only ) and 7 ( elephant effects only ) had similar DIC values ( Table 1 ) . The role of fire was supported , however , by an examination of coefficient credible intervals . The fire coefficient ( γ1 ) differed from zero both when it appeared alone or with elephants as a covariate ( Table 1; values for model 3 given in Table 2 ) , but the credible intervals for the elephant , mean annual rainfall , and atmospheric CO2 coefficients included zero in all models . To further test the explanatory power of fire versus other factors in driving tree density changes , we ran model 3 again , but fitted only to tree data for the period 1981–2003 ( see Methods ) , and then validated by comparing its predictions with the reserved 1960–1980 data . We also ran two competing single-factor models ( elephants and mean annual rainfall ) with the same dataset . In all cases we included wildebeest and intra-annual rainfall variation as explanatory variables for fire . The fire model performed equally well with the reduced and full datasets ( Figure 3 ) , closely tracking the trajectory of the original model and predicting the decline in tree density that occurred in the 1960s and 70s ( Figure 3 ) . The other two models , however , while fitting the 1981–2003 data quite well , performed poorly for the validation period ( Figure 3 ) . We extended our analysis to include the role that the eradication of rinderpest ( a Morbillivirus closely related to measles and distemper [16] ) played in causing a shift from top-down disease control to bottom-up resource limitation in wildebeest . The prevalence of rinderpest , which causes high levels of mortality in wildebeest calves , declined rapidly following vaccination of the cattle that were a reservoir for the pathogen ( Figure 4A ) [16] . Eradication of the pathogen permitted the wildebeest population to erupt , ultimately driving the trophic cascade ( driven by grazing-mediated fire suppression ) that resulted in a marked increase in tree density . The rinderpest-triggered trophic cascade may have had far-reaching functional consequences for the role of savanna ecosystems as carbon ( C ) sources or sinks . The soil C ( SOC ) and plant biomass C pools contain most of the C in terrestrial ecosystems , and a decline in the size of these pools would make the ecosystem a net source of C . Grazing intensity ( GI ) and fire have been shown theoretically [18] and empirically ( unpublished data ) [19] to enhance and reduce the size of the soil organic matter pool in the Serengeti , respectively . We redefined tree density in units of C per km−2 , and used functions relating fire and GI to changes in SOC to simulate changes in the size of these two C pools with a modified version of our best-fit Bayesian state-space model ( BSS ) model ( model 3 ) . The model predicted changes in ecosystem-level C stocks in the Serengeti between 1960 and 2003 based on annual estimates of GI , fire extent , and changes in tree density over this period ( Figure 5 ) .
Our results suggest that long time series , examined over appropriate spatial scales , can identify strong signals in the relationships among herbivores , fire , climate , and vegetation . Our model explained about three-quarters of the variance in both fire and per capita tree density change ( Figure 2F–2H ) . This is particularly striking in the case of fire , which depends not only on fuel loads , but also on the occurrence of ignition events . Here we show that grazer population size ( and by implication grazer-determined fuel loads ) is a key determinant of fire frequency , a finding documented in at least one other savanna system [20] , and thus grazer abundance is an important indirect driver of tree population dynamics , supporting findings from previous modelling and empirical studies [21]–[23] . Although much of the relationship between wildebeest population size and fire extent is arguably driven by the widespread changes that occurred up to 1975 in the immediate aftermath of rinderpest eradication , to the best of our knowledge no other plausible driver of fire extent has exhibited a temporal pattern that might explain the historic decline in fire . For example , marked increases in human population density around the Serengeti [24] and changes in park fire management policies over the past few decades ( both of which alter the frequency of ignition events ) [11] might have been expected to overwhelm the effects of grazers in determining fire occurrence , but this was clearly not the case . An important caveat to our model results is the lack of direct data on grass biomass across the ecosystem . The link between wildebeest population size and standing grass biomass is implicit in our model , and would no doubt be strengthened by the availability of time-series data for grass biomass . Other studies , however , have shown both directly [25] and indirectly ( by estimating grass production and wildebeest consumption [21] ) that wildebeest can exert a very strong regulatory effect on grass cover in the Serengeti at landscape scales . This finding is consistent with the observation that at large enough spatial scales , it is fuel loads rather than ignition events that determine fire occurrence in savannas [26] . Our results also support the hypothesis that savannas are primarily regulated by fire ( and not rainfall ) above a mean annual rainfall threshold of 650–700 mm ( most of the Serengeti woodlands fall above this limit ) [5] , [22] . Variation in rainfall failed to directly explain patterns of tree density change , but it did play an indirect role by modulating the fire regime [27] . Notably , our results suggest that although elephants are known to exert important local effects on tree dynamics in Serengeti woodlands [12] , [13] , [28] , there is only weak support for the notion that elephants have influenced ecosystem-wide temporal patterns in tree density over the past half-century . Our model suggests that fire , rather than elephants , has been the key driver of tree density change in the Serengeti over the past half-century . A separate simulation model , drawing on different sources of data , predicted that both fire and elephants ( at their present-day population size , which is relatively high by historical standards ) can determine tree cover in the Serengeti , with fire being of greater importance [21] . There are , however , additional factors that must be considered in evaluating the overall importance of elephants for tree density . First , the elephant population of the Serengeti has historically been kept low by poaching . It is rapidly expanding at present , and in the future elephants could potentially exert large-scale impacts on vegetation . Second , elephants are patchily distributed in Serengeti [21] , [29] , and global assessments as summarized in our model do not capture spatial heterogeneity in their effects ( the same observation applies to fire ) , or localized interactions with fire and other factors [30] , [31] . An important future challenge will be to reconstruct and explain spatial patterns of tree cover change in this system . Third , elephants may have impacts on tree cover in savannas that are not reflected by changes in tree density . This is because they often feed on medium to large trees [12] , and their impact can reduce canopy cover ( thus having an impact on vegetation structure ) while maintaining density ( or even potentially increasing it , as a single large tree is replaced by several smaller recruits ) . Our results are consistent with the rinderpest trophic cascade hypothesis [15] , [16] , which proposes a linear chain of causality of remarkable simplicity operating in the Serengeti , one that zigzags vertically across three “trophic” levels: decreased pathogen→increased specialist consumer ( wildebeest ) →decreased producer ( grass ) →decreased generalist “consumer” ( fire ) →increased producer ( trees ) , mediating the relative dominance of two functional producer groups , trees and grasses ( Figure 4B ) . On the face of it , that a pathogen could regulate such a fundamentally important aspect of ecosystem structure as woody cover ( through its effects on an herbivore that does not even consume trees ) might seem improbable , but there is growing evidence of trophic cascades via subtle links in other ecosystems [32] , [33] , and , more broadly , increasing recognition of the role of pathogens in regulating plant communities [34] . We propose that the dominant factors controlling tree density in the Serengeti are top down , and that episodic top-down regulation of the herbivores by infectious disease has historically played an important role in restructuring this and ( potentially ) other ecosystems . In essence , the period of rinderpest enzoosis that prevailed throughout the first half of the 20th century in the Serengeti matches the scenario of the HSS ( Hairston , Smith , and Slobodkin ) “Green World” model [35] , but with a pathogen playing the role of predator and fire dynamics modulated by herbivory constituting a critical piece of the puzzle [16] . Although the scheme we propose in Figure 4B simplifies the range of possible interactions and feedbacks that could occur in an ecosystem as complex as the Serengeti ( e . g . , food availability as mediated by rainfall could affect the susceptibility of herbivores to disease ) , it captures what we believe to be some of the salient features of the system . Our simulations of C stocks suggest that the changes in wildebeest population density , fire prevalence , and tree density that have occurred over the past half-century may have had important effects on the C stocks in woody biomass ( Figure 5A ) . Furthermore , new field studies show that current densities of wildebeest and resident grazers stimulate storage of soil C ( unpublished data ) . Thus , our analysis allows us to estimate C loss and accumulation in the Serengeti ecosystem as a function of its trophic organization . A caveat to our estimates of tree biomass C is that , lacking data on changes in the size class distribution of tree over time , we must assume for simplicity that C stocks are directly proportional to density . This assumption might hold true when the size distribution is stable over time , but when changes in density are asymmetric across size classes ( e . g . , fire tends to remove small trees , elephants large ones ) , this assumption is violated . Better estimates of historic changes in tree C stocks will require data on tree size distribution changes over time , which are not yet available . Nevertheless , given our data , our best estimate is that Serengeti trees and soils constitute a net C sink , removing on the order of 40–70 Mg C km−2 y−1 from the atmosphere ( Figure 5B ) . Across 25 , 000 km2 of mostly protected woodland habitat across the entire ecosystem , this is equivalent to 106 Mg C y−1 . In contrast , our model suggests that in the past , when rinderpest was endemic and grazer densities were low , the Serengeti was a net C source . Rinderpest eradication may thus have had ecological consequences in the Serengeti that extend beyond the impact on habitat and landscape structure in this system . Furthermore , any future epizootic ( or any population crash from whatever cause , including disease , hunting , or drought ) may rapidly reverse the changes that have occurred over the past few decades and would release the C from its present stored form back into the atmosphere . A fundamental insight that emerges from the Serengeti longitudinal dataset is the value of the occurrence of external perturbations as proxies for manipulative experiments . The emergence and subsequent eradication of rinderpest resulted in multivariate transient dynamics , the pattern of which provides valuable information about the causal links that drive the system . At large spatial scales , manipulative experiments are infeasible , and deriving insights from natural experiments is an essential alternative for understanding the dynamics of complex systems at the landscape scale , which is a necessary step towards devising scientifically informed conservation policy in protected areas . Our results also show that wildlife conservation ( via control of illegal hunting and exotic diseases ) has an evident potential to make the Serengeti a substantial C sink in both wood and soils . This status could possibly allow the Serengeti to draw revenue for its management; the annual amount of C removed from the atmosphere by the system operates as a sink that could offset seats taken by tourists on flights from Europe and the rest of the world to East Africa or be marketed as CO2 offsets on carbon markets . This suggests a novel approach to maintaining the conservation status of this region by coupling park revenues to the economics of C offsets . Furthermore , even though the current status of the Serengeti as a C sink is unlikely to hold indefinitely ( the system will eventually saturate and become C-neutral ) , incentives are required that minimize the risk of the system becoming a net source of C should further disease outbreaks occur . The key point here is that the Serengeti may only work as an efficient C sink in the short term if it is grazed by over one million wildebeest ( Figure 5C ) . Their abundance is intimately dependent upon the control of infectious diseases and game-meat poachers [36] , as well as the continued viability of the migration , which is increasingly disrupted by land-use changes along the northern and western boundaries of the park [37] . The management of top-down trophic cascades can thus have important implications for how local ecological dynamics impact global-scale processes .
Serengeti National Park and the broader Serengeti-Mara ecosystem ( Serengeti hereafter ) have been described in detail elsewhere [11] , [25] , [38] . The ecosystem comprises an area of ∼25 , 000 km2 in Tanzania and Kenya in East Africa , and is characterized by a marked southeast to northwest rainfall gradient , as well as a roughly parallel gradient of increasing soil depth , sand to clay ratio , and declining fertility . It can be divided into areas of pure grassland in the southeastern plains and woodland in the rest of the ecosystem . The grasslands are the product of edaphic constraints [39] , and the woodlands vary spatially and temporally in terms of tree cover [28] , [40] . Wildebeest ( Connochaetes taurinus ) and elephants ( Loxodonta africana ) are dominant grazers and browsers , respectively , and can be regarded as keystone species in their respective feeding guilds , although giraffe ( Giraffa camelopardalis ) also have locally significant effects on trees [12] . We obtained wildebeest and elephant population estimates from census data [41] , [42] for the entire Serengeti ecosystem , and calculated the proportion of area burned in any given year from published [28] , [43] fire maps and our own database . We estimated mean per capita annual changes in tree density from sequential photopanoramas collected by A . R . E . S . at 51 sites in the Serengeti woodlands [11] , [44] between 1960–2003 ( Figure 1 ) . The sites were chosen in northern and central Serengeti to match photopanoramas that had been established at earlier dates ( pre-1960 ) and/or to achieve a good representation of road-accessible areas of the park . The time gaps between successive photopanoramas varied from site to site , and ranged between 2 and 31 y , resulting in sequences of between two and six photos per site ( Dataset S1 ) . To calculate observed annualized per capita changes in tree density ( ) across sites ( i ) and time periods ( j ) , we used the following equation: ( 1 ) where Ni , j and Ni , j+1 are the numbers of trees counted within a fixed frame ( reproducible across time periods ) inserted in photos j and j+1 , respectively , and yj+1−yj is the time elapsed between photos . Note that N are counts , but because trees were counted within fixed areas , we could treat r as a density change . It is a relative and not absolute density change because we could not measure the absolute areas covered by the frames . We used monthly rain gauge data to generate rainfall surfaces with inverse distance weighting , and estimated mean ecosystem-wide annual rainfall ( Rann , in mm ) , dry-season ( June–October ) rainfall ( Rdry , in mm ) , and the ratio of wet ( November–May ) to dry season rainfall ( Rw∶d ) for the period 1960–2003 . We used a poaching index ( P , dimensionless ) reconstructed from carcass data in the Serengeti [36] to model elephant population dynamics . We set P to 0 starting in 1990 on the basis of reports of negligible elephant poaching in the park following the ivory ban instituted in 1989 . To incorporate the effects of atmospheric CO2 on tree population growth , we used published values of CO2 ( C , in ppm ) from the Mauna Loa long-term dataset in Hawaii [45] . We reconstructed the history of rinderpest seroprevalence in the Serengeti for the periods 1958–1963 and 1982–1989 from the literature [46]–[48] . Raw data values for the model covariates used in the analysis are given below as text files for R and WinBUGS input . A technique that is increasingly gaining currency in ecological studies for the analysis of time series data with nonlinear dynamics , process and observation error , missing data , and latent variables is the BSS model using Gibbs sampling [49]–[52] . Given that our data analysis confronted all of these challenges , we adopted this approach to make inferences about the factors driving fire and tree population dynamics in the Serengeti . This framework allowed us to jointly model the population dynamics of the herbivores , which we treated as covariates , and fire and tree population dynamics . Some of the environmental covariates available for the Serengeti , such as annual rainfall , have been monitored continuously over the period of analysis , but herbivores have been censused unevenly over time; and for both elephants and wildebeest , the proportion of missing data exceeds 50% . To impute values for these missing data ( with appropriate error estimates ) , we required nonlinear population dynamics models incorporating both process error ( accounting for demographic and environmental uncertainty ) and observation error [50] . There were four dynamic variables that needed to be modeled: the total numbers of wildebeest ( W ) and elephants ( E ) , fire ( F ) , expressed as the proportion of the ecosystem that burns year−1 , and tree density ha−1 ( T ) . The BSS model allowed us to model probability distributions for the true values of these variables , both for years with and without missing data , by specifying probability models for each variable in year t conditional on: ( i ) its value in year t−1; ( ii ) the values of other variables hypothesized to affect it; and ( iii ) the observations [50] . We treated W and E as modeled covariates and F and T as dependent variables . We modeled W and E by drawing on past work supporting key effects of dry-season rainfall on wildebeest carrying capacity [53] , and of poaching [29] on elephant dynamics [54] . A BSS model generally comprises three components: a process equation describing the dynamics of the variable of interest ( e . g . , the true size of an animal population over time ) , an observation equation linking the process equation to the data , and prior distributions for the unknown parameters [50] . In this case , we have a multivariate time series of linked variables , so we have multiple process and observation equations [49]: Our immediate objective was to find , , Ft , and ( the actual values of interest , which we can express as a vector X ) together with the model parameters and error estimates ( which we jointly refer to as the vector θ ) that produced the best fit to the data , , , ( the vector of observations Y ) . The model described in Equations 2–13 can be expressed in terms of the joint likelihood of the variables and parameters for the period 1960–2003 , given the observations , or p ( X , Y| θ ) . The joint posterior distribution is proportional to this likelihood times the priors , and estimates of the X's and θ's can be obtained by sampling from this joint posterior [50] , which is difficult or impossible to do analytically . We used WinBUGS 1 . 4 [49] , [62] , which uses Gibbs sampling , a Markov Chain Monte Carlo ( MCMC ) technique [63] , to generate these estimates . We ran ten versions of the model ( Table 1 ) , combining alternative forms of Equations 4 and 5 , allowing for two different drivers of fire ( wildebeest and wet∶dry rainfall ratio ) and five of per capita tree density change ( fire , elephants , rainfall , wildebeest impact not explained by effects on fire , and atmospheric CO2 ) . We did not assess the potential contribution of human population increase on fire patterns for two reasons: first , we lacked sufficient data on human population change over the period in question; second , what we did have suggested that fire declined as the human population increased , making this explanation a poor a priori candidate for our fire model . We compared the fits of alternative models with the DIC , analogous to the AIC used in an information theoretic framework [64] , [65] . Our alternative versions of Equations 4 and 5 allowed us to simultaneously determine the relative importance of climate and herbivory on fire occurrence , and of climate , herbivory , fire , and atmospheric CO2 on tree population dynamics . The WinBUGS code for the best model ( model 3; see Table 1 ) is given in Protocol S1 . We ran each model for 106 iterations and discarded the first half of these as “burn-in . ” We used multiple initial values for each parameter and checked for model convergence with the Gelman-Rubin diagnostic [66] . We verified that our sampling interval did not lead to autocorrelation between successive realizations of each variable . We also examined the posterior distributions of all model parameters and variables to ensure that that they were not unduly constrained by the limits imposed by the priors ( in the case of uniform distributions ) and that they were approximately normally distributed . To put our results into perspective for readers unfamiliar with Bayesian approaches , we plotted observed versus predicted ( by the state-space model ) values for fire and tree cover change and calculated adjusted-R2 values as approximate indicators of the amount of variance explained by the best model ( Figures 1F and 1G ) . We took as our predicted values the mean of the posterior distribution for each response variable . Although the Bayesian approach generates distributions rather than point estimates , we treated these means as our best estimates of model predictions . We noted a number of outliers in the plot of observed versus predicted values of ri , j ( Figure 1G ) , even after accounting for site differences in tree population change . We hypothesized that these particularly high observed values of annualized relative growth might be associated with the initial tree densities in these sites , so we plotted the model residuals ( robs−rpred ) against the logarithm of N1 , the tree count at the beginning of each paired photo sequence . Although we found that initial tree abundance explained almost an additional fifth of the total variance in tree population growth ( Figure 1H ) , we could not parameterize the exact magnitude of this effect because we were unable to standardize tree densities across photos . To estimate ecosystem-level C fluxes in the Serengeti as a result of changes in wildebeest population size , fire , and tree density , we simulated changes in the size of the two dominant ecosystem C pools , tree C , and SOC . Our own analysis indicated large shifts in tree density , and recent empirical and modeling studies support the existence of dominant fire and grazing effects on SOC ( unpublished data ) [18] , [19] , so we focused our analysis on these three effects . We explicitly simulated the dynamics of tree C to calculate annual changes in biomass C , and estimated gains/losses from the soil C pool caused by grazing and fire from equations derived empirically ( unpublished data ) and through modeling of soil nutrient dynamics [18] , respectively . We did not explicitly model the dynamics of the SOC compartment because the fluxes we report are small in relation to the absolute size of the total soil C pool , and we could significantly simplify our analysis by treating SOC as a pool of constant size ( to a first approximation ) over the relatively short time scale of the analysis . We modified the best overall state-space model ( model 3 in Table 1 of the main text ) by expressing tree density T in C units ( Mg C km−2 ) . We obtained a point estimate of tree C for 1999 ( of 997 Mg C km−2 ) in the woodland portion of the ecosystem by combining our plot data [40] with allometric equations relating stem and crown diameter with aboveground and belowground biomass in Acacia tortilis [67] , the most common tree species in the ecosystem . We then converted biomass into tree C per km2 in the survey plots across tree size classes . Fire effects vary widely across tree size classes [12] , and much of the woody biomass in large trees does not burn and volatilize in the short term [68] . Because our tree data does not discriminate among size classes , however , we can not incorporate this size distribution effect and treat our estimates of biomass C fluxes only as approximations . We used the estimated 1999 value in combination with the model to estimate woody biomass C for the entire period 1960–2003 , the same way we previously did with density . To simulate the effect of grazing on the soil C pool , we used the following empirically derived polynomial equation ( unpublished data ) relating SOC flux to ( GI ) : ( 14 ) where ΔSOC is in units of Mg C km−2 y−1 and GI equals the proportion of aboveground net primary production ( NPPt , grasses only ) consumed by grazers ( CONSt ) . To estimate GI we first had to estimate NPPt and CONSt on the basis of rainfall and the size of the wildebeest population . We used an empirically derived equation relating NPPt to annual rainfall ( Rann ) to estimate annual production [25] in Mg km−2 y−1: ( 15 ) The correction factor of 0 . 6 adjusts the production estimate to account for bare ground , topography , rivers , etc . [21] , [38] . To estimate CONSt ( in MG DM km−2 y−1 ) on a unit area basis ( assuming a total area of 25 , 000 km2 ) we used the following equation: ( 16 ) where 1 . 79 ( in Mg DM ) is our estimate of annual consumption for an average wildebeest based on empirically derived functions relating daily voluntary intake to body mass [69] . In our analysis we only model wildebeest , but numerous other grazing species ( such as buffalo ) have covaried numerically with wildebeest as a result of rinderpest eradication and poaching pressure [15] , [36] . On the basis of census data , we estimate that wildebeest represent 54% of the biomass of Serengeti grazers on a metabolic basis ( which maps to consumption ) , and use this value in Equation 16 to generate a realistic estimate of historic consumption patterns for all grazers . We used Equations 15 and 16 to estimate GI = CONSt/NPPt on an annual basis , and applied this estimate to Equation 14 to estimate ΔSOCt . Our mean simulated estimate of GI for the period 1974–1977 ( 0 . 55 ) compared favorably with a mean field-based estimate of 0 . 52 obtained for this period [25] . To estimate ΔSOCt as a function of fire , we first used a published model of Serengeti soil organic matter ( SOM ) dynamics [18] to estimate mean annual SOM changes in the top 10 cm of soil ( the layer most susceptible to fire-induced SOM losses [70] ) as a function of fire frequency . We estimated maximum annual SOM ( and SOC ) losses of 0 . 8% y−1 with an annual fire regime ( Ft = 1 ) . These estimates are consistent with long-term values measured elsewhere [71] . We used a linear interpolation ( with ΔSOCt = 0 with no fire ) to estimate ΔSOCt as a function of area burned ( Ft ) , as follows: ( 17 ) based on mean values of SOM of 7 . 8% [19] and a mean bulk density of 1 . 21 [25] . To estimate changes in total ecosystem C , we modified our best-fit state-space model ( model 3 in Table 1 ) to simulate Δ tree Ct+ΔSOCt over the period 1960 to 2003 based on inferred values of Wt , Ft , and Tt and Equations 14–17 . We adjusted Δ tree Ct in our calculations of total ecosystem C change by a factor of 2/3 to account for the fact that one third of the ecosystem consists of edaphic , tree-less grasslands . To smooth out the high degree of inter-annual variation in Δ tree Ct+ΔSOCt , we present our results as mean annual changes calculated over decadal intervals . | Diseases are known to play important roles in regulating and structuring populations , but the consequences of disease outbreaks for entire communities and ecosystems are not as well understood . The Serengeti wildebeest were historically kept at low numbers by the rinderpest virus , but underwent a population explosion ( irruption ) after rinderpest was eradicated in the 1960s . We examined nearly a half-century of data to test the hypothesis that this irruption was responsible for a decline in the frequency of fires in this ecosystem ( through increased grazing and a reduction in fuel loads ) , and that this in turn increased the density of trees . We found strong evidence for this indirect link between rinderpest and tree density , and less support for the role of other factors such as elephants and climate . We also investigated the consequences of this chain of events for ecosystem carbon , and suggest that the combined effects of increased grazing intensity by wildebeest , reduced fire , and increasing tree density may have shifted the Serengeti from being a net source to a net sink for carbon . This would imply that seemingly small ecological perturbations such as disease outbreaks have the potential to profoundly affect ecosystem function . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"ecology/plant-environment",
"interactions",
"ecology/global",
"change",
"ecology",
"infectious",
"diseases",
"ecology",
"ecology/community",
"ecology",
"and",
"biodiversity",
"ecology/population",
"ecology",
"ecology/ecosystem",
"ecology"
] | 2009 | A Disease-Mediated Trophic Cascade in the Serengeti and its Implications for Ecosystem C |
Meteorological factors influence dengue virus ecology by modulating vector mosquito population dynamics , viral replication , and transmission . Dynamic modeling techniques can be used to examine how interactions among meteorological variables , vectors and the dengue virus influence transmission . We developed a dengue fever simulation model by coupling a dynamic simulation model for Aedes aegypti , the primary mosquito vector for dengue , with a basic epidemiological Susceptible-Exposed-Infectious-Recovered ( SEIR ) model . Employing a Monte Carlo approach , we simulated dengue transmission during the period of 2010–2013 in San Juan , PR , where dengue fever is endemic . The results of 9600 simulations using varied model parameters were evaluated by statistical comparison ( r2 ) with surveillance data of dengue cases reported to the Centers for Disease Control and Prevention . To identify the most influential parameters associated with dengue virus transmission for each period the top 1% of best-fit model simulations were retained and compared . Using the top simulations , dengue cases were simulated well for 2010 ( r2 = 0 . 90 , p = 0 . 03 ) , 2011 ( r2 = 0 . 83 , p = 0 . 05 ) , and 2012 ( r2 = 0 . 94 , p = 0 . 01 ) ; however , simulations were weaker for 2013 ( r2 = 0 . 25 , p = 0 . 25 ) and the entire four-year period ( r2 = 0 . 44 , p = 0 . 002 ) . Analysis of parameter values from retained simulations revealed that rain dependent container habitats were more prevalent in best-fitting simulations during the wetter 2010 and 2011 years , while human managed ( i . e . manually filled ) container habitats were more prevalent in best-fitting simulations during the drier 2012 and 2013 years . The simulations further indicate that rainfall strongly modulates the timing of dengue ( e . g . , epidemics occurred earlier during rainy years ) while temperature modulates the annual number of dengue fever cases . Our results suggest that meteorological factors have a time-variable influence on dengue transmission relative to other important environmental and human factors .
In the last decade dengue infections have increased dramatically in the Americas with cases now occurring in the southern U . S . , Mexico and Central America , across the Caribbean , and as far south as Argentina in South America [1] . The Pan American outbreak in 2010 resulted in 1 . 7 million cases of dengue fever ( DF ) including 21 , 206 cases in Puerto Rico [1] . There are four serotypes of the dengue virus ( DENV ) , and subsequent infection with a new serotype increases the risk of severe dengue which can manifest as hemorrhagic fever ( DHF ) . While DHF case-fatality is fairly low , ranging between 0 . 5% and 5 . 0% [2] , the global burden of DENV infection is extremely high with an annual estimate of 390 million infections of which 96 million result in symptomatic disease [3] . Given the high burden of disease and that transmission is directly and indirectly regulated by meteorological factors , understanding how dengue dynamics may shift under different meteorological conditions is a key public health question [4–6] . Meteorological factors influence many components of DENV ecology , most directly through the Aedes ( Ae . ) mosquito vector [7] . Temperature and precipitation are important drivers of mosquito population dynamics [8] . Temperature influences development rates , mortality , and reproductive behavior [9–12] . Precipitation often provides water in containers that serve as larval and pupal habitat . Container water volume influences development rates and low water volume can increase mortality through enhanced competition between larvae and pupae at higher population densities [13 , 14] . Ambient air temperature also influences viral replication within the adult mosquito and is a key regulator of the length of the extrinsic incubation period ( EIP ) , the time between when a mosquito is infected and becomes infectious [15–17] . A shorter EIP reduces the length of the cycle of transmission and increases the probability of its completion during the lifespan of Ae . aegypti . Despite the established connections between meteorological variables and many components of DENV ecology , the relative influence of weather versus human factors on dengue epidemics is still unclear . For example , although much of the southern United States is inhabited by the vector Ae . aegypti and lies in close proximity to areas where dengue is endemic , locally acquired infections are rare . Reiter at al . [18] contend that interactions between the vector and human are limited in Laredo , Texas as compared to across the border in Nuevo Laredo , Tamaulipas , Mexico due to better infrastructure which prevents vector-human contact and , thus , limits transmission . Other studies have also found human-related factors to be of greater importance than climatic suitability . Keating [19] and Brunkard at al . [20] both found meteorological variables to influence DF cases but suspect that herd immunity , circulating serotype , and strain play a larger role in transmission . Additionally , human responses to climate may be as important as climate itself . In Australia , Beebe at al . [21] and Kearney at al . [22] found that DENV vector habitat may increase due to human water storage designed to combat drought . The complexity and uncertainty of the many factors involved in the disease system make predictions and effective interventions to reduce transmission difficult . Process-based models can be useful in determining the relative influence of human versus meteorological factors on DENV transmission . Vector dynamics are frequently tied to variations in temperature and precipitation [23] . Dynamic models are built using known biophysical relationships between the vector , virus and the environment . Mathematical relationships are derived from studies on rates of vector development , mortality , and generational progression as well as thermal limits for population survival [11 , 24] . Vaidya et al . [25] , for example , built a model of mosquito population dynamics using temperature dependent maturation and mortality rates and precipitation dependent carrying capacity . Such models have successfully used meteorological inputs to simulate vector-borne disease dynamics [26–28] including models for Aedes mosquitoes that are the vectors of dengue [29 , 30] . Other uses include exploration of outbreaks based on weather scenarios , and investigations of land use/cover change , and the evaluation of intervention strategies [31] . Because many parameters involved in DENV transmission are unknown , the ability to perform simulations under a variety of conditions and scenarios is an important and powerful aspect of dynamic models that can help characterize parameter uncertainty and whether parameter values vary in time . Process-based models that simulate vector dynamics are rare and most do not calculate the EIP nor do they simulate transmission between the vector and human populations , in large part due to the complexity of the single and combined models [7] . But vector density does not always correlate with disease incidence [6 , 32] . A review of dengue transmission modeling approaches by Andraud et al . [33] promotes the use of models that include a vector component for informing public health policy . One of the first models that integrates the two components ( human and vector ) was developed by Focks et al . [34] . It integrates a dynamic vector population model ( the container inhabiting mosquito simulation model , CIMSiM ) with a dengue transmission model ( the dengue simulation model , DENSiM ) . While CIMSiM model outputs have been validated in the field , some parameters remain highly uncertain , emphasizing the importance of developing alternative process-based models of dengue transmission [35] . Skeeter Buster , for example , has expanded on CIMSiM to include population genetics , spatial heterogeneity in habitat availability , and stochastic effects [36] . Andraud et al . [33] and another review by Johansson et al . [37] both identify model parameterization as a key challenge for simulating dengue transmission . However , employing numerous models , or many slightly-differing versions of the same model , allows researchers to explore and quantify uncertainty in the parameter space , and model ensembles have been shown to be more accurate than any single model [38] . In this study we present a new modeling framework that couples a dynamic mosquito life-cycle model calibrated for Ae . aegypti with a SEIR ( susceptible-exposed-infectious-recovered ) model for dengue transmission to simulate dengue outbreaks in San Juan , Puerto Rico from 2010–2013 ( available for download at https://sites . google . com/site/dymsimmodel/home ) . This work expands upon the existing Dynamic Mosquito Simulation Model ( DyMSiM ) [8] by incorporating a virus transmission component and parameterizing the model for the mosquito species Ae . aegypti . The model incorporates newly collected and consolidated research related to temperature effects on vector survival and development [39] and the dengue virus extrinsic incubation period [17] . Using a Monte Carlo approach , simulations are performed for numerous combinations of parameter values and the results are evaluated using dengue case data from San Juan , PR reported to the Centers for Disease Control and Prevention ( CDC ) Dengue Branch . The best model simulations over the entire time period and for individual years are analyzed to determine the relative influence of different meteorological and human-mediated factors on DF case numbers and to identify how and why parameter values changed between years .
San Juan is a municipality of Puerto Rico and is located in the northeastern part of the Caribbean island . The population was 395 , 326 in 2010 ( US Census Bureau , 2010 Census ) . San Juan has a humid tropical climate with minimal variation in seasonal temperature . Precipitation occurs all year , but it is notably drier during boreal winter and early spring . San Juan is an ideal study location because dengue is endemic in the municipality , and weekly dengue case data are available through CDC ArboNet ( USGS , 2014 ) from 2010–2013 . During that time , annual case numbers ranged from 500 ( 2011 ) to 919 ( 2012 ) . These data are crucial for model evaluation . The mosquito population was simulated using the general structure of the Dynamic Mosquito Simulation Model , ( DyMSiM ) [8] but the model was parameterized for Ae . aegypti mosquitoes with additional components added , including the epidemiological SEIR model ( S2 Table ) . This enabled the simulation of virus transmission between the human and mosquito populations in DyMSiM . The model is deterministic and was implemented using Euler’s method of integration in Stella 10 . 06 software . A conceptual diagram of model processes is provided in Fig 1 . Parameter value constants and equations are provided in S1 Table , governing equations for human and mosquito populations are provided in S2 Table , and further details are discussed below . Because it is difficult to obtain precise measurements for many of the DyMSiM parameters , especially those related to immature container habitat , we used a Monte Carlo approach to assess outcomes for a range of possible parameter values . We performed 9600 simulations with DyMSiM using discrete parameter values representative of the distribution of values identified in the literature and the preliminary model simulations ( Bold in Column 3 of S1 Table ) to replicate DENV transmission in San Juan , Puerto Rico from 2010–2013 . During each simulation , one value was changed until all parameter value combinations were used . All humans and mosquitoes were assumed to be susceptible at the beginning of the runs , however , runs were started in 2009 to provide spin-up time to build infection and immunity ( human only ) within the human and mosquito population . Additionally , a background infection rate was used to initiate infections and is based off of the average number of dengue cases that occur during the low transmission season according to the 2010–2013 case data for San Juan . The assumption that the population is completely susceptible is not accurate , however , given the low incidence ( 3192 total cases over a population of 395 , 326 ) and short time interval over which the model is run , the influence on outcomes should be minor . Weekly CDC reported case data were used to evaluate the model . Container habitat area estimates were determined by first performing preliminary model simulations under parameter values that would produce a maximum and minimum number of dengue cases in San Juan . By comparing these runs with the reported dengue data , an estimated range of habitat area was determined for the study . Because there is great uncertainty in the container habitat area and it strongly influences mosquito population size , the number and range of values chosen greatly exceeds that of the other parameters . Fewer values and smaller ranges were chosen for parameters where more research has been performed resulting in greater confidence in the selected values .
There were 3 , 192 cases of DF reported during the study period with 901 cases occurring in 2010 , 500 cases in 2011 , 919 cases in 2012 , and 872 cases in 2013 . Reported cases began increasing during week 20 in 2010 , 2012 , and 2013 but occurred later ( week 25 ) in 2011 which was relatively cool compared to the other years and experienced the fewest number of cases ( Fig 2 ) . During most years ( 2010 , 2011 , and 2013 ) the peak in reported DF cases occurred between weeks 32 and 37 , however , during 2012 the increase in cases was much more gradual and the peak did not occur until week 49 and did not decline to minimum season case numbers until week 15 of 2013 . Temperatures were also much warmer during the second half of 2012 compared to other years , while precipitation was notably lower , especially during summer and fall . Although 2013 experienced similar reported case numbers as 2010 and 2012 , most cases occurred during the first 15 weeks of the year ( Fig 2 ) with fewer cases occurring during its summer peak . The decline to reported minimum season case numbers also occurred earlier than the other years . Aside from 2012 , increases in reported DF followed rains occurring in late spring . The earliest increase and peak in reported DF cases occurred in 2010 which also experienced the warmest spring temperatures and earliest precipitation . Statistical measures of accuracy between the reported data and ensemble averages using the top 1% ( n = 96 ) best-fit simulations , for both the entire time period ( S2 Dataset ) and individual years ( S3–S6 Datasets ) , are reported in Table 1 . Although inter-annual variability in DF cases was well-simulated in the ensemble average from the 96 model simulations over the entire time period ( r2 = 0 . 44 , p = 0 . 002 ) , the annual epidemic curves for individual years often lacked precision ( Fig 2 ) . When evaluating ensemble averages from the simulations over individual years ( same simulations but the statistics are only performed on reported case data for individual years ) , however , intra-annual variability was replicated with much higher accuracy for 2010 ( r2 = 0 . 90 , p = 0 . 03 ) , 2011 ( r2 = 0 . 83 , p = 0 . 05 ) , and 2012 ( r2 = 0 . 94 , p = 0 . 01 ) ( Fig 3 ) . Simulations for 2013 were markedly less accurate ( r2 = 0 . 25 , p = 0 . 25 ) compared to the other years ( Fig 3 ) . By contrast , the yearly model accuracy assuming no a-priori knowledge to guide parameter selection was lower and more variable: the r2 values for 2010 , 2011 , 2012 , and 2013 were 0 . 77 , 0 . 64 , 0 . 05 , and 0 . 12 respectively . Fig 4 displays the number of times each parameter value was used in the top 1% of simulations . Notably , in 2010 and 2011 the number of simulations retained that used a high proportion of open containers ( i . e . , more precipitation dependent sources ) vs . a low proportion of open containers ( i . e . , more human-managed sources ) is much larger than in 2012 and 2013 ( 76 and 59 vs . 0 and 0 ) . Additionally , patterns in container habitat area , background infection rate , and carrying capacity vary considerably between years . The retained simulations for most years used the higher adult daily survival rate , and the shorter length of infectious period ( except 2012 ) . With the exception of 2011 , the range of values tested for container height and host infection probability are used with equal frequency between years indicating that they did not have a strong impact on temporal variation in dengue transmission dynamics . Fig 5 displays the average rank of simulations using each parameter value among all 9600 simulations . Simulations that used a high proportion of open containers ranked slightly better for 2011 , while the opposite is true for 2012 and all years . This is consistent with the trend in the top 1% of simulations , however , there is no strong difference for 2010 and 2013 . Simulations using higher values for container habitat area were markedly worse , though this effect was lesser for 2012 . Additionally , the lower carrying capacity value tended to produce better simulations . Average ranks among different values did not vary markedly for the other parameters . Precipitation was highest in 2010 ( 227 . 5 cm ) and 2011 ( 224 . 0 cm ) , with the onset of the rainy season starting a few weeks earlier in 2010 ( Fig 6 ) . Precipitation was by far the lowest in 2012 ( 140 . 3 cm ) , however , temperatures were warmest in 2012 , especially during late fall ( ~1°C warmer in November ) . Precipitation was 216 . 3 cm in 2013 –only slightly drier than 2010 and 2011—but the timing of the precipitation was different , exhibiting a marked drop off during late summer/early fall .
While we used the best available information from the literature to specify values of the DyMSiM parameters , some values are based on the results of only a few studies or were reasonable estimates when quantitative results were unavailable . Papers that consolidated and considered data from multiple studies proved especially useful because they accounted for variations arising from conducting research on different colonies of mosquitoes and strains of virus or using different methods to answer the same question [17 , 39] . Often averages were used to enhance simplicity and interpretability of the model; yet , many parameters may have values across an unknown distribution surrounding their mean due to genetic and environmental variability . For example , genetic variations have been found to alter temperature regulation of West Nile virus transmission [63] . If a similar variation in temperature sensitivity exists between or within dengue serotypes it could be an important component to the severity of DF epidemics . We attempted to mitigate our limited knowledge of some parameter values by using a Monte Carlo approach when they were particularly difficult to ascertain . The model showed that some parameters , such as container habitat area and composition of containers , were particularly important for simulating transmission . The carrying capacity density was also important as it influenced the mosquito productivity within containers , and was directly related to their area . For example , simulations were degraded when using larger container areas , but this effect was lessened when using the lower maximum larval density value . The value of the background infection rate was also important when selecting the best fit models . This may indicate the importance of varying virus introduction or virulence between years . Vertical transmission of the virus was not considered in this model though it has been observed in the laboratory and field setting and could help maintain the virus during unfavorable conditions [64 , 65] . However , Adams et al . [66] found it did not have considerable importance in their model and the frequency of its occurrence in nature is still uncertain [67] . As more evidence is generated establishing links among meteorological factors , vectors , virus , and humans , parameter values can be better defined . Some of the inconsistencies between modeled and reported case data may also be attributed to the nature of surveillance data . Because the surveillance system is passive , the reported case data are only an approximation of total DF incidence for the municipality . Incomplete or inaccurate reporting can arise due to sub-clinical infections , misdiagnosis , failure to report cases , differences between where virus transmission is reported and where it was acquired , and variability in the time between transmission of the virus , development of symptoms , and clinical visits . Relative increases in dengue reporting may be seen if there is increased media attention , circulation of a more severe strain , or increases in laboratory testing ability , etc . The serotype of the virus is often not reported but could be of particular importance if a high proportion of the population is immune to the circulating serotype . It is likely that not all cases are being reported to the CDC and , therefore , the reported case data is only a subset of the actual number of cases occurring in San Juan though it is difficult to estimate the amount of underreporting . The influence of under-reporting should be minimal if the patterns of reported cases represent the yearly variability even if the overall magnitude of reported cases is inaccurate . The largest bias in the calculation of the correlation between modeled output and reported case numbers would occur if there were significant changes to the reporting system or if there are interannual differences in reporting . There were no significant changes to the surveillance system protocol during this time period . It is difficult to determine inter-annual variability in reporting , but there is the possibility for bias . For example , if individuals at the beginning of the season were more likely to seek care and be reported through the surveillance system , or if physicians were less likely to diagnose and report cases during typically lower periods of dengue , the shape of the curve might shift . Given the uncertainty of these types of scenarios , trying to determine the relative impact they may have on the correlation between model output and reported case numbers would be highly speculative . We acknowledge , however , that under-reporting of dengue is an important issue and therefore , the model output should not be interpreted as an estimate of actual dengue case numbers but as a representation of inter-annual variability of DENV transmission . If more reliable estimates of reporting bias and information on the temporal variability in asymptomatic cases or the reporting of symptomatic cases become available , it would be useful to compare numbers of modeled and reported dengue case data . Despite these limitations , the reported case data are currently the best metric available for tracking DENV transmission and have been used in other studies to represent levels of DENV transmission in Puerto Rico [4 , 61 , 68] . The results of this study provide important information for dengue control . First , epidemics can occur during both wet and dry years . Dengue cases were comparatively high in 2010 , one of the wettest years on record for San Juan , and in 2012 , a drier-than-normal year in which the rains came later than average . The warmer winter and spring temperatures are likely responsible for the greater number of DF cases during 2010 while warmer fall and winter conditions in 2012 likely helped propagate dengue transmission later in the year . It is common for dry years to be hotter due to a combination of factors that include: 1 ) fewer clouds facilitating higher daytime temperatures and 2 ) less evapotranspiration freeing a greater amount of solar radiation for heating . Second , the timing of the dengue cases varies substantially from year-to-year and appears to be linked with the timing and magnitude of rainfall . In the wetter 2010 year , DF case numbers rose sharply after the onset of the rains ( Week 10 ) and peaked near Week 30 , whereas in the drier 2012 year the cases began later ( Week 20 ) after a highly sporadic beginning to the rainy season , and gradually climbed to a much later peak near Week 48 . In both instances , the onset of DF cases coincides with or begins shortly after the onset of the rainy season , and rises steadily due to the time required for the mosquito population to development in newly formed habitats and completion of the EIP . Lastly , despite a tropical climate , small fluctuations in temperature can have considerable effects on DENV transmission . Often tropical environments are thought to be warm enough for year-round transmission with precipitation being the limiting factor . Our results indicate that nonlinear relationships result in greater temperature impacts on DENV transmission than would be expected . Monitoring climate as a proxy of dengue risk may provide public health workers with a simple tool to prepare and execute transmission intervention methods and arrange testing and treatment protocols before the DF season peaks . Early warning systems can be developed based on the identification of rainfall and temperature patterns that promote rapid dengue transmission [69–72] . Dominant Ae . aegypti container habitat appeared to vary with precipitation patterns . During the wetter years , 2010 and 2011 , simulations using a smaller container area and more rain dependent containers were ranked better; however , during 2012 simulations using a greater amount of human managed water sources ranked better . This suggests that Ae . aegypti inhabit rain filled containers when available , in addition to human managed water sources , but drier conditions may limit immature mosquito habitats to permanent ( septic tanks , large water tanks , etc . ) or human managed ( animal drinking pans , plant trivets ) water sources . This pattern has been observed in field studies [58] . From a public health perspective , the magnitude of precipitation in a given year may inform how to refine the message conveyed to the public about how to limit mosquito habitat . During wet years , mosquito control strategies should focus on eliminating old tires , buckets , and other items that can be filled via precipitation . Drier years should shift the focus to treating permanent or human managed water sources and covering stored water to prevent mosquito infestation . Further investigation of how weather and rainfall patterns influence human behaviors such as water storage are needed to determine if this could be incorporated into the model . Understanding weather influences on DENV transmission is important in the context of climate change especially as DENV range expands in the Americas . Outbreaks in Hawaii , along the Texas/Mexico border , and in southern Florida indicate a reemergence of dengue in the U . S . [73–75] . Additionally , sero-prevalence studies indicate that dengue cases are likely underreported in the U . S . [73 , 75] due to subclinical infections , insufficient resources for testing , lack of sensitization of the medical community and patients seeking medical attention across the border in Mexico [75] . Although infrastructure differences , such as the prevalence of air conditioning , were shown to reduce transmission on the US side of the Mexico border in Laredo , autochthonous cases still occur [18] . Viral introduction in response to frequent travel across the border by residents may be a large source of infections [76] and weather may still have an important influence on transmission risk [20] . Whether DENV is endemic to an area or currently only poses a threat , understanding the influence of weather and climate on DENV ecology can facilitate strategies to prevent or mitigate transmission . Using known relationships between climate variables , Ae . aegypti dynamics , DENV replication and transmission , and a basic SEIR model within a modeling framework known as DyMSiM we were able to simulate inter-annual variability in DF cases in San Juan County , PR for the years 2010–2013 . When optimizing the DyMSiM parameters for a single year , we were able to simulate intra-annual variability well for three of the four years . Meteorological factors were important influences on dengue transmission for 2010 , 2011 , and 2012 but not as important in 2013 . Parameter values changed in response to meteorological conditions , illustrating the complexity of DENV ecology . The results of this study quantify meteorological impacts on DENV transmission and demonstrate the variance in parameter values that result from changing environmental conditions . Rain patterns were especially important for determining the timing of epidemics and the primary habitat for immature Ae . aegypti vectors . Abnormally high temperatures have the potential to extend the transmission season ( even during dry years ) as exemplified in 2012 . In general , temperature had the greatest influence on annual case numbers . The sensitivity of DENV ecology to meteorological variables and their interactions underscores the utility of process-based modeling when studying the impacts of climate variability and climate change on vector-borne disease . Information from this study can be used by public health officials to improve DENV transmission intervention strategies in San Juan , PR by way of advanced preparedness through model predictions and creation of more targeted vector control campaigns . As our understanding of the ecology of the virus and its thresholds improves , so will our ability to implement effective mitigation strategies against DENV transmission and reduce the disease burden on human populations . Future research endeavors will seek to enhance model performance and utility . Validation of DyMSiM in other areas with longer term datasets would be useful in determining its ability to simulate DENV transmission in other environments and its ability to simulate DF incidence over long time periods . The deficit of long , reliable dengue surveillance records has forced training of dengue models over periods of less than 10 years and validation on one or only a few years [70 , 77 , 78] . Inclusion of serotypes would also enhance the model given that immunity to specific virus serotypes can affect transmission dynamics by limiting the susceptible population . The current SEIR model is simple but can still be used to examine the dynamics of dengue over short time periods . Incorporating surveillance of multiple virus serotypes , including short term cross-immunity , will be a priority in future research given the importance of immunological status on virus transmission dynamics . Additionally , the inclusion of social factors that influence population risk , such as migration , age structure , personal and household level prevention behaviors and other factors influencing vector-human contact , may provide further insights into the nature of DF epidemics . Finally , future work will concentrate on the possible impacts of climate change on dengue transmission . For instance , under climate change , winter temperatures may become suitable for completion of the EIP allowing DENV transmission to continue . Understanding the effects of future weather and climate conditions on DENV transmission is vital for limiting increases in incidence of disease and range expansion . | Numerous studies have investigated meteorological and climatic influences on mosquito transmitted viruses . However , dengue ecology is complex , necessitating an understanding of the interactions among components in the system . We estimate dengue fever cases in San Juan , Puerto Rico using a mathematical model informed by relationships among meteorology , land cover , and interactions among human hosts , mosquitoes , and the dengue viruses identified from the literature . Because some of these relationships are not well known or static , we performed several thousand simulations and compared model output to dengue fever cases reported to the Centers for Diseases Control and Prevention . The model replicated reported dengue cases well , but factors related to dengue transmission patterns varied between years . During wetter years , precipitation-filled containers were the primary immature mosquito habitat in the model . Conversely , during drier years , containers filled with water by humans were the most important habitat . In warmer years there was an increased number of dengue cases that peaked following higher rainfall . These results reveal that current climatic conditions modify the relative influence of human and climatic factors on dengue transmission patterns . This knowledge can be used to develop forecasting tools for dengue outbreaks and enhance mosquito control campaigns based on weather predictions . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | Meteorologically Driven Simulations of Dengue Epidemics in San Juan, PR |
The binding of human complement inhibitors to vaccine antigens in vivo could diminish their immunogenicity . A meningococcal ligand for the complement down-regulator , factor H ( fH ) , is fH-binding protein ( fHbp ) , which is specific for human fH . Vaccines containing recombinant fHbp or native outer membrane vesicles ( NOMV ) from mutant strains with over-expressed fHbp are in clinical development . In a previous study in transgenic mice , the presence of human fH impaired the immunogenicity of a recombinant fHbp vaccine . In the present study , we prepared two NOMV vaccines from mutant group B strains with over-expressed wild-type fHbp or an R41S mutant fHbp with no detectable fH binding . In wild-type mice in which mouse fH did not bind to fHbp in either vaccine , the NOMV vaccine with wild-type fHbp elicited 2-fold higher serum IgG anti-fHbp titers ( P = 0 . 001 ) and 4-fold higher complement-mediated bactericidal titers against a PorA-heterologous strain than the NOMV with the mutant fHbp ( P = 0 . 003 ) . By adsorption , the bactericidal antibodies were shown to be directed at fHbp . In transgenic mice in which human fH bound to the wild-type fHbp but not to the R41S fHbp , the NOMV vaccine with the mutant fHbp elicited 5-fold higher serum IgG anti-fHbp titers ( P = 0 . 002 ) , and 19-fold higher bactericidal titers than the NOMV vaccine with wild-type fHbp ( P = 0 . 001 ) . Thus , in mice that differed only by the presence of human fH , the respective results with the two vaccines were opposite . The enhanced bactericidal activity elicited by the mutant fHbp vaccine in the presence of human fH far outweighed the loss of immunogenicity of the mutant protein in wild-type animals . Engineering fHbp not to bind to its cognate complement inhibitor , therefore , may increase vaccine immunogenicity in humans .
Neisseria meningitidis causes sepsis and meningitis with relatively high rates of fatalities or severe permanent sequelae [1] , [2] . Licensed quadrivalent polysaccharide-protein conjugate vaccines are available against four capsular groups: A , C , W135 and Y . Development of conjugate vaccines against group B strains , however , has been hampered by cross-reactivity of the group B polysaccharide with host molecules [3] , [4] , and safety concerns about the potential to elicit auto-reactive antibodies . Development of a vaccine against group B strains is important since these strains are responsible for about one-third of cases of meningococcal disease in the U . S . [1] and up to 90% in some European countries [5] . Several non-capsular antigen-based vaccines are being developed against group B meningococci ( reviewed in [6] , [7] ) . One of the most promising antigens is factor H-binding protein ( fHbp ) [8] , [9] . Vaccines containing recombinant fHbp [10]–[12] or native outer membrane vesicles ( NOMV ) from mutant meningococcal strains with over-expressed fHbp [13] , [14] are being tested in humans . After clinical testing had started , fHbp was discovered to bind complement factor H ( fH ) [15] . Further , binding was found to be specific for human fH [16] . Binding of a host protein to a vaccine antigen could in theory decrease immunogenicity by covering important epitopes or decreasing uptake , processing or presentation of the antigen . Also , the implications of binding a human complement protein to a vaccine antigen with respect to its effect on immunogenicity or the potential safety concern of eliciting auto-antibodies had not been considered at the time of starting the clinical trials with these vaccines . Using transgenic mice , we recently reported that the presence of human fH impaired immunogenicity of a recombinant fHbp vaccine that bound human fH [17] . In that study we also described a mutant fHbp antigen in which substitution of arginine 41 with serine ( R41S ) abrogated binding of fH to the fHbp vaccine . Serum bactericidal antibody responses of human fH transgenic mice immunized with the recombinant R41S mutant fHbp vaccine were higher than those of mice immunized with the recombinant wild-type fHbp vaccine that bound fH . In wild-type mice , NOMV vaccines from mutants with genetically attenuated endotoxin and over-expressed fHbp elicited high titers of serum anti-fHbp antibodies , which had broader ( i . e . in terms of strain coverage ) complement-mediated bactericidal activity than anti-fHbp antibodies elicited by recombinant fHbp vaccines [18]–[23] . Thus , NOMV vaccines from mutants with increased fHbp expression appear to be a promising approach to elicit broad protective immunity against meningococcal group B strains . An important limitation of these studies was that they were done in wild-type mice whose mouse fH did not bind to the fHbp antigens . Therefore , the studies did not assess the possible effect of human fH on vaccine immunogenicity . In the present study we used human fH transgenic mice to investigate the immunogenicity of NOMV vaccines containing wild-type fHbp , or a mutant fHbp antigen that had no detectable binding of human fH .
Vaccine immunogenicity was evaluated in mice in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . The protocols were approved by the Institutional Animal Care and Use Committees at Children's Hospital Oakland Research Institute and the University of Massachusetts Medical School . Blood collection was performed under anesthesia , and all efforts were made to minimize suffering . The human complement source for measuring serum bactericidal activity was serum from an adult who participated in a protocol that was approved by Institutional Review Board ( IRB ) of Children's Hospital Oakland Research Institute . Written informed consent was obtained from the subject . The NOMV vaccines were prepared from mutants of group B vaccine strain H44/76 ( Table 1 ) . The parent strain was isolated from a patient with invasive disease during an epidemic in Norway , and was used to prepare a detergent-extracted OMV vaccine [24] , [25] . The wild-type H44/76 strain naturally expresses a high level of fHbp amino acid sequence variant ID 1 as classified in the public database of fHbp variants ( http://pubmlst . org/neisseria/fHbp/ ) . To attenuate endotoxin [26] , the LpxL1 gene was deleted as previously described [23] , which resulted in penta-acylated instead of hexa-acylated lipo-oligosaccharide ( LOS ) [26] , [27] . The resulting mutant strain ( H44/76ΔLpxL1 , Table 1 ) was then transformed with one of three plasmids: pBSΔfHbp::erm [8] , pBS-erm-fHbp wild-type ( WT ) , or pBS-erm-fHbp R41S . The pBS-erm-fHbp WT plasmid was constructed by subcloning a cassette containing a modified PorA promoter , in which the polyguanine tract between the −35 and −10 sites was replaced with the corresponding sequence from the NadA gene , followed by the fHbp ID 1 gene , downstream of the erm cassette of pBSΔfHbp::erm . The R41S substitution was introduced into WT fHbp as described previously [17] . To increase over-expression of fHbp , a second copy of the cassette containing the modified PorA promoter and the WT fHbp gene was subcloned into pFP12 using the SphI and StuI restriction sites as described previously [28] . The modified plasmid was used to transform the strain with WT fHbp integrated into the chromosome . Another version of this plasmid containing R41S mutant fHbp was used to insert a second copy into the strain with the R41S fHbp gene integrated into the chromosome . Characteristics of the three mutants used to prepare NOMV vaccines are summarized in Table 1 . Binding of human fH to the surface of live bacteria was determined by flow cytometric detection of fluorescence , which was performed as described previously [29] . Bacteria were grown to mid-exponential phase in Mueller-Hinton broth supplemented with 0 . 25% glucose and 0 . 02 mM cytidine monophosphate N-acetyl neuraminic acid ( CMP-NANA ) , harvested by centrifugation , and suspended in blocking buffer ( Dulbecco's PBS containing 1% ( w/v ) BSA ) at approximately 7 . 5×108 cfu/ml . Human fH ( 10 µg/ml ) or , as controls , anti-fHbp mAb ( 4 µg/ml ) , or anti-PorA mAb ( 1 µg/ml ) was incubated with the cells for 30 min at room temperature . The anti-fHbp mAb was JAR 5 [30] and the anti-PorA mAb was P1 . 7 ( National Institute for Biological Standards and Controls , Potters Bar , UK; NIBSC code: 01/514 ) [31] . To detect bound human fH , the cells were incubated with affinity purified goat anti-human fH ( 2 µg/ml ) , followed by AlexaFluor 488 rabbit anti-goat IgG H+L ( 1∶500; Invitrogen ) . To detect murine mAbs , the cells were incubated with AlexaFluor 488 goat anti-mouse IgG H+L ( 1∶500; Invitrogen ) . The preparations were fixed with 0 . 5% formaldehyde in PBS buffer , and the bacterial cells were analyzed by flow cytometry ( LSRFortessa , BD Biosciences ) . NOMV vaccines were prepared from membrane blebs released into bacterial culture supernatants as previously described [32] . To test binding of fH or mAbs by ELISA , the NOMV vaccines ( 2 µg/ml in PBS ) were adsorbed to the wells of a microtiter plate ( Immulon 2HB; Thermo Scientific ) by incubation overnight at 4°C . Non-specific binding was blocked with PBS containing 1% BSA for 1 h at room temperature . Different concentrations of purified human fH ( Complement Technology ) were added to the plate , and were incubated for 2 h at room temperature . After washing , bound fH was detected with goat anti-human fH ( affinity purified , 1 µg/ml ) followed by rabbit anti-goat IgG conjugated to alkaline phosphatase ( Sigma; 1∶5 , 000 ) . Murine mAb binding to the NOMV vaccines was tested by a similar procedure except that the mAbs ( anti-fHbp mAb JAR 5 or anti-PorA mAb P1 . 7 ) were added instead of fH , and bound mAbs were detected with goat anti-mouse IgG conjugated to alkaline phosphatase . We characterized the three NOMV preparations further by Western blotting to detect fHbp expression , and by SDS-PAGE . The proteins in the NOMV preparations were separated by SDS-PAGE ( 10% acrylamide NuPAGE; Invitrogen ) and stained with Coomassie blue ( SimplyBlue; Invitrogen ) . For the Western blot , the proteins were transferred to a polyvinylidene fluoride membrane ( Immobilon-FL; Millipore ) using an XCell II Blot Module ( Invitrogen ) . The membrane was blocked using PBS ( Roche ) containing 1% ( w/v ) blocking grade nonfat dry milk ( Bio-Rad ) , and then washed with PBS containing 0 . 1% Tween-20 ( Sigma ) . fHbp was detected using anti-fHbp mAb JAR 5 ( 0 . 1 µg/ml ) and goat anti-mouse IgG conjugated to IRDye 800CW ( Li-Cor Biosciences ) . The IRDye was detected at 800-nm wavelength on an infrared scanner ( Odyssey; Li-Cor Biosciences ) . Groups of sixteen to twenty female BALB/c mice , aged 5 weeks , were immunized with one of three NOMV vaccines ( fHbp knock-out ( KO ) ; over-expressed wild-type fHbp ( OE WT ) ; or over-expressed R41S mutant fHbp ( OE R41S ) ) . As described above , all three vaccines were prepared from mutants in which the LpxL1 gene had been inactivated to attenuate endotoxin activity [27] , Each NOMV vaccine dose contained 2 . 5 µg of protein and 600 µg of aluminum hydroxide ( Alhydrogel; Brenntag Biosector ) . Additional groups of control mice were immunized with 20 µg of recombinant fHbp ID 1 ( N = 10 mice ) or aluminum hydroxide adjuvant alone ( N = 7 mice ) . Two doses were given intraperitoneally at three-week intervals and blood was collected by cardiac puncture three weeks after the second dose . Details of the construction and characterization of human fH transgenic mice have been described [17] . In the present study , the number of available human fH transgenic BALB/c mice was limited , in part because we excluded mice with serum human fH concentrations <240 µg/ml as measured by a fHbp capture ELISA [17] . The rationale for exclusion was our previous data that suggested that the effect of human fH on decreasing fHbp immunogenicity was greatest in mice with high serum human fH concentrations ( >250 µg/ml , with the typical range of fH concentrations in sera from humans being between 200 to 400 µg/ml [17] ) . To maximize the sizes of the treatment groups of fH transgenic mice , we included both male and female animals , ages 2 to 4 months . The 32 available mice were randomized to one of four vaccine groups using the random number function in Microsoft Excel . Thirteen animals were assigned to receive the NOMV vaccine with over-expressed WT fHbp , and 13 animals to receive the NOMV vaccine with over-expressed R41S mutant fHbp . The remaining six mice served as controls ( three were immunized with aluminum hydroxide adjuvant alone and three were immunized with an NOMV vaccine from a fHbp knock-out mutant ) . For the immunogenicity study in human fH transgenic mice , three doses of NOMV vaccine ( 2 . 5 µg of protein per dose ) were given intraperitoneally at three-week intervals . Blood was collected by cardiac puncture three weeks after the third dose at which time the animals were euthanized . During the study , five mice were removed before completing the vaccination schedule ( three in the group immunized with the NOMV vaccine with over-expressed R41S fHbp , one in the group immunized with the NOMV vaccine from the fHbp KO , and one in the group immunized with aluminum hydroxide ) . These animals either were males injured from fighting and were euthanized to minimize pain or distress , or died for other reasons not related to the procedures . Sera from individual mice were tested for IgG anti-fHbp antibody titers by ELISA , which was performed as described previously [33] . Complement-mediated serum bactericidal activity was measured using washed , exponential growth-phase bacteria grown to OD620 = 0 . 6 in Mueller-Hinton broth supplemented with 0 . 25% glucose and 0 . 02 mM CMP-NANA . The group B test strain , Cu385 , expressed fHbp ID 1 that matched the amino acid sequence of fHbp ID 1 in the vaccines , and had a PorA variable region ( VR ) type heterologous to strain H44/76 , which was used to prepare the NOMV vaccines ( Table 1 ) . For measuring bactericidal activity , the mouse sera were heated at 56°C for 30 min to remove endogenous complement activity . The exogenous complement source was human serum from a donor lacking intrinsic bactericidal activity . To eliminate possible effects of naturally acquired non-bactericidal IgG antibodies in the human complement on the bactericidal titers of the mouse sera [34] , the human serum was depleted of IgG by passing one ml of serum over a protein G column ( HiTrap Protein G HP 1 ml; GE Healthcare ) as previously described [17] . The final 40-µl bactericidal reaction included different dilutions of the mouse sera , 20% ( v/v ) human complement and ∼103 cfu of the test strain . After one hour of incubation at 37°C in the presence of 5% CO2 , 12 µl of the reaction were plated . Titers were assigned by the interpolated dilution of test serum that gave 50% survival of the bacteria after 60 min incubation relative to CFU of negative controls at time 0 . Adsorption of serum anti-fHbp antibody was performed using recombinant fHbp , or recombinant human albumin as a mock treatment , each coupled separately to CnBr-activated Sepharose 4B ( Sigma ) , which was performed as described [34] . Two serum pools were prepared from wild-type mice immunized with the NOMV vaccine with over-expressed WT fHbp and two serum pools from mice immunized with the NOMV vaccine with over-expressed R41S mutant fHbp ( each pool contained sera from 9 to 10 mice ) . The pools were diluted 1∶5 and each pool was divided into three aliquots; one aliquot was incubated with the conjugated fHbp , the second with the conjugated human albumin ( mock control ) , and the third was retained as a non-adsorbed sample . After incubation with the conjugated Sepharose overnight at 4°C , the beads were transferred to disposable columns ( Bio-Rad ) . Three column washes were collected and concentrated to the original sample volume by ultrafiltration ( Spin-X UF 6 , 10K MWCO; Corning ) . By ELISA , the fHbp column removed >99% of the anti-fHbp antibodies ( see Results ) .
As described in the Methods , we prepared three meningococcal mutant vaccine strains for preparation of NOMV vaccines: one strain was a LpxL1 knock-out ( KO ) mutant with attenuated endotoxin that also had the fHbp gene inactivated ( negative control ) ; a second was an LpxL1 KO mutant engineered to over-express wild-type ( WT ) fHbp ID 1; and a third was an LpxL1 KO mutant with over-expressed fHbp ID 1 containing the R41S substitution to eliminate fH binding to fHbp ( Table 1 ) . As expected , the mutant strain with over-expressed WT fHbp bound fH ( Figure 1 , Panel A ) . In contrast , binding of fH to the mutant with over-expressed R41S fHbp was indistinguishable from that of the fHbp knock-out mutant ( Figure 1 , Panel A ) . The control anti-fHbp mAb JAR 5 showed similar binding to the strains with over-expressed WT or R41S fHbp , which showed that both of these mutants had similar levels of fHbp expression ( Figure 1 , Panel B ) . With the fHbp knock-out mutant , there was only background signal with the anti-fHbp mAb . An additional control anti-PorA mAb P1 . 7 bound equally well to all three strains ( Figure 1 , Panel C ) . By flow cytometry , the mutant vaccine strain with over-expressed wildtype fHbp had about 4-fold greater binding of anti-fHbp mAbs than that of the parent wild-type H44/76 strain , and 4-fold greater binding of human fH ( Supplemental Figure S1 , Panels A and B , respectively ) . We prepared NOMV vaccines from membrane blebs released into broth culture supernatants by each of the three mutant vaccine strains using methods previously described [32] . By ELISA , there was similar binding of the control anti-fHbp mAb , JAR 5 , with the NOMV vaccines from the mutants with over-expressed WT or R41S mutant fHbp , whereas there was only background binding with the NOMV vaccine from the fHbp knock-out mutant ( Figure 1 , Panel E ) . The NOMV vaccine from the mutant with WT fHbp bound ∼50 fold-more fH than the NOMV vaccine with mutant R41S fHbp ( Figure 1 , Panel D ) . Further , binding of fH to the NOMV vaccine with mutant R41S fHbp was indistinguishable from that of the NOMV vaccine from the fHbp knock-out mutant . The residual fH binding to the NOMV vaccines prepared from the mutants with over-expressed R41S fHbp or the fHbp knock-out likely was mediated by Neisserial surface protein A ( NspA ) , which recently was identified is a second meningococcal ligand for fH [35] . The NOMV vaccines from the two mutants with over-expressed WT fHbp or fHbp knock-out had similar binding with the control anti-PorA mAb ( Figure 1 , Panel F ) . With the NOMV from the third mutant with fHbp R41S , there was slightly lower binding with the control anti-PorA mAb ( Figure 1 , Panel F ) . By Western blotting ( Figure 2 , Panel A ) , the NOMV vaccine from the fHbp knock-out strain did not contain detectable fHbp ( lane 3 ) , whereas the NOMV vaccines from the mutants with over-expressed WT or R41S fHbp contained similar amounts of fHbp ( lanes 4 and 5 , respectively ) . Although not shown in Figure 2 , the amount of over-expressed fHbp in these NOMV vaccines was approximately 5-fold higher than that of an NOMV vaccine prepared in a previous study from the parent ( wild-type ) H44/76 strain [20] . The H44/76 strain naturally expressed relatively high amounts of fHbp [36] . By SDS-PAGE with Coomassie blue staining , the respective protein profiles of the three NOMV vaccines were similar ( Figure 2 , Panel B ) . The only exception was a band with a mass of ∼25 kDa , which was present in the NOMV from the fHbp KO mutant ( lane 3 ) but not in the two other NOMV vaccines with over-expressed fHbp ( lanes 4 and 5 ) . This band was not identified . Note also , that a 28 kDa protein co-migrated with fHbp since SDS-PAGE the 28 kDa band was present in the of the NOMV vaccine from the fHbp knock-out mutant ( Panel B , lane 3 ) but not detected by Western blot ( Panel A , lane 3 ) . In previous experiments , a band resolving in this portion of SDS-PAGE of NOMV preparations from wild-type and mutant strains of H44/76 with over-expressed fHbp was identified by mass spectrometry to contain both fHbp and OpcA [20] . We immunized wild-type BALB/c mice with two doses of each of the NOMV vaccines . Control mice received two doses of the recombinant fHbp ID 1 vaccine adsorbed with aluminum hydroxide , or aluminum adjuvant alone . Since mouse fH does not bind to fHbp [16] , the antibody responses provided information on vaccine immunogenicity in an animal model in which mouse fH did not bind to fHbp in any of the vaccines . Although three doses of the vaccines likely would have been more immunogenic than two doses , we chose to use two doses as a more sensitive indicator of possible loss of fHbp immunogenicity from the amino acid substitution introduced to eliminate fH binding . Both of the NOMV vaccines with over-expressed WT or mutant R41S fHbp elicited higher serum IgG anti-fHbp titers than mice immunized with the control recombinant WT fHbp vaccine ( P≤0 . 0002 , Figure 3 , Panel A ) . However , the mice immunized with the NOMV vaccine with mutant fHbp had a 2-fold lower IgG anti-fHbp geometric mean titer ( GMT ) than the mice immunized with the NOMV vaccine with WT fHbp ( P = 0 . 001 , Figure 3 , Panel A ) . The NOMV vaccines with over-expressed WT or mutant R41S fHbp elicited higher serum bactericidal responses than to the recombinant fHbp vaccine . The low responses to the recombinant protein may have reflected the use of a suboptimal two-dose schedule . The mice immunized with the NOMV vaccine with over-expressed mutant fHbp had a 4-fold lower bactericidal GMT than the mice immunized with the NOMV vaccine with over-expressed WT fHbp ( P = 0 . 003 , Figure 3 , Panel B ) . Serum bactericidal activity was not detectable in mice immunized with the NOMV vaccine prepared from the fHbp knock-out mutant ( titers <1∶10 ) . The Cu385 strain used to measure bactericidal activity had an identical fHbp sequence but a different PorA VR type than the strains used to prepare the NOMV vaccines ( Table 1 ) . Thus , the high bactericidal responses elicited by the NOMV vaccines containing fHbp suggested that the target antigen was fHbp . To determine the contribution of anti-fHbp antibodies to serum bactericidal activity , we depleted anti-fHbp antibodies in serum pools from mice vaccinated with the NOMV vaccines with over-expressed WT or R41S mutant fHbp by adsorption to immobilized recombinant WT fHbp ( see Methods ) . By ELISA , the fHbp column removed greater than 99% of the anti-fHbp antibodies , compared with a negative control ( mock ) column containing immobilized human albumin ( Figure 4 , Panel A ) . The fHbp adsorption procedure did not affect significantly the anti-NOMV ELISA titer using NOMV from the fHbp KO mutant as the antigen adsorbed to the wells ( Figure 4 , Panel B ) . Depletion of the serum anti-fHbp antibodies resulted in a nearly 100-fold decrease in bactericidal titers against strain Cu385 with fHbp from variant group 1 ( Figure 4 , Panel C ) . As a control , we tested bactericidal activity against a second strain , which was susceptible to anti-PorA bactericidal activity ( a mutant of group B strain H44/76 in which fHbp ID 28 in variant group 3 had been substituted for fHbp ID 1 in variant group 1; Table 1 ) [37] . The adsorption of the serum anti-fHbp antibody had no effect on the titers against this second strain ( Figure 4 , Panel D ) . Thus , the serum bactericidal antibodies against strain Cu385 were directed largely against fHbp . Collectively , the immunogenicity results from the studies in wild-type mice showed that the NOMV vaccine with over-expressed mutant fHbp elicited lower serum anti-fHbp responses than the NOMV vaccine with over-expressed WT fHbp . Thus , in wild-type mice whose mouse fH did not bind to fHbp in either NOMV vaccine , the R41S substitution diminished fHbp immunogenicity . In humans , fH would be expected to bind to the WT fHbp vaccine antigen . To investigate the effect of human fH , we measured the immunogenicity of the NOMV vaccines in human fH transgenic mice . The respective ranges of the serum human fH concentrations of the mice in the different vaccine groups , which were measured in sera obtained before immunization , were similar ( Figure 5 ) , as were the respective mean ages and gender distributions of the groups . We immunized the transgenic mice with three doses of each of the NOMV vaccines . The group immunized with the NOMV vaccine with over-expressed mutant R41S fHbp had a 5-fold higher serum IgG anti-fHbp GMT ( P = 0 . 002 ) , and a 19-fold higher serum bactericidal GMT ( P = 0 . 001 ) , than the group immunized with the NOMV vaccine with over-expressed WT fHbp ( Figure 6 , Panels A and B , respectively ) . There was no correlation between the magnitude of the serum concentrations of human fH in the individual animals and the serum bactericidal titers ( r = 0 . 15 , P = 0 . 62 ) . Thus , in human fH transgenic mice , there was superior immunogenicity of the NOMV vaccine with the R41S mutant fHbp , which was opposite to the results obtained in the wild-type mice .
In previous studies in wild-type mice , NOMV vaccines prepared from mutant meningococcal strains with attenuated endotoxin and over-expressed fHbp elicited broader serum bactericidal antibody responses than recombinant fHbp vaccines [18] , [20] , [21] , [23] . The higher activity was particularly evident against strains with lower fHbp expression [18] . Although there are conflicting data [38] , by adsorption most of the serum bactericidal antibody elicited by the NOMV vaccines against test strains with heterologous PorA VR sequence types was directed at fHbp [20] , [21] , [23] , [28] . The underlying mechanism for the broader anti-fHbp bactericidal activity is unknown although one study suggested that the IgG anti-fHbp antibodies elicited by the NOMV vaccine had greater ability to inhibit binding of fH to fHbp than antibodies elicited by a recombinant fHbp vaccine [18] . These results suggested that the serum anti-fHbp antibody repertoire to the NOMV vaccine better targeted the fH-binding surface , as compared with other sites on the fHbp molecule when recombinant fHbp vaccines were used [18] . In the present study , we investigated the immunogenicity of NOMV vaccines prepared from mutants with over-expressed WT fHbp that bound human fH , or over-expressed R41S mutant fHbp that did not bind human fH . Introduction of the R41S mutation resulted in decreased fHbp immunogenicity in wild-type mice whose mouse fH did not bind to fHbp ( Figure 3 ) . In contrast , the NOMV vaccine with the mutant fHbp , which did not bind human fH , elicited nearly 20-fold higher serum bactericidal antibody responses than the control NOMV with wild-type fHbp that bound fH ( Figure 6 ) . Thus , the modest loss of fHbp immunogenicity , which was evident in the WT mice whose mouse fH did not bind to either vaccine was more than compensated for in the transgenic mice by the larger effect of human fH on decreasing immunogenicity of the NOMV vaccine containing fHbp that bound human fH . These results extend our earlier findings in human fH transgenic mice , which demonstrated an adverse effect of human fH on immunogenicity of a recombinant fHbp vaccine that bound human fH [17] . In the present study we found an even larger effect of human fH on lowering immunogenicity of the NOMV vaccine with the wild-type fHbp but , in contrast to our earlier study [17] , we did not find a significant correlation between immunogenicity and serum human fH levels . The greater effect of human fH on lowering immunogenicity in the present study , and the lack of an inverse correlation with the serum human fH levels , may be explained by our exclusion of transgenic mice with low serum concentrations of human fH ( <240 µg/ml ) , which was not done in our previous study . Two lines of evidence indicated that the serum bactericidal activity measured in the present study against strain Cu385 was directed against fHbp . First , this strain was resistant to serum bactericidal antibodies elicited by the control NOMV vaccine from the fHbp knock-out strain ( Figure 3 , Panel B ) . Second , depletion of anti-fHbp antibodies from serum pools from wild-type mice immunized with NOMV vaccines that contained WT or R41S mutant fHbp resulted in ≥98% loss of bactericidal activity ( Figure 4 , Panel C ) . Similar adsorption studies were not feasible with the sera from the human fH transgenic mice because of insufficient volumes from the fewer animals in each vaccine group than used in the studies of the WT mice . We expect , however , that the serum bactericidal responses of the transgenic mice against strain Cu385 also depended on the presence of antibodies to fHbp . The recombinant R41S mutant fHbp vaccine used in our previous study was slightly less immunogenic in wild-type mice than the recombinant WT fHbp vaccine , although the difference was not statistically significant [17] . In the present study , we immunized larger groups of wild-type mice and detected statistically significant 2-fold lower IgG anti-fHbp titers and 4-fold lower bactericidal titers elicited by the NOMV vaccine with over-expressed R41S mutant fHbp , compared with NOMV vaccine with WT fHbp . Thus , substitution of serine for arginine at residue 41 in fHbp ID 1 , which resulted in no detectable binding of human fH , had an adverse effect on fHbp immunogenicity in mice lacking human fH . Potentially , alternative mutant fHbp molecules can be identified that eliminate fH binding without this loss of immunogenicity and that these vaccines will be even more effective than the R41S mutant fHbp vaccine . Vaccines containing fHbp molecules that bind human fH are reported to elicit serum bactericidal antibodies in infants and adults and , therefore , are likely to confer protection against disease [10] , [11] . Next-generation , recombinant chimeric fHbp molecules , which combine domains or regions from different variants , offer the prospect of extending protection against strains with fHbp from different variant groups [39] , [40] . NOMV vaccines from mutant strains with over-expressed fHbp also can be used to increase breadth of protection [19] , [21] . The present and previously published data from studies in human fH transgenic mice [17] indicate that the immunogenicity of fHbp antigens in humans also can be improved by incorporating mutations that decrease or eliminate fH binding . Finally , a number of other pathogenic microbes including N . gonorrhoeae , Borrelia burgdorferi , Francisella tularensis , Haemophilus influenzae , Streptococcus pneumoniae , S . pyogenes and Yersinia enterocolitica are reported to bind complement factor H to their surface ( Reviewed in [41] ) . Although the amino acid sequences of the fH ligands of these pathogens are not homologous to fHbp , genetic approaches similar to those employed in the present study could be employed to eliminate fH binding of these antigens and potentially increase their immunogenicity in humans . The results of testiing fHbp immunogenicity in human fH transgenic mice highlight the potential value of this mouse model for testing immunogenicity of these vaccines . | Vaccines containing factor H-binding protein ( fHbp ) are being developed for protection against bacterial meningitis and sepsis caused by meningococci . The antigen was identified from genomic sequences and only later found to bind a human complement protein , factor H ( fH ) , but not fH from non-human species . In previous studies , native outer membrane vesicle ( NOMV ) vaccines from mutants with over-expressed fHbp elicited broadly protective serum antibodies in mice whose fH did not bind to fHbp in the vaccine . In this study , the authors immunized transgenic mice and showed that the presence of human fH decreased serum bactericidal antibody responses to a NOMV vaccine with fHbp that bound human fH . In contrast , a NOMV vaccine containing fHbp with a single amino acid substitution that eliminated fH binding elicited nearly twenty-fold higher protective antibody responses . Thus , a simple change in a vaccine antigen to eliminate binding to a host protein can increase immunogenicity . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"vaccines",
"medicine",
"vaccination",
"complement",
"system",
"clinical",
"immunology",
"immunity",
"immunology",
"biology",
"microbiology",
"immune",
"system"
] | 2012 | The Effect of Human Factor H on Immunogenicity of Meningococcal Native Outer Membrane Vesicle Vaccines with Over-Expressed Factor H Binding Protein |
Circadian disruption has multiple pathological consequences , but the underlying mechanisms are largely unknown . To address such mechanisms , we subjected transformed cultured cells to chronic circadian desynchrony ( CCD ) , mimicking a chronic jet-lag scheme , and assayed a range of cellular functions . The results indicated a specific circadian clock–dependent increase in cell proliferation . Transcriptome analysis revealed up-regulation of G1/S phase transition genes ( myelocytomatosis oncogene cellular homolog [Myc] , cyclin D1/3 , chromatin licensing and DNA replication factor 1 [Cdt1] ) , concomitant with increased phosphorylation of the retinoblastoma ( RB ) protein by cyclin-dependent kinase ( CDK ) 4/6 and increased G1-S progression . Phospho-RB ( Ser807/811 ) was found to oscillate in a circadian fashion and exhibit phase-shifted rhythms in circadian desynchronized cells . Consistent with circadian regulation , a CDK4/6 inhibitor approved for cancer treatment reduced growth of cultured cells and mouse tumors in a time-of-day–specific manner . Our study identifies a mechanism that underlies effects of circadian disruption on tumor growth and underscores the use of treatment timed to endogenous circadian rhythms .
In response to day–night cycles produced by earth's 24-hour rotation around its axis , almost all living organisms have evolved circadian clocks , endogenous timekeeping systems that adapt physiology to daily changes in the environment . In mammals , the circadian timing system consists of a central light-entrained clock in the suprachiasmatic nucleus ( SCN ) of the brain and numerous peripheral clocks located in most body organs , all of which are typically coordinated to constrain sleep and feeding , metabolism , and immune functions to appropriate times of the day . The basic timekeeping unit is the cell in that even single neurons and fibroblasts harbor a conserved , cell-autonomous circadian clock . At the molecular level , the basic mechanism consists of transcription-translation feedback loops , with the major loop in mammals comprised of the Period ( Per ) and Cryptochrome ( Cry ) genes , which are rhythmically transcribed by circadian locomoter output cycles protein kaput ( CLOCK ) —brain and muscle Arnt-like protein-1 ( BMAL1 ) transcription factors and repressed by their own protein products ( PER , CRY ) . With further fine-tuning at transcriptional , posttranscriptional , and translational levels , the intrinsic molecular oscillator integrates multiple external signals to regulate expression of clock-controlled genes , which differ dramatically from tissue to tissue despite usage of largely the same core clock genes across the organism [1] . Consistent with the adaptive physiological and cellular benefits of the circadian timing system , accumulating evidence indicates that disruption of circadian homeostasis by genetic alteration or irregular lifestyle has pathological consequences [2] . For instance , epidemiological studies show that frequent misalignment of circadian rhythms caused by lifestyle factors such as chronic sleep deprivation , jet lag , or shift work is a potential risk factor for cancer [3] . Even in experimental animal studies , genetic or environmental disruption of circadian rhythms highly increases the incidence or growth rate of various types of tumors , including lung , breast , skin , oral , and prostate cancers [4–6] . The wide variety of tumors affected suggests that circadian disruption alters basic cellular physiology in a fundamental way that increases susceptibility to diseases like cancer; however , the underlying mechanism is not understood . Increased tumorigenesis and accelerated tumor growth are often linked to a dysregulated cell cycle [7 , 8] . The cell cycle consists of cell growth ( G1 ) , DNA replication ( S ) , and cell division ( G2/M ) phases . Generally , control of cell proliferation occurs at the G1 phase , during which cells integrate multiple signals , ranging from growth factors to DNA damage , to determine entry into the S phase or exit to the G0 non-cycling quiescent phase [9] . Most human cancers are thought to develop from disruption of G1/S cell cycle control [10] . The G1/S phase boundary is regulated by the retinoblastoma ( RB ) tumor suppressor protein , which is sequentially phosphorylated by cyclin D1/cyclin dependent kinase ( CDK ) 4/6 in early G1 and by cyclin E/CDK2 in late G1 [11] . Links between cell cycle regulators and the circadian clock have been identified [12 , 13] , but how the circadian clock regulates cell cycle transitions is unclear , as is the mechanism by which circadian disruption affects tumor growth . In addition to the relevance of circadian disruption for the development of disease , circadian rhythms can be quite important for the treatment of cancer and other disorders . In recent years , some studies have investigated the idea of exploiting the circadian clock in tumors for therapy , for instance , by modulating activity of circadian clock molecules [14] , enhancing intra-tumor circadian rhythms [15] , and optimizing anticancer drug delivery by timing it to the host’s circadian rhythms [16 , 17] . Delivery timed to the appropriate time of day ( chronotherapy ) is expected to have better efficacy and possibly reduce the dose required , thereby limiting side effects , but the practice of chronotherapy is still in its infancy and would be facilitated by mechanistic insights into the temporal regulation of drug action . In the present study , we sought to determine , on a cellular and molecular level , how circadian disruption affects basic cell function . Starting with cell-based circadian desynchronization experiments , we conducted extensive molecular , cellular , and biochemical analyses to reveal that chronic circadian disturbance promotes pro-proliferative signaling events , which converge to stimulate G1/S phase progression via CDK4/6-dependent RB phosphorylation . We also found that RB phosphorylation , particularly at S807/S811 sites , cycles in phase with rhythmic cyclin D1 gene expression , and it is increased and altered in fluctuation by chronic circadian perturbation . Using a mouse model , we show that circadian disruption also increases the growth of carcinogen- or melanoma-induced tumors in mice . Consistent with a role for RB-CDK4/6 , we find that PD0332991 , a selective CDK4/6 inhibitor , has time-of-day–specific effects on proliferation in cells and mouse tumors , but this time dependence is abrogated by a chronic jet-lag protocol . Together , our findings indicate that the circadian clock regulates G1/S phase progression via the cyclin D1-CDK4/6–RB pathway , and provide a mechanism for the application of chronotherapeutic approaches to cancer patients .
Experimental disruptions of the circadian clock , such as chronic jet lag in mice and rats , have detrimental effects on health , causing aberrant behavioral and hormonal rhythms , cognitive deficits , weight gain , diabetes , apoptosis , and accelerated tumor development and growth [18] . However , molecular , biochemical , or cellular mechanisms underlying these effects of circadian disturbance remain ambiguous . In order to determine consequences of circadian disruption on the most basic level , we sought to develop a cell-based , chronic circadian desynchrony ( CCD ) assay mimicking the experimental jet-lag protocol used in mouse models [4 , 19] . To test the feasibility of an in vitro CCD approach , we chose a commonly used circadian model [20] , human U2 osteosarcoma ( U2OS ) cells expressing a Period2 ( Per2 ) promoter-driven destabilized luciferase reporter ( pPer2-dLuc ) . This reporter cell line enables real-time measurement of cellular rhythms after synchronization with dexamethasone ( dex ) , a synthetic glucocorticoid ( GC ) and potent clock synchronizer widely used in cell culture models ranging from fibroblasts to astrocytes [21–23] . First , to assess the effect of CCD on cellular rhythmicity , we exposed control cells to 10 days of a regular change of dex ( 100 nM ) -containing media at 24-hour intervals ( Control; CTL ) , while experimental cells received serial 8-hour advances of the dex treatment cycle every 2 days ( Jet lag; JL ) ( Fig 1A ) . After the CCD schedule , dex synchronization was done on a weekly basis to prevent decoupling of circadian phases across cells , and real-time activity of the Per2 promoter was recorded in both CTL and JL cells ( Fig 1B ) . Analysis of bioluminescence rhythms during the first week ( 1stwk ) post CCD revealed significantly dampened rhythms in JL cells , with lower amplitude and , notably , 6 . 8-hour delayed onset of rhythms ( acro-phase ) ( Fig 1C , 1D and 1E ) . However , rhythms in JL cells were not different from those of CTL cells in subsequent weeks ( 2ndwk , 3rdwk ) , throughout which all cells showed nearly identical rhythmic cycles in luminescence ( Fig 1B , 1C and 1D ) . These effects are similar to those that occur with travel across different time zones , when rhythms are affected in the first week following travel but subsequently adapt to the new , stable environment [24] . A phase delay ( 8 . 2 hours ) and reduced amplitude of JL cell rhythms were also observed in the first week after a short-term circadian desynchronization schedule of only 6 days ( S1 Fig ) . The observations above concur with a previous mouse study showing that temporal patterns of circadian clock genes ( Per1 , Per2 , Cry1 ) in chronically jet-lagged mice exhibit lower amplitude as well as phase delays of 5 . 5–9 . 0 hours and 7 . 0–11 . 2 hours in the SCN and liver , respectively [19] . Moreover , in a recent human study , a similar set of circadian genes ( PER1 , PER2 ) as well as hormones ( plasma cortisol , melatonin ) showed between a 7- and 9-hour phase delay with decreased amplitude in human blood samples after a night shift protocol [25] . Taking this evidence together , we conclude that our cultured cell-based CCD strategy closely mimics a physiological jet-lag model . To determine the effects of CCD on cellular function and physiology , we conducted a wide range of assays that tested for oxidative stress/senescence ( H2O2 , glutathione [GSH]/glutathione disulfide [GSSG] ) , metabolism ( NADP/NADPH ) , proteolysis , and apoptosis , but we did not find any significant differences between CTL and JL cells ( S2 Fig ) . To take a more unbiased approach , we investigated genome-wide changes of rhythmic gene expression following the CCD schedule . To this end , we performed RNA sequencing ( RNA-Seq ) analysis of CTL and JL cell samples over a 24-hour cycle collected every six hours ( Fig 1A; black arrowheads ) . MetaCycle [26] analysis revealed significant circadian cycles of expression ( q < 0 . 1 ) for 44 transcripts in CTL cells ( Fig 1F and 1G , and S1 Table ) . Interestingly , in JL cells , the transcriptional rhythms of these cyclers were severely compromised or lost ( Fig 1F and 1G , and S2 Table ) . The expression of several known circadian genes also showed dampened or phase-altered oscillations in JL cells ( Fig 1H ) . Differential expression ( DE ) analysis comparing all CTL samples to all JL samples revealed many genes significantly up-regulated at all circadian times in JL cells ( S3 Table ) . Interestingly , many genes up-regulated in JL cells function in disease pathways implicated in neurodegeneration and cancer: SRC homology domain-containing protein 21 ( SH3D21 ) ( ataxia-telangiectasia and colorectal cancer [27] ) , exocyst complex component 3-like protein 2 ( EXOC3L2 ) ( Alzheimer disease [28] ) , DEXH ( Asp-Glu-X-His ) box polypeptide 58 ( DHX58 ) ( mammary tumor [29] ) , proprotein convertase subtilisin/kexin type 1 inhibitor ( PCSK1N ) ( skin carcinogenesis , Alzheimer disease and parkinsonism-dementia [30] ) , and jagged 2 ( JAG2 ) ( myeloma [31] ) ( Fig 1I and S3 Table ) . These observations indicate that in vitro chronic circadian disturbance significantly perturbs cellular rhythmicity as well as disease-relevant pathways . Pathway analysis of DE genes ( CTL versus JL samples; q < 0 . 4 ) found enrichment of gene networks involved in cellular growth , proliferation , and cancer ( Fig 2A , 2B and 2C ) . Prompted by this , we compared gene expression patterns between CTL and JL samples for well-characterized oncogenes and tumor suppressor genes . We found that several oncogenes ( 56/95 ) were up-regulated in JL cells , with significant induction noted for 11 genes ( e . g . , musculoaponeurotic fibrosarcoma oncogene B [MAFB] , Jun proto-oncogene related gene d [JUND] , musculoaponeurotic fibrosarcoma oncogene B [MAFA] , oligodendrocyte transcription factor 2 [OLIG2] , nuclear factor kappa B subunit 2 [NFKB2] , jun oncogene [JUN] , cyclin D3 [CCND3] , high mobility group AT-hook 1 [HMGA1] , erbb2 receptor tyrosine kinase 2 [ERBB2] , myelocytomatosis oncogene cellular homolog [MYC] , alpha serine/threonine kinase 1 [AKT1]; p < 0 . 05 ) . On the other hand , tumor suppressor genes ( 58/66 ) were largely unaffected or significantly down-regulated ( e . g . , annexin A1 [ANXA1] , neurofibromin 1 [NF1] , cluster of differentiation 36 [CD36] , breast cancer type 1 susceptibility protein [BRCA1]; p < 0 . 05 ) ( Fig 2D and S4 Table ) . This gene expression pattern is more clearly visible when plotting the log2 fold-change values for oncogenes and tumor suppressor genes ( Fig 2E ) . Overall expression of oncogenic retinoic acid syndrome ( RAS ) signaling factors as well as receptors and ligands of several growth factor signaling pathways ( epidermal growth factor [EGF]/fibroblast growth factor [FGF]/vascular endothelial growth factor [VEGF] ) , which contribute to cancer progression , was also induced in response to CCD ( Fig 2F–2I; S5 , S6 , S7 and S8 Tables ) . These data indicate that CCD reprograms the transcriptional landscape of multiple tumor progression or suppression pathways to facilitate cell proliferation and survival . Consistent with the transcriptome analysis , a trypan blue exclusion assay after CCD revealed a significant increase in cell number in JL samples compared with CTL ( Fig 3A ) . Importantly , the effect of CCD on molecular cycles and cell number did not depend on the use of dex , as similar data were obtained when forskolin was used as a synchronizing agent ( S3 Fig ) . To determine whether CCD-induced increased cell number was mediated by the molecular clock , we depleted the core clock activator ( BMAL1 ) or repressors ( CRY1/CRY2 ) using transient RNA interference ( RNAi ) in CTL and JL cells 48 hours before CCD ( Fig 3B–3G ) . Notably , the knockdown of either set of core clock genes significantly compromised the cellular response to CCD ( Fig 3B and 3C ) . Similarly , cell viability , as measured with a thiazolyl blue tetrazolium bromide ( MTT ) -based cell viability assay , was increased following CCD in control cells stably expressing siRNA targetting green fluoresecent protein ( si-GFP ) cells but not in U2OS cells stably expressing small interfering RNA ( siRNA ) targeting BMAL1 ( si-BMAL1 ) ( Fig 3D and 3E ) . Consistent with these results , a bromodeoxyuridine ( BrdU ) incorporation assay during the first week following CCD revealed a significantly higher number of proliferating cells in JL samples relative to CTL samples in the presence of control siRNA ( si-CTL ) , but the relative effect was reduced by knockdown of a core clock regulator ( si-BMAL1 , si-CRY1/2 ) ( Fig 3F and 3G ) . Although BMAL1 can have non-circadian effects , the reduced effect of CCD on cell number with two different clock knockdowns that have opposing roles within the clock ( BMAL1 and CRY ) supports the idea that the core clockwork contributes to the effects of CCD on cell proliferation . To explore the distribution of the cell cycle following CCD , we utilized fluorescence ubiquitination-based cell-cycle indicator ( FUCCI ) imaging , which uses fluorescent probes to demarcate stages of the cell cycle ( Cdt1-RFP for G1/S and Geminin-GFP for G2/M ) . We transduced these probes into the cells during the last day of the CCD schedule and found that both FUCCI sensor–positive cells were significantly increased by the CCD/JL protocol ( Fig 3H ) . However , the number of Cdt1-RFP–expressing cells was significantly higher than that of Geminin-GFP–expressing cells ( Fig 3I ) . Besides the CCD-induced changes in protein levels of the FUCCI probes , mRNA levels of CDT1 were also higher in the JL cells , relative to CTL , across multiple circadian times , whereas mRNA levels of Geminin ( GMMN ) , a posttranslational inhibitor of CDT1 , showed little change ( Fig 3J ) . The proto-oncogenic effect of CDT1 involves its up-regulation and hyperactivity in DNA replication in malignant tumor cells or tissues [32 , 33] . Thus , we speculated that activation of specific cell cycle genes contributed to the enhanced proliferation in response to CCD . Indeed , our transcriptomic analysis showed that genes involved in the G1/S phase transition ( CDK3 , MYC , CCND3 , CCND1 , and CDT1 ) were highly induced upon CCD , more so than genes that function in other parts of the cell cycle ( S , G2 , M ) ( Fig 3K , S4 Fig and S9 Table ) . These results together suggest that chronic disturbance of cellular rhythms influences the transcriptional network to preferentially activate G1/S phase progression and thereby enhance cell proliferation . In addition to transcript levels , the G1 to S phase transition is tightly regulated by phosphorylation of a number of cell cycle regulatory proteins [11] . In fact , a key event for G1/S cell cycle progression is phosphorylation of RB by cyclin D-CDK4/6 , which leads to activation of the G1/S transition-driving E2 transcription factor ( E2F ) ( S5A and S5B Fig ) . Several phosphorylation site mapping studies have reported that RB undergoes extensive phosphorylation during the G1/S cell cycle progression [34] ( S5A Fig ) . We sought to examine how CCD affects the phosphorylation status of RB using available phosphorylation site-specific antibodies ( pRB-S807/S811 , pRB-S795 , pRB-S780 , pRB-S612; S5A Fig ) . Western blot analysis using phospho-RB antibodies revealed that CCD induced significantly higher phosphorylation levels at many sites ( S807/S811 , S795 , S780 ) in RB ( S5C and S5D Fig ) . However , phosphorylation at S612 was not significantly increased . Interestingly , total RB protein levels were also significantly increased in JL cells compared with CTL cells despite similar levels of RB mRNA ( RB1 ) in both cell samples ( S5E Fig ) . Thus , the elevated RB protein level may result from increased protein stability mediated by CCD-induced phosphorylation . Phosphorylation of specific residues of RB at the carboxy terminus by different G1 Cyclin-CDK complexes contributes to its differential function in cell cycle progression and proliferation [34–36] . In this regard , we observed relatively higher phosphorylation at positions S807/S811 , compared with the other phosphosites , in response to CCD ( S5C Fig ) . To determine if the regulation of RB phosphosites in U2OS cells is consistent with previous reports in other systems , we used expression constructs encoding wild-type RB , non-phosphorylatable RB with 15 putative CDK sites converted to Ala residues ( RB-ΔCDK ) , and single CDK site monophosphorylatable RB proteins ( RB-ΔCDK+S612 , RB-ΔCDK+S780 , RB-ΔCDK+S807 , RB-ΔCDK+S811 ) ( S6 Fig ) . The expression of the wild-type and phosphosite–modified RB proteins was confirmed by western blot ( WB ) using phospho-specific antibodies ( S6B Fig ) . FUCCI cell cycle analysis of cells expressing each of these RB constructs revealed that either or both of the S807 and S811-monophosphorylated RB proteins prominently increased G1/S phase cells relative to wild-type , non-phosphorylated , and other single CDK site RB proteins ( S6A and S6C Fig ) . Interestingly , RB lacking all phosphorylation sites ( RB-ΔCDK ) resulted in more G1/S phase cells than wild-type RB , suggesting that some phosphorylation sites inhibit this transition . In parallel , MTT cell viability analysis also showed significantly enhanced cell proliferation in cells expressing the S807 and S811 monophosphorylated RB proteins alone or in combination ( S6D Fig ) . Indeed , phosphorylation of S807/S811 , which alleviates the suppressor function of RB , has been proposed as a representative marker for G1 progression through S phase [37–39] . Moreover , phosphorylation of RB at S807 protects against apoptotic cell death by regulating B-cell lymphoma 2 ( Bcl2 ) -associated X ( Bax ) activity [40] . To directly assess whether phosphorylation of RB of S807/S811 contributes to cell proliferation upon circadian disturbance , we used C33A cervix carcinoma cells that express a truncated unstable RB protein and so are considered null for RB [41] . We expressed wild-type RB or alanine mutants S807A and S811A of RB ( RB-S807A , RB-S811A ) in C33A cells and measured cell density with and without CCD ( S7A and S7B Fig ) . As expected , when a tumor suppressor is removed , RB-null cells showed high cell numbers and a very robust response to CCD . C33A cells rescued with wild-type RB showed a smaller but still significant response , but effects of CCD were blunted in C33A cells expressing RB-S807A or RB-S811A ( S7C Fig ) . Together , these data suggest that phosphorylation at S807/S811 plays a unique role in enhancing cell proliferation upon CCD and so likely contributes to the cell proliferation phenotype caused by chronic circadian disturbance . Multiple phosphoproteins that function in cell signaling , metabolic , and cell cycle pathways exhibit 24-hour oscillations [42 , 43] . To determine if RB phosphorylation was affected by CCD in a time-of-day–specific manner , we exposed cells to a CCD schedule and found that phosphorylation at targeted sites of RB , with the possible exception of S612 , exhibited a consistent increase across circadian time in JL cells compared with CTL cells ( Fig 4A ) . Notably , phosphorylation at S807/S811 was found to cycle in CTL cells , but the circadian phase was reversed in JL cells ( Fig 4B ) . To understand the mechanism underlying the effects of CCD on RB phosphorylation , we examined the expression of cell cycle genes under CCD . Cyclin D1 ( CCND1 ) protein expression displayed increased expression and a reversed cyclic pattern in response to CCD , while other cell cycle molecules ( CDK4 , cyclin B1 [CCNB1] ) were relatively unchanged ( Fig 4A and 4B ) . mRNA expression of cyclin D1 was also persistently induced over most circadian time points in JL cells relative to CTL cells , but this was not seen for other RB pathway components ( RB1 , CDK4 , CDK6 ) ( S8A Fig ) . These results point to a principal role of cyclin D1 in promoting CDK4/6 kinase activity to drive increased expression and oscillation of RB phosphorylation in response to CCD . The cyclin D1 gene responds directly to and integrates multiple signaling pathways via its various enhancer elements during the G1/S transition [9] . To address the mechanism responsible for CCD-driven cyclin D1 gene activation , we examined the RNA-Seq data for differences between CTL and JL samples with respect to signaling pathways known to regulate cyclin D1 expression ( S8B–S8F Fig , and S10 , S11 , S12 , S13 and S14 Tables ) . Expression of a large number of key signaling molecules that activate cyclin D1 ( wingless/integrated [Wnt] , extracellular signal-regulated kinase/mitogen activated protein kinase [ERK/MAPK] , phosphatidylinositol 3-kinase/alpha serine/threonine-protein kinase [PI13K/AKT] , hippo signaling pathway [HIPPO] , G protein-coupled receptor [GPCR] ) was up-regulated , whereas most genes that inhibit cyclin D1 were relatively unaffected or down-regulated in response to CCD ( S8B–S8F Fig , and S10 , S11 , S12 , S13 and S14 Tables ) . Correspondingly , several transcriptional activators that respond to such upstream signals and directly target enhancer elements of the cyclin D1 promoter ( activator protein 1 [AP-1] , nuclear factor kappa B [NFKB] , notch signaling pathway [NOTCH] ) exhibited up-regulated gene expression relative to repressors , as a result of CCD ( Fig 4C , and S15 Table ) . To verify CCD-induced activation of cyclin D1 in vivo , we induced a fibrosarcoma in mice using methylcholanthrene ( MCA ) , a potent chemical carcinogen , and exposed the mice to a chronic jet-lag schedule [4] . As described below , tumor growth showed a notably higher rate in jet-lagged mice compared with controls . Western blot analysis revealed that cyclin D1 protein expression as well as RB phosphorylation were markedly increased at multiple time points in tumor tissue , but not in liver tissue , relative to control mice samples assayed in parallel ( Fig 4D ) . This is consistent with previous reports of the cyclin D1-CDK4/6 dependence of RB phosphorylation [39 , 44] . These data strongly suggest that chronic circadian disturbance promotes an intracellular signaling environment that stimulates cyclin D1 expression in tumors . Around-the-clock assays of mouse liver extracts at 4-hour intervals showed constitutive expression of CDK4 and CCNB1 . Diurnal variations were observed in cyclin D1 expression and RB 807/811 phosphorylation , but were not as robust as seen in cells , perhaps because the adult liver is a non-proliferating tissue ( Fig 4E ) . We also asked if cyclin D1 is a direct target of the clock , and so sought to investigate if a canonical enhancer box ( E-box ) sequence , located at −588 in the upstream region of the human cyclin D1 gene ( Fig 4C ) , is relevant for clock-mediated expression . To address this , we conducted transcriptional analysis using a luciferase reporter ( phCCND1-Luc ) in HEK293T cells that are typically used for such assays [22] . We found that cyclin D1 promoter activity was significantly up-regulated by CLOCK/BMAL1 and repressed by CRY1 when co-expressed ( Fig 4F ) . Thus , cyclin D1 is a direct target of the core clock machinery , supported by recent microarray analysis identifying cyclin D1 as a rhythmically expressed gene in most mouse tissues [45] . Taken together , these data suggest that cyclin D1 is a critical component required for rhythmic RB phosphorylation by CDK4/6 , mediating circadian coordination or alteration of G1/S phase progression . The cyclin D1-CDK4/6–RB pathway is a major target for G1/S phase inhibitors in cancer chemotherapy [46] . For instance , palbociclib ( PD0332991 ) is a potent and selective CDK4/6 inhibitor used for the treatment of specific cancers [47] . We found that palbociclib efficiently and dose-dependently blocked RB phosphorylation at several residues , in conjunction with reduced levels of total RB protein , in U2OS cells ( Fig 5A ) . This supports our prior hypothesis of phosphorylation-dependent stabilization of RB . Conversely , the phospho-RB inhibitor gradually increased protein abundance of cyclin D1 or CDK4/6 dose dependently ( Fig 5A ) . This is consistent with previous reports showing that palbociclib can stabilize cyclin D-CDK4/6 complexes by inducing adaptive responses in mammalian target of rapamycin ( mTOR ) or phosphatidylinositol 3-kinase/alpha serine/threonine-protein kinase PI3K/AKT pathways [48 , 49] . Considering the cyclic RB phosphorylation in our data above , we sought to examine circadian kinetics of the antiproliferative effect of palbociclib on dex-synchronized U2OS osteosarcoma cells ( Fig 5B ) . The time course of an MTT cell viability assay revealed that palbociclib exhibited significantly higher antiproliferative activity at 30 hours , relative to other time points , following dex synchronization in both 48-hour and 72-hour drug incubation protocols ( Fig 5C , and S9A and S9B Fig ) . Palbociclib did not show a substantial effect on circadian rhythmicity , as assayed by determining its effect on the Per2 reporter in U2OS cells , although slightly prolonged rhythms were observed at higher doses , probably due to severe cell growth arrest by the drug ( S7C , S7D , S9E and S9F Figs ) . To determine whether rhythmic variations in drug efficacy were circadian clock dependent , we subjected U2OS cells stably expressing BMAL1 siRNA ( si-BMAL1stable ) , or GFP siRNA as a control ( si-GFPstable ) , to CCD and then followed up with the timed pharmacological experiments as above . MTT analysis showed a circadian pattern of drug sensitivity ( pMetaCycle < 0 . 05 ) in control si-GFPstable cells , but the rhythm of drug response was phase reversed in JL GFPstable cells , reflecting altered cycling in these cells ( Fig 5D , left panels ) . Furthermore , rhythmic effects of the inhibitor were abrogated in CTL and JL si-BMAL1stable cells ( Fig 5D , right panels ) , although they were not totally eliminated—likely because siRNA does not produce complete knockdown . These results indicate that the antiproliferative effect of palbociclib , which targets CDK4/6 activity , is regulated by the circadian clock and can be severely altered by rhythm desynchrony . Based on these findings , we sought to determine if a G1/S cell cycle–specific inhibitor shows time-of-day–specific effects in an in vivo mouse model . To this end , 30–60 days after subcutaneous injection of the MCA carcinogen , mice were subjected to a chronic jet-lag ( CJL ) protocol , the effects of which were verified by testing the mice for locomotor rhythms ( Fig 6A and 6B ) . Consistent with a previous study [4] , jet-lagged mice ( Jet-lag ) formed tumors faster than controls ( Control ) ( Fig 6C and 6D ) . To investigate the time-dependent efficacy of palbociclib , we gave daily oral treatment to Control and Jet lag mice for 5 days at three hours after lights on ( zeitgeber time [ZT3] ) ( morning ) or three hours after lights off ( ZT15 ) ( night ) following the chronic jet-lag schedule ( Fig 6A ) . A significant reduction of tumor growth rate was observed in Control mice receiving the drug at ZT3 , but not those treated at ZT15 ( Fig 6E and 6G ) . Moreover , the time-dependent antitumor activity of palbociclib was compromised in Jet lag mice ( Fig 6F and 6H ) . To explore time dependence of this drug with respect to other types of tumors , we induced tumors in mice through subcutaneous injection of B16 mouse melanoma cells , which possess intrinsic circadian function [15] . Interestingly , similar to the results observed in MCA-induced tumors , the inoculated melanoma grew faster in Jet-lag mice than Control during the chronic jet-lag regime ( S10A , S10B , S10C and S10D Fig ) . Furthermore , the melanoma tumors exhibited significantly higher drug sensitivity at ZT3 than ZT15 in Control mice when subjected to daily treatment with palbociclib at these respective time points ( S10E and S10G Fig ) . As with the MCA-induced tumors , time-of-day effects of the drug were severely compromised in Jet-lag mice ( S10F and S10H Fig ) . In conjunction with the cell-based data , our in vivo animal data indicate that circadian regulation of G1/S progression confers time-of-day sensitivity to antitumor agents that act at this step , but the rhythm of sensitivity is lost under conditions of circadian desynchrony .
In this study , we demonstrate the feasibility of a cell-based assay to investigate the impact of circadian desynchrony on a cellular/molecular level . The use of this model allowed us to comprehensively assess impairments in cell physiology induced by a jet-lag protocol , and it revealed a molecular mechanism that is not only relevant for the role of circadian dysregulation in cancer but also provides a molecular target for chronotherapy . Importantly , we show that the mechanism as well as time-of-day effects of a drug targeting the mechanism are manifest in mouse tumor models . The dex synchronization model is used largely to induce free-running circadian oscillations in mammalian cell lines . However , GCs such as dex are also relevant for circadian synchrony in the organism . Besides photic entrainment driven by the SCN , mammalian body clocks can be directly reset by multiple stimuli such as stress , exercise , and nutrition [50] , of which GC hormones released from the adrenal cortex are among the most potent non-photic synchronizers [51] . GC levels exhibit robust daily oscillations , with peak expression in the morning and a trough at night , driven by circadian regulation of GC production by a local clock in the adrenal gland as well as by systemic innervation [52] . GCs mediate circadian responses to various psychosocial stresses [53–55] , and thus disruption of GC rhythms by shift work or jet lag could be relevant to pathological conditions [56] . Indeed , disruption of daily GC rhythms is even linked to cancer [57] , perhaps related to the inhibitory effects of GC on cell proliferation [58 , 59] . These critical roles of GC in normal circadian physiology provide an important rationale for the use of dex to impose CCD in a cell-based model . Notably , in our circadian transcriptome data ( Fig 1F ) , a significant number of genes exhibited circadian cycling in the daily dex-synchronized U2OS cells ( n = 44 ) , compared with cycling genes estimated in the same cells after a single dex treatment in a previous report ( n = 7 ) [60] . This could be primarily due to different experimental and rhythm assessment procedures , for instance , with respect to the measured time points and analysis criteria used in each of the studies . Another possibility for the discrepancy is that the regular rhythmic environment , with daily dex treatment , induced more robust cycling of responsive genes . This is reminiscent of previous circadian gene expression profiling data showing that a large number of genes exhibit diurnal cycling , but not all cycle in free run , nor are all directly clock controlled [45 , 61] . Given the robust diurnal fluctuation of GC , it is tempting to speculate that our daily dex-synchronization regimen closely mimics physiological cycling in the presence of light–dark cycles . Malignant cell proliferation and tumor growth is a major pathological consequence of chronic jet lag [7] . Underscoring the link between disrupted clocks and cancer , our unbiased approach , which consisted of characterizing many different cellular functions as well as conducting genome-wide RNA-seq to determine how cell physiology is perturbed by circadian disruption , reveals that cell proliferation , along with concomitant changes in gene expression , is the most highly affected function . Our chronic jet-lag protocol results in a dramatic increase in the ratio of oncogenic genes to tumor suppressor genes . Given the tumor suppressive effect of enhanced circadian function in a recent study [15] , we suggest that chronic circadian perturbation tips the transcriptional balance of tumor-progressive and -suppressive genes in favor of promoting tumor survival and proliferation . This explains in part why carcinogen-induced or -injected tumors grow faster under chronic jet lag ( Fig 6C and 6D , S10C and S10D Fig ) [4] . We note that clock proteins may also directly regulate tumor suppressors and oncogenic proteins [62 , 63] . However , as our chronic jet-lag experiments reported here are performed with cancer cells and induced tumor tissues , we cannot speak to how a jet-lag paradigm would impact nontransformed cells . Also , because not all cancer cells have functional circadian clocks [64] , the proposed mechanism would likely only apply to the subsets of cancers that maintain a clock function . On the other hand , recent studies have shown that circadian rhythms can be reprogrammed or otherwise affected by tumors or multiple tumor components [65] . For example , lung adenocarcinoma and breast cancer reprogram circadian rhythms in liver metabolism and transcription , respectively [66 , 67] . Furthermore , cell cycle–related oncogenes ( MYC , RAS ) were found to affect the circadian clock [68 , 69] . These findings indicate a dynamic bidirectional relationship between circadian disruption and cancer progression . Several mitogenic and oncogenic pathways signal to the cell cycle machinery through D-type cyclins [9] . Consistent with this , our transcriptomic analysis reveals extensive up-regulation of upstream signals and downstream transcriptional activators of the cyclin D1 gene in U2OS cells exposed to chronic jet lag ( Fig 4C and S8 Fig ) . More interestingly , cyclin D1 expression was markedly elevated in tumors , but not in other tissue , in chronically jet-lagged mice ( Fig 4D ) . These results suggest that the expression of cell cycle proteins is more sensitive to chronic jet lag in cancerous cells . We note that cyclin D3 , another D-type G1 cyclin important for cancer cell survival and progression [70 , 71] , was also up-regulated by CCD in our transcriptomic data ( Fig 3K and S4B Fig ) . Our findings regarding cyclin D corroborate the reported oncogenic role of cyclin D in triggering spontaneous tumors or inducing tumors in response to mitogenic and oncogenic signals [72] . Taken together , these data suggest that cyclin D mediates abnormal tumor proliferation in response to chronic jet lag . Molecular links between the circadian clock and the cell cycle have been suggested in a few previous studies . For instance , in a murine hepatectomy model , circadian regulation of the G2-M transition was proposed to be driven by clock-controlled expression of Wee1 G2 checkpoint kinase ( Wee1 ) , a key cell cycle inhibitor that phosphorylates and thereby inactivates the CDK1–cyclinB1 complex [73] . Timely coordination of the G1/S phase is suggested by circadian transcriptional regulation of p21 and p16 , which encode inhibitors of cyclin-dependent kinases , including CDK4/6 [74 , 75] . In addition , the myelocytomatosis oncogene cellular homolog ( Myc ) proto-oncogene is thought to be regulated by BMAL1/neuronal PAS domain protein 2 ( NPAS2 ) ( a CLOCK paralogue ) action on its E-boxes , which may contribute to enhanced tumor growth in Per2 mutant mice in response to DNA damage [76] . Myc can act through p27 ( Kip1 ) , another multifunctional CDK inhibitor , to facilitate the activation of CDK4/6 [77] . Our in vitro and in vivo data demonstrate that CLOCK/BMAL1 directly regulate cyclin D1 , which likely contributes to rhythmic CDK4/6 activity ( Fig 4 ) , so multiple circadian mechanisms may converge on CDK4/6 . We show also that chronic jet lag alters the phosphorylation of RB , which is the rate-limiting substrate of cyclin D1–CDK4/6 in the cell cycle progression [35] . Importantly , RB phosphorylation is predominantly affected at S807/811 , which is a known CDK4/6 site [78] . S807/811 on RB serves as a priming site whose induced phosphorylation releases the tumor-suppressive effect of RB and promotes cell cycle entry , as well as cell survival [37–40] . In this regard , rescue of RB-null cells ( C33A ) with RB phospho mutants ( S807/811A ) significantly blunted but did not completely abrogate the enhanced cell proliferation upon CCD ( S7 Fig ) . We speculate that RB-like protein 1 ( RBL1 , p107 ) , which has structural and functional similarities to RB and is regulated by cyclin D–CDK4/6 to de-repress E2F activity during G1/S cell cycle progression [79] , could contribute to enhanced cell proliferation upon chronic circadian disturbance . Emerging cancer therapeutic efforts target G1/S cell cycle progression , with a focus on the development of selective CDK 4/6 inhibitors [46] . For instance , palbociclib ( PD0332991 , Ibrance; Pfizer ) is a potent oral inhibitor of CDK4/6 initially approved by the United States Food and Drug Administration ( FDA ) for the treatment of metastatic breast cancer in combination with endocrine therapy [80 , 81] . Numerous preclinical and clinical studies show that palbociclib has antiproliferative activity in several types of RB-positive tumors , and it is becoming more widely used to treat multiple cancers [47 , 82] . Despite the growing clinical use of palbociclib and other CDK4/6 inhibitors such as abemaciclib ( Lilly ) and ribociclib ( Novartis ) , these drugs are typically prescribed without any time-of-day indications . Our in vitro and in vivo chronopharmacological studies demonstrate that the cyclin D1-CDK4/6–RB pathway is a target for chronotherapy , likely because of circadian regulation of the G1/S phase . This pathway could even be a circadian marker for cancer therapy targeting the G1/S phase ( Figs 5 and 6 ) . Pharmacodynamic variability among cancer patients remains a daunting challenge in cancer drug therapy [83] . Our findings strongly suggest that environmental or physiological perturbation of circadian rhythms such as shift work , abnormal sleep timing , or irregular psychosociological stresses can underlie interindividual variability in both cancer growth and response to cancer drugs . Circadian disruption may also be relevant for the chronic sleep loss and depression suffered by many cancer patients following diagnosis and treatment [84] . Given this , it is reasonable to expect that resetting of the body clock by scheduled light exposure , mealtimes , or exercise , alongside a carefully timed chemotherapy regimen , would improve antitumor treatment . Taken together , our study provides a mechanism for tumor susceptibility conferred by circadian dysregulation and highlights the importance of judicious application of cancer chronotherapy .
All live animal experiments were performed according to protocols ( 806387 ) approved by Institutional Animal Care and Use Committee of the University of Pennsylvania in accordance with guidelines set by the NIH . U2OS , HEK293T , or C33A cells purchased from ATCC were cultured in Dulbecco's Modified Eagle Medium ( DMEM ) supplemented with 10% FBS and 1% penicillin-streptomycin ( 15140122 , Thermo Fisher Scientific , Waltham , MA ) at 37 °C under 5% CO2 . The cells were transfected with siRNAs using Lipofectamine RNAi tranfection reagent ( 13778075 , Thermo Fisher Scientific , Waltham , MA ) and DNA plasmids using transfection reagents ( E2311 , Promega , Madison , WI ) . The plasmids encoding wild-type RB , non-phosphorylatable RB with 15 putative CDK sites converted to Ala residues ( RB-ΔCDK ) , and single CDK site monophosphorylatable RB proteins ( RB-ΔCDK+S612 , RB-ΔCDK+S780 , RB-ΔCDK+S807 , RB-ΔCDK+S811 ) were purchased from Addgene ( Cambridge , MA ) , and were designated as follows: pCMV-HA-hRB-WT ( 58905 ) , pCMV-HA-hRB-ΔCDK ( 58906 ) , pCMV-HA-hRB-ΔCDK+S811 ( 58919 ) , pCMV-HA-hRB-ΔCDK+S807 ( 58918 ) , pCMV-HA-hRB-ΔCDK+S780 ( 58915 ) , and pCMV-HA-hRB-ΔCDK+S612 ( 58914 ) . The other set of plasmids encoding wild-type RB and alanine mutants S807A and S811A of RB ( RB-S807A and RB-S811A ) were kindly provided by Dr . Nancy A . Krucher at the Department of Biology and Health Science , Pace University . For cyclin D1 promoter analysis , a −1748 human cyclin D1 promoter pGL3 basic ( 32726 ) and a −962 human cyclin D1 promoter pGL3 basic ( 32727 ) were purchased from Addgene . For luciferase assays , pCMV Sport6 plasmids encoding full-length cDNAs of CLOCK , BMAL1 , and CRY1 were obtained from the Mammalian Gene Collection ( MGC , Thermo Fisher Scientific , Waltham , MA ) . A total of 1 . 5 × 105 pPer2 reporter U2OS cells were seeded per 35-mm dish . After the chronic 100 nM dex ( D2915 , Sigma , St . Louis , MO ) or forskolin ( S2449 , Selleckchem , Houston , TX ) synchronization schedule , as depicted in Fig 1A , the cells were treated with dex every seventh morning , delivered in the medium mentioned above . Real-time bioluminescence of the control and jet-lag cells was monitored using a LumiCycle luminometer ( Actimetrics , Wilmette , IL ) , as previously described [21] , and period and amplitude of the luminescence data were determined through the LumiCycle software program ( Actimetrics , Wilmette , IL ) . HEK293T cells transfected with plasmids encoding proteins as indicated in the figures were assayed for luciferase reporter activity using the luciferin reagent ( E2920 , Promega , Madison , WI ) 24 hours post-transfection according to the manufacturer's protocol . Immunoblot analysis for U2OS cells was performed as described [85] using the following antibodies: anti-BMAL1 ( 14020 , Cell Signaling , Danvers , MA ) , CRY1 ( A302-614A , Bethyl Laboratories , Montgomery , TX ) , anti-CRY2 ( 13997-1-AP , Proteintech , Chicago , IL ) , anti-RB ( 9309 , Cell Signaling , Danvers , MA ) , anti-pRB-S807/811 ( 8516 , Cell Signaling , Danvers , MA ) , anti-pRB-S795 ( 9301 , Cell Signaling , Danvers , MA ) , anti-pRB-S780 ( 8180 , Cell Signaling , Danvers , MA ) , anti-pRB-S612 ( AP3236a , Abgent , San Diego , CA ) , anti-CCNB1 ( #4138 , Cell Signaling , Danvers , MA ) , anti-cyclin D1 ( ab134175 , Abcam , Cambridge , MA ) , anti-CDK4 ( 12790 , Cell Signaling , Danvers , MA ) , anti-CDK6 ( 14052-1-AP , Proteintech , Chicago , IL ) . For liver tissue extracts , anti-CDK4 ( ab137675 , Abcam , Cambridge , MA ) , anti-RB ( ab24 , Abcam , Cambridge , MA ) , and anti-pRB-S807/811 ( ABC132 , MilliporeSigma , Burlington , MA ) were used . Anti-GAPDH ( sc25778 , Santa Cruz Biotech , Dallas , TX ) was used as loading control antibody for both U2OS cell and liver tissue extracts . Twenty-four hours after the final dex stimulation , as per the experimental schedule depicted in Fig 1A , control and jet-lag cells were harvested and subjected to a cell viability assay with 0 . 4% tryptopan blue ( 15250061 , Thermo Fisher Scientific , Waltham , MA ) to determine cell numbers according to the manufacturer’s protocol . Similarly , after 5 × 103 U2OS cells were seeded per well in 96-well plates , cell viability in control or jet-lag cells following the dex ( 100 nM ) or forskolin ( 100 nM ) synchronization schedule was determined colorimetrically by MTT [3- ( 4 , 5-dimethylthiazol-2-yl ) -2 , 5-diphenyl-2H-tetrazolium bromide] using the MTT cell Proliferation assay per the kit protocol ( V13154 , Thermo Fisher Scientific , Waltham , MA ) or alamar blue cell viability reagent ( DAL1025 , Thermo Fisher Scientific , Waltham , MA ) . Briefly , prior to addition of the MTT reagent , microplates were subject to the removal of media by decantation followed by the addition of 100 μL of fresh media and 10 μL of freshly prepared MTT reagent . The plates were incubated at 37 °C for 4 hours prior to the addition of 100 μL freshly prepared SDS-HCl solution , followed by mixing with a pipettor . The plates were incubated overnight at 37 °C and absorbance read at 570 nm on the Epoch Microplate Spectrophotometer . The same experimental procedures were performed for cell viability analysis of U2OS cells transfected with plasmids encoding RB and various phosphor-RB mutants , as described above . To determine S-phase distribution , after 2 . 5 × 104 U2OS cells were seeded per well of 24-well plates , control and jet-lag cells following the dex ( 100 nM ) synchronization schedule were labeled with 10 μM BrdU ( ab142567 , Abcam , Cambridge , MA ) for 6 hours . Cells were fixed in 4% paraformaldehyde ( PFA ) ( in PBS ) for 15 minutes at room temperature , washed three times , and incubated with 0 . 5% Triton X-100 permeabilization buffer for 20 minutes . The fixed cells were incubated with 1N HCl for 10 minutes on ice , 2N HCL for 20 minutes at room temperature to denature , and neutralized with 0 . 1 M sodium borate ( pH 8 . 5 ) for 2 minutes . The cells were incubated with anti-BrdU antibody ( ab6326 , Abcam , Cambridge , MA ) overnight and stained by an Alexa Fluor 568–conjugated secondary antibody ( A-11077 , Thermo Fisher Scientific , Waltham , MA ) . The cells were imaged with a fluorescence microscope using TRITC ( excitation [Ex . ] = 579 nm and emission [Em . ] = 604 nm ) and DAPI ( Ex . = 370 nm and Em . = 460 nm ) filter sets with the same light exposure time . Proliferating cells represented by Alexa Fluor 568–stained nuclei were counted as a percentage of all nuclei over 10 microscopic fields . For measurement of cellular death rate , after 1 . 5 × 105 U2OS cells were plated per 35-mm dish and exposed to the dex ( 100 nM ) synchronization schedule , control and jet-lag cells were harvested for apoptosis analysis performed according to the Annexin V-FITC Early Apoptosis Detection Kit protocol ( 6592 , Cell Signaling , Danvers , MA ) . Fluorescence was analyzed using a FACS Calibur ( BD Biosciences , San Jose , CA ) and FlowJo software ( FlowJo , LLC , Ashland , OR ) . For measurement of H2O2 level , GSH/GSSG ratio , NADP/NADPH ratio , and protease activity in the control and jet-lag cells seeded in a 96-well plate , luminescence-based assays were performed using H2O2 assay kit ( G8820 , Promega , Madison , WI ) , NADP/NADPH assay kit ( G9081 , Promega , Madison , WI ) , GSH-Glutathione assay kit ( V6911 , Promega , Madison , WI ) , and Proteasome-Chymotrypsin-Like Cell-based assay kit ( G8660 , Promega , Madison , WI ) according to manufacturer's protocols . At the final dex-containing media change during the chronic desynchronization schedule after 2 . 5 × 104 U2OS cells were seeded per well of 24-well plates , the control and jet-lag cells were transduced with baculovirus-expressing FUCCI cell cycle sensors ( Cdt1-RFP , Geminin-GFP ) provided by a FUCCI cell cycle sensor kit ( P36237 , Thermo Fisher Scientific , Waltham , MA ) for 48 hours and were fixed in 4% PFA in PBS for microscopic analysis . They were imaged with a fluorescence microscope using GFP ( Ex . = 492 nm and Em . = 514 nm ) , RFP ( Ex . = 579 nm and Em . = 604 nm ) , and DAPI ( Ex . = 370 nm and Em . = 460 nm ) filter sets with the same light exposure time . For quantification of the fraction of FUCCI cell cycle indicator–positive cells , the proportion of Geminin-GFP–or Cdt1-RFP–positive cells from the total number of Dapi-stained nuclei ( >200 ) in the CTL or JL cells was averaged from four optical fields scanned with a 20× objective . The same experimental procedures were performed for cell cycle analysis of U2OS cells transfected with plasmids encoding RB and various phosphor-RB mutants , as described above . siRNAs targeting human BMAL1 ( GS406 ) , CRY1 ( SI02757370 ) , CRY2 ( GS1408 ) , cyclin D1 ( SI02654540 ) , CDK6 ( SI00605052 ) , and CDK4 ( SI00299789 ) were purchased from Qiagen . Transfection of 100 pmol siRNAs per well in U2OS cells was conducted with Lipofectamine RNAi Transfection Reagent ( 13778075 , Thermo Fisher Scientific , Waltham , MA ) according to the manufacturer's instructions . Immunoblotting was performed 48 hours after transfection . U2OS cells were stably transfected with lentiviral particles expressing scrambled GFP siRNA as a control ( LVP015-G ) or pooled human BMAL1 ( ARNTL ) siRNA ( iV001368 ) purchased from Applied Biological Materials ( Richmond , Canada ) . The cells were grown with the addition of selection marker ( G418; Invitrogen ) for 4 weeks . After selection , western blot and chronopharmacological experiments were performed as described . After the final dex stimulation in the experimental schedule depicted in Fig 1A , control and jet-lag cells were collected at every 6-hour point ( 24 hours , 30 hours , 36 hours , 42 hours , 48 hours ) for 24 hours ( black arrows in Fig 1A ) . Three replicate dishes of cells were collected at each time point and processed as separate samples . Total RNA was extracted from each sample using the RNA extraction kit ( 74134 , Qiagen , Germantown , MD ) according to the manufacturer’s protocol . RNA quality was assessed by Bioanalyzer , and samples with a high RNA integrity number ( >8 ) were used for library construction . For each sample , Illumina sequencing libraries were prepared from 100 ng of total RNA using the TruSeq Stranded mRNA Sample Prep Kit ( Illumina , San Diego , CA ) , according to the manufacturer’s protocol ( TruSeq RNA Sample Preparation V2 Guide ) . Each library was prepared using a unique adapter bar code index to allow for multiplexing . All libraries were pooled together and sequenced across three lanes of an Illumina HiSeq 2000 Sequencer running in single-end mode ( 1 × 100 bp ) . Raw RNA-Seq reads were mapped to the GRCh38 build of the human reference genome using STAR v2 . 5 . 3a [86] , with the following command line arguments:—outSAMtype BAM Unsorted—outSAMunmapped Within KeepPairs—outFilterMismatchNmax 33—seedSearchStartLmax 33—alignSJoverhangMin 8 . STAR was provided with gene models from the Ensembl v90 genome annotation [87] . Note that the three fastq files per library ( one for each sequencing lane ) were aligned separately . Following alignment , the three BAM files for each library were assigned read groups and merged together using the AddOrReplaceReadGroups and MergeSamFiles commands from Picard Tools v2 . 7 . 1 ( http://broadinstitute . github . io/picard ) , respectively . For both commands , BAM files were maintained in unsorted order ( i . e . , sorted by read ID ) . The Pipeline of RNA-Seq Transformations ( PORT ) v0 . 8 . 4-beta ( https://github . com/itmat/Normalization ) was used to perform gene-level normalization and quantification of the aligned data . Raw fastq files and quantification tables are available from the Gene Expression Omnibus database ( GSE119668 ) . To identify genes cycling with a 24-hour period , PORT-normalized gene counts from control and jet-lag samples were processed separately using the meta2d function of the MetaCycle R package v1 . 1 [26] . The meta2d function was run with the following arguments: adjustPhase = "predictedPer" , combinePvalue = "fisher" , weightedPerPha = FALSE , cycMethod = c ( "JTK" , "LS" ) , minper = 24 , maxper = 24 . Genes with significant 24-hour rhythms in expression were identified using a meta2d combined false discovery rate ( FDR ) cutoff of 0 . 1 . To identify genes with broad DE between the control and jet-lag conditions , PORT-normalized gene counts were analyzed using the limma R package v3 . 34 [88] . Differentially expressed genes ( DEGs ) were identified using a limma FDR cutoff of 0 . 4 . This cutoff was selected to yield a sufficient number of DEGs for Ingenuity Pathway Analysis ( IPA; Qiagen , https://www . qiagenbioinformatics . com/products/ingenuity-pathway-analysis/ ) . For IPA core analysis , the entire set of input genes served as the background for the enrichment tests . C57BL/6J mice ( Jackson Laboratory , Bar Harbor , ME ) were housed ( ≤5/cage ) under 12-hour light–12-hour dark conditions with food and water available ad libitum . Male mice ( 2–3 months old ) were used in all experiments . After acclimation with standard lighting conditions of LD 12:12 , with lights on from 5 AM ( Zeitgeber time 0 ) to 5 PM ( Zeitgeber time 12 ) for 2 weeks , the mice were injected with 1 × 106 B16-10 cells subcutaneously along the right flank and separated into control and chronic jet-lag groups . For MCA-induced tumors , mice were injected subcutaneously along the right flank with 400 μg of 3-MCA in peanut oil ( 213942 , Sigma , Louis , MO ) . After 30–60 days of the injection , the mice were separated into control and chronic jet-lag groups . The chronic jet-lag mice underwent repeated 8-hour advances of the light–dark cycle every 2 days for 11 days , while the control mice further remained in the LD 12:12 lighting regimen . Their locomotor activity was monitored as described [89] . Tumor growth was measured with a digital caliper three times per week . Tumor weight was computed as follows: ( length × width2 ) /2 . In all animal experiments , mice were euthanized when the tumor exceeded 20 mm in diameter ( approximately a volume of 3 , 500 mm3 ) . Tumor and liver tissues were collected at ZT3 , ZT9 , ZT15 to measure circadian changes of RB phosphorylation and expression of cell cycle proteins with western blot analysis , as described . For pharmacological inhibition of CDK4/6 activity in U2OS cells , palbociclib hydrochloride ( PD-0332991 HCL ) ( S1116 , Selleckchem , Houston , TX ) was used in various experimental settings as described ( Fig 5 ) . For in vivo animal studies , palbociclib was dissolved in 50 mmol/L sodium lactate , pH 4 , and administered by oral gavage daily ( 120 mg/kg ) at ZT3 and ZT15 after measurable tumors were formed following the chronic jet-lag schedule , as described [90] . For evaluation of antitumor activity of the drug , tumor growth was measured with a digital caliper daily during drug administration . Additional details for each experiment are given in the Fig 6 legend . All statistical tests used in this study were completed with Prism7 GraphPad Software . For making multiple comparisons , we used one-way or two-way ANOVA followed by Bonferroni , Sidak , and Tukey multiple comparisons tests . For comparing the average of two means , we used the Student t test ( two-tailed paired or unpaired ) to reject the null hypothesis ( p < 0 . 05 ) . | Circadian misalignment caused by altered sleep–wake cycles , shift work , or frequent jet lag increases susceptibility to several disorders , including cancer . However , the mechanisms by which circadian disruption contributes to disease are not well understood , and so we addressed this issue by investigating the molecular , cellular , and biochemical consequences of chronic circadian desynchronization . Our studies using cancer cell or tumor tissue models show that chronic circadian desynchronization induces multiple oncogenic pathways to promote cell proliferation . In particular , chronic circadian desynchronization promotes phosphorylation of the retinoblastoma ( RB ) protein , thereby favoring G1/S phase cell cycle progression . Consistent with these findings , the antiproliferative activity of a selective inhibitor of the enzyme that phosphorylates RB has time-of-day–specific effects on cancer cells and mouse tumors , but this time dependence is abrogated by chronic jet-lag conditions . These data suggest a circadian regulation of G1/S cell cycle progression and provide an important molecular rationale for time-of-day–specific treatment of cancer patients , also known as chronotherapy . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"sequencing",
"techniques",
"phosphorylation",
"gene",
"regulation",
"cell",
"cycle",
"and",
"cell",
"division",
"cell",
"processes",
"circadian",
"oscillators",
"chronobiology",
"molecular",
"biology",
"techniques",
"rna",
"sequencing",
"research",
"and",
"analysis",
"methods",
"small",
"interfering",
"rnas",
"proteins",
"gene",
"expression",
"molecular",
"biology",
"circadian",
"rhythms",
"biochemistry",
"rna",
"cell",
"biology",
"post-translational",
"modification",
"nucleic",
"acids",
"genetics",
"biology",
"and",
"life",
"sciences",
"non-coding",
"rna",
"cyclins"
] | 2019 | G1/S cell cycle regulators mediate effects of circadian dysregulation on tumor growth and provide targets for timed anticancer treatment |
While the geographic range of a species is a fundamental unit of macroecology and a leading predictor of extinction risk , the evolutionary dynamics of species' ranges remain poorly understood . Based on statistical associations between range size and species age , many studies have claimed support for general models of range evolution in which the area occupied by a species varies predictably over the course of its life . Such claims have been made using both paleontological data and molecular estimates of the age of extant species . However , using a stochastic model , we show that the appearance of trends in range size with species' age can arise even when range sizes have evolved at random through time . This occurs because the samples of species used in existing studies are likely to be biased with respect to range size: for example , only those species that happened to have large or expanding ranges are likely to survive to the present , while extinct species will tend to be those whose ranges , by chance , declined through time . We compared the relationship between the age and range size of species arising under our stochastic model to those observed across 1 , 269 species of extant birds and mammals and 140 species of extinct Cenozoic marine mollusks . We find that the stochastic model is able to generate the full spectrum of empirical age–area relationships , implying that such trends cannot be simply interpreted as evidence for models of directional range size evolution . Our results therefore challenge the theory that species undergo predictable phases of geographic expansion and contraction through time .
The geographic area occupied by a species is known to vary through time [1] , [2] , but whether these dynamics follow any regular trends or are instead largely idiosyncratic remains controversial [3]–[6] . A number of recent studies , however , have revealed a remarkable pattern whereby the relative range size of a species does appear to vary predictably with its evolutionary age [7]–[15] . The results of these studies are intriguing because they suggest that species have a geographical ontogeny , akin to the life cycle of an individual organism [14] . Similar ideas have been proposed many times before [1] , [2] , [4] , [16]–[19] . For example , in Willis's [1] theory of “age and area” , the geographic range of a species continues to expand over the course of its life , while in the “taxon cycle” , newly formed species rapidly expand their distributions before undergoing a protracted decline that ends either in extinction or the initiation of a new wave of the cycle [4] , [17] . In the fossil record , where the trajectories of individual lineages can be traced through time , the predominant trend appears to be for newly formed species to gradually expand their ranges , only to later undergo a gradual decline to extinction . This pattern of “rise and fall” has been reported from a taxonomically broad set of groups including mammals [10] , [20] , plankton [8] , [9] , and marine mollusks [7] . Molecular phylogenies have also been used to infer the mode of range evolution within lineages by comparing the range sizes of extant species of varying age , where age is estimated as the time since divergence from the closest extant relative [11] ( for an alternative method of inferring range evolution see [21] , [22] ) . Such “age–area” correlations have been tested in a wide variety of groups including birds [11] , [14] , mammals [12] , plants [13] , frogs [15] , aquatic beetles [23] , and reef fish [24] . A pattern emerging from these studies is that no single model of range evolution appears to apply across groups [3]: while in some cases range size is independent of species age , in others there is evidence for increasing , decreasing , or cyclical range size trends . The assumption underlying all these previous studies is that , in the absence of any tendency for ranges to either expand or contract , range size and species age should be independent [7] , [8] , [11] . Evidence for directional trends in range evolution is therefore provided when the slope of the relationship between age and area departs significantly from zero . However this assumption may be violated under a model of stochastic range evolution , speciation , and extinction . This is because , in phylogenies containing only extant species , range sizes may increase with age simply because species with small or declining ranges are less likely to have survived to the present [25] , [26] . Further biases may also be expected because of the interactions between geographic range size and the process of speciation [22] , [27] . Even in the fossil record , the appearance of regular trends in range size could arise if studies restrict their analysis to those species that go extinct before the present [7]–[9] or if species with small geographic ranges are less likely to be sampled [28]–[30] . To examine the extent to which these processes could account for the patterns observed between range size and species age , we developed a stochastic model of range evolution that incorporates the effects of speciation and extinction on geographic range size [27] , [31]–[33] . We then compared the age–area correlations arising from this model to those observed across the tips of 39 avian and mammalian molecular phylogenies containing 1 , 269 species from across 64 genera and 17 orders . These groups provide an ideal case study because there are abundant data on their phylogenetic relationships and geographic distributions [34]–[36] , and as a result they have also featured prominently in previous studies reporting trends between range size and evolutionary age ( e . g . , [12] , [14] , [37] ) . Finally , we tested the extent to which the null model can account for the patterns observed in the fossil record using the occupancy trajectories of 140 species of extinct marine mollusks [7] . This dataset was previously presented as evidence that species exhibit regular patterns of range size evolution [7] , but a departure from a stochastic model has not yet been tested . We demonstrate that under a stochastic model , range size and species age are not expected to be independent and that correlations between age and area observed in phylogenies and the fossil record therefore need not imply a deterministic trend in range size evolution within lineages .
We modelled a process of range evolution and species diversification in which range sizes within lineages evolved according to a random walk through time ( see Materials and Methods for details ) . Within the simulation , we considered two models of speciation rate ( either independent or positively correlated with range size ) and varied both the rate of change in range size and the asymmetry of range division amongst the daughter lineages during speciation ( see Materials and Methods for details ) . Extinction occurs when the range size of a species walks to or below zero . For each combination of speciation model , rate of range evolution , and range inheritance asymmetry , we ran 500–5 , 000 replicates of clade diversification and range evolution . Pooling the results from across replicate clades shows that range size may show strong correlations with species age even under a random model of range size evolution ( Figure 1 ) . Range size is positively correlated with evolutionary age across much of the parameter space , and it is evident that high rates of range evolution give rise to stronger positive correlations ( Figure 1 , unfilled circles ) , resulting from increased extinction of small-ranged species . In rare circumstances , our stochastic model predicts negative age–area correlations ( Figure 1 , filled circles ) . This requires slow rates of range evolution , an increase in the probability of speciation with range size , and high asymmetry in range splitting ( Figure 1A ) . Under this scenario , which resembles a peripatric speciation model [38] , small-ranged species are less likely to speciate than large-ranged species but are also unlikely to go extinct . They will therefore tend to occur on the end of longer terminal branches than large-ranged species [27] , [39] . In contrast , when speciation rate is independent of range size ( Figure 1B ) , positive age–area correlations are expected even when extinction is negligible . This occurs because range sizes decrease at speciation , and so species with the largest ranges will tend to be those that have not recently undergone speciation . Although these correlations reveal the overall expected relationship under different scenarios , individual simulations exhibit considerable variation in the correlation between age and area , and pooling smaller sets of simulations shows the extent of this variation ( Figure 1C ) . Therefore , we also calculated the relative frequencies with which individual simulations fall into five broad categories of age–area relationships . To do this , we fitted range size as a quadratic function of species age for each clade in our simulated dataset , using F-tests of the significance of terms to simplify to a linear regression or a null model where appropriate . We aggregated the results of these models into five broad classes: no relationship , increasing relationship , decreasing relationship , intermediate peaks , and intermediate troughs ( see Materials and Methods for details ) ( Figure S1 ) . Our results were qualitatively similar when using more detailed curve classifications ( Figures S1 and S4; Table S1 ) . We assessed the age–area relationships of individual genera of birds and mammals using the same curve classification procedure as for our simulated clades . Observed clades exhibit a variety of relationships , including positive linear and intermediate troughs , but in only a minority of cases are these age–area relationships significant ( 9 . 4%; Tables 1 , S1 , and S2 ) . In accordance with the effects of sample size on statistical power revealed by our simulations ( Figure 1C ) , grouping species into orders ( median richness = 29 species ) rather than genera ( median richness = 12 species ) increases the proportion of clades exhibiting significant age–area correlations to 23 . 5% ( Table 1 ) . Different combinations of parameter values in the null model give rise to differences in the relationship expected between range size and evolutionary age ( Figures 1 , 2A , and 2B ) . One explanation for the variety of age–area correlations observed in the empirical data is therefore that rates of range evolution or the geographic mode of speciation have differed amongst clades . However , when clade sizes are relatively small , as is typical of empirical datasets , we expect to see substantial variation in age–area correlations simply because of chance . To explore this effect , for each point in parameter space , we compared the proportion of observed and simulated clades exhibiting a particular age–area correlation . Because of the strong effect of clade richness on the patterns ( Figure 1C ) , we randomly aggregated our simulated clades until average species richness was similar to that of our empirical dataset . The results for a representative sample of parameter space show that the full spectrum of observed age–area correlations can often arise because of stochastic sampling of the overall null expectation ( Figure 2A and 2B ) . However , the precise frequency of the different age–area curves across avian and mammalian clades cannot generally be explained by any single combination of parameters ( Figure 2A and 2B ) . Overall the results suggest that while substantial variation in age–area correlations observed across clades may be due to chance , differences in the rates of range evolution or modes of geographic speciation across clades may also be required . To investigate whether the stochastic model can account for the patterns observed in the fossil record , we extracted the extinct lineages from our simulations and assigned each one of these to a range size trajectory . To ensure that the definition of species age in our simulations was consistent with the empirical data , we measured absolute species ages under a model of ancestral persistence: upon each speciation event the lineage with the larger range size retained the ancestral species name , with the smaller ranged lineage designated as a new species . Species ages thus represent the time between the first and last occurrence of a species and are not affected by the speciation or extinction of other lineages [40] . The empirical dataset consists of the occupancy trajectories of extinct marine mollusks provided by Foote et al . [7] . Because many ( 44% ) of the species occurred in only three statigraphic stages we did not attempt to fit curves to these and instead assigned each species to one of three possible range trajectories , depending on whether its peak mean range size was reached in the first , second , or third tercile of its life ( see Materials and Methods for details ) . If range size is independent of evolutionary age , then a similar proportion of species should reach their maximum extent across each of the three sampling intervals . We used exactly the same procedure for assigning range size trajectories in our simulated dataset . We found that approximately 57% of mollusk species reached their peak range size in the middle of their lives , with the number of peaks in the first and final third relatively evenly split ( Figure 2C and 2D; Table S3 ) . Our analysis therefore supports the pattern of “rise and fall” previously reported for this group ( multinomial model: p<0 . 017 ) . However , our stochastic model shows that when geographic ranges have evolved randomly through time , range size is not expected to be independent of species age ( Figure 2C and 2D ) . Instead , most species are expected to reach their maximum geographic extent in either the first or second interval of their lives , and typically have smaller ranges at the end of their durations ( Figure 2C and 2D ) . This occurs because extinct species must have undergone a net decline in range size through time , and , when ranges evolve according to a random walk , extinction is likely to be preceded by rarity unless rates of range evolution are extremely high . Whether species undergo a continuous decline or exhibit an intermediate peak depends on the probability of extinction amongst newly formed species ( Figure 2C and 2D ) . When young species have a low probability of extinction , either because they inherit a large range or because ranges are relatively stable , then the predominant pattern is for range sizes to simply decline through time ( Figure 2C and 2D ) . In contrast , when species inherit a small geographic range or when rates of range evolution are high , then only those species that initially expand their distributions are likely to persist for a sufficient length of time to be included in the analysis ( i . e . , more than two time steps; see Materials and Methods for details ) ( Figure 2C and 2D ) . Under these conditions the relative frequency of the different range size trajectories observed across marine mollusks is consistent with that expected under a stochastic model ( Figure 2C and 2D ) .
Directional trends between range size and species age in both molecular phylogenies and the fossil record have generally been interpreted as evidence that species undergo a predictable sequence of geographic expansion and contraction over the course of their life [7]–[14] , [37] . However , using a stochastic model of range evolution we show that trends in range size with evolutionary age are expected even if range sizes have evolved randomly through time . Across the tips of reconstructed phylogenies , significant age–area correlations arise because of both the process of geographic speciation and the censoring of small-ranged species that went extinct before the present ( Figure 3A and 3B ) . For extinct species , range sizes may appear to vary predictably with evolutionary age because of the censoring of those species that either survived to the present or were too rare to be detected ( Figure 3 ) . Our results demonstrate that correlations between range size and species age cannot be reliably interpreted as evidence for deterministic models of range evolution that predict directional trends in range size through time [1]–[4] , [17] . Previous studies have interpreted a positive relationship between range size and evolutionary age as evidence that geographic ranges tend to expand through time [11] , [13] , apparently vindicating Willis's [1] theory of “age and area” . Here we show that this same pattern is also expected under a stochastic model because , when range size undergoes a random walk through time , only those species that , by chance , happened to expand their ranges or were initially widespread are likely to have survived to the present ( Figures 1A , 1B , 3A , and 3B ) . Given the evidence from the fossil record for high rates of extinction [41] and the dependence of these rates on range size [42] , [43] , we suggest that extinction may be a more parsimonious explanation than any underlying trend for range expansion per se . The illusion that geographic ranges tend to expand through time can also arise from the division of geographic ranges during speciation ( Figure 1B ) . In this case the patterns are particularly misleading , for while average range sizes within the clade may be declining across each speciation event [27] , an age–area correlation would instead suggest that ranges have been expanding through time . Negative relationships between range size and evolutionary age have commonly been interpreted as evidence for a taxon cycle model [17] , in which ranges rapidly expand following speciation before gradually contracting to extinction [11] , [12] , [14] . Our results show that a negative age–area correlation can also arise in the absence of any intraspecific range size trends , when newly formed species are geographically restricted but have a low probability of either speciating or going extinct ( Figure 1A ) . These conditions may , in fact , be most likely to occur on island archipelagos , where taxon cycles have been most commonly invoked ( e . g . , [4] , [17] , [44] ) . Here , a species from the mainland gives rise to a number of daughter species , each endemic to an individual or small number of islands , and these persist for long periods of time in isolation [45] . Our results show that a negative relationship between range size and evolutionary age need not imply taxon cycle dynamics . We might expect the fossil record to provide a more reliable signal of range evolution than molecular phylogenies , which rely on inferring these dynamics from a comparison of present-day range sizes [46] . Our analysis suggests , however , that interpreting the patterns in the fossil record is equally challenging because it , too , is likely to be biased with respect to geographic range size . First , when newly formed species are geographically restricted , as expected under a variety of speciation models [47] , then only those species that , by chance , happened to expand their distributions are likely to be sampled . In contrast , those species that immediately declined to extinction are unlikely to be detected [28] , [29] or will be represented in only one or a few sampling intervals ( “singletons” ) , precluding an analysis of their range dynamics [7] , [8] . As in molecular phylogenies , this leads to the illusion that geographic ranges have a tendency to expand post-speciation ( Figure 3 ) . Second , studies examining the fossil record have often excluded extant species on the basis that their full histories have yet to be played out [7] , [8] . By excluding those species that survived to the present , a subsequent decline of geographic range size becomes inevitable ( Figure 3 ) . Our results suggest that , together , these two sampling biases can generate the pattern of “rise and fall” so frequently reported in the fossil record ( Figures 2C , 2D , and 3 ) [7]–[10] , [20] . One of the most surprising aspects of our simulations is the wide variety of relationships between range size and evolutionary age that can be generated by varying only a few biological parameters . For instance , when rates of range size evolution are slow , then—depending on the asymmetry of range division during speciation—our model predicts both negative and positive age–area correlations across extant species ( Figure 2A ) and negative and hump-shaped relationships amongst extinct species ( Figure 2C ) . Differences in rates of range evolution or in the geography of speciation may therefore explain the heterogeneity in empirical age–area relationships reported both in previous and the current analyses [11]–[15] , [24] . Our results also show , however , that even if all clades had evolved under the same conditions , substantial heterogeneity in age–area relationships can arise simply due to chance ( Figure 2 ) . Under some regions of parameter space this stochasticity is able to capture much of the variation observed across the different taxonomic groups ( e . g . , high rates of range size evolution , asymmetric range division , and when probabilities of speciation are independent of range size ) , whereas in other regions the match is extremely poor . For instance , patterns resembling those observed across extinct marine mollusks can be produced by a stochastic model only when range inheritance is highly asymmetric , as expected under a peripatric or micro-vicariance model of speciation . Independently derived estimates of these key parameters will therefore be required to assess the extent to which a single global model , versus a model with varying conditions across clades , is able explain the observed patterns . Here we have focused on the range dynamics of individual species , but it has also been suggested that the areas occupied by entire clades may undergo regular patterns of expansion and contraction through time ( e . g . , [48]–[51] ) . Our results raise the possibility that these higher level dynamics may also not require deterministic explanations . Testing whether the geographic distributions of entire clades depart from a null expectation will require a more complex , spatial version of the model employed here that can account for both species' range sizes and their geographic overlap [27] . That patterns resembling the empirical data can arise from a random model does not imply that changes in species distributions occur at random [52] . Rather , the processes regulating geographic range size may be so complex and subject to historical contingency that the patterns of range evolution across species appear random [6] . Alternatively , even if species ranges did vary predictably through time the diagnostic tests that we have used may not be sufficiently refined to detect this . One reason for this may be that definitions of species age do not correspond to a realistic estimate of the evolutionary time from the birth to the death of a lineage [53] . For instance , under a peripatric model of speciation , the age of the parent species in a reconstructed phylogeny is reset to zero every time a dispersal event gives rise to a new species . Furthermore , given the sensitivity of the stochastic model to the underlying biological assumptions , stronger tests of departure from the null expectation will require constraining the possible parameter space based on a priori knowledge of the particular focal group . Regardless of these assumptions , however , our findings demonstrate that under a biologically realistic null model of range evolution , range sizes should not be expected to be independent of species age .
We used a discrete time simulation of species diversification and range size evolution , starting from a single species with a range size drawn from an empirically realistic [54] log-normal distribution ( log mean μ = 1 , log variance σ2 = 1 , 95% CI = [0 . 4 , 19 . 3] ) . At each time step of the simulation , changes in range size for each extant species were drawn from a normal distribution with μ = 0 , with extinction occurring if the range size of a species drifted to or below zero . By modifying the standard deviation of the distribution , we modelled the effects of differences in the rate of range size evolution ( σ2 = 0 . 0001 , 0 . 01 , 0 . 04 , 0 . 09 , and 0 . 16 ) , and the extremes of these rates lead , respectively , to extinction of 0% and >90% of species by the end of the simulation . At each time step of the simulation , speciation could occur with a per-lineage probability ν . We simulated two speciation models , both of which have strictly bifurcating speciation . In the first model , ν varied as a set proportion of range size ( 0 . 005 ) , corresponding to a value of ν = 0 . 5 when range size equalled 100 units . Such an increase in the probability of speciation with range size is expected under certain probabilistic models of range splitting when range sizes are strongly right skewed [27] , [55] , [56] . To prevent runaway clade growth associated with large range sizes , we truncated the range size used to calculate ν at 100 . However , achieving such range size is rare over the lifetime of the simulation , even under the greatest degree of range change . In the second model , ν was equal across lineages and constant through time ( ν = 0 . 02 ) , analogous to the birth–death model [57] , [58] . This latter value of ν was chosen to result in similar rates of speciation as expected in the range-size-dependent speciation model assuming the initial log-normal range size distribution . Varying the value of ν ( 0 . 014 , 0 . 02 , and 0 . 033 ) in the range-size-independent model or the rate at which ν increased with range size ( 0 . 0025 , 0 . 005 , and 0 . 01 ) did not qualitatively alter our results ( Figures S2 and S3 ) . Upon speciation the geographic range of the parent was split amongst the two daughter lineages , with relative range sizes as a proportion ( p ) of the parent ( p and 1−p ) drawn from a beta distribution ( α = 1 and β = 1 , 5 , 10 , and 70 ) . Increasing values of β resulted in more asymmetric splits , allowing us to model a variety of geographic scenarios from vicariance ( β = 1 ) to peripatry and micro-vicariance ( β = 70 , 95% CI of p = [0 . 0003 , 0 . 05] ) [31] , [32] , [38] , [59] . For each combination of parameter values we generated 500–5 , 000 replicate clades contingent on either the survival of at least six species to the present ( the minimum richness in our empirical dataset of phylogenetic trees; Table S2 ) or the extinction of at least one species ( the minimum required for comparison to our empirical fossil dataset ) . The number of replicates varied across parameter space because of the variation in computational time arising from differences in expected clade size and the probability of obtaining an extinct species . In total we simulated over 1 million clades . For each clade we obtained the reconstructed phylogeny of those species extant at the end of the simulation and recorded the age and area of extant species . In a reconstructed phylogeny , the age of a species is defined as the tip length: the time from the present to the most recent node . This estimate of species age is thus sensitive to the extinction of other lineages and does not account for the possibility of ancestral persistence during speciation . However , these limitations are common to both the observed and simulated phylogenies and so will not bias our results . We constructed trees based on mitochondrial protein coding genes ( cytb , COI , COII , ND2 ) for genera and families of birds and mammals according to standard avian [60] and mammalian taxonomies [61] . Our selection of bird phylogenies largely follows that of Phillimore and Price [36] , but we included additional sequences and species and an expanded taxonomic coverage where possible . Sequences were downloaded from GenBank and aligned by eye in MEGA v4 [62] . Trees were constructed using a relaxed clock Bayesian method in BEAST v1 . 5 . 4 [63] . For each tree we specified a Yule prior on branching times and that variation in rates of substitution amongst branches was uncorrelated and followed a log-normal distribution . For mammals we used an HKY model of substitution with four gamma categories and separate rates for codon positions 1+2 versus 3 . Rates of molecular evolution are known to vary substantially amongst mammalian lineages [64] , and so for each clade , trees were dated using fossil calibrations obtained from the literature . On all calibrations we used a log-normal prior on the age of the split with a mean and standard deviation of one and with the minimum age set to that used in the original study; usually corresponding to the first appearance of a group [65] . For birds we used a GTR-γ model of substitution , assuming an average rate of sequence substitution of 0 . 01 per site per lineage per million years [66] . We performed two runs of 5–10 million generations depending on the time required for convergence as assessed in Tracer v1 . 4 [67] . The first 10% of the generations were discarded as burn-in , and then , depending on the length of the model run , we sampled every 4 , 000–8 , 000 generations to produce a posterior distribution of trees . The trees were combined from the separate runs and summarized as the maximum clade credibility tree , from which we estimated species age as the tip length corresponding to each extant species . In three avian genera ( Cinclodes , Hemispingus , and Muscisaxicola ) , runs failed to converge , and so in these cases we used an HKY model of substitution . Although this simpler model will tend to overestimate species ages , the frequency of significant age–area correlations across clades was similar when we re-ran all trees using the HKY model , showing that our results are robust to model assumptions . All trees are provided in Dataset S1 . Estimates of species ages from clades with large numbers of missing extant species will be relatively inflated , and so from our trees we included only those genera containing at least six species and where at least 80% of the species had been sampled . These clades generally corresponded to monophyletic , taxonomically recognized genera [60] , [61] , but we also included polyphyletic clades if at least 80% of the species from each constituent genera had been sampled . In total , we obtained estimates of species ages for 1 , 269 species from 64 genera across 17 orders ( Table S1 ) . For each species , we obtained the range size ( km2 ) from previously published range databases of birds [68] and mammals [34] . Because range sizes and species ages tend to be positively skewed within clades , we log-transformed both variables prior to analysis . For each clade we fit a quadratic model simplifying to a linear and then null model on the basis of an F-test . We aggregated the results of these models into five broad classes ( Figure S1 ) . Linear models provide three classes: no relationship ( Type 1 ) and increasing ( Type 2 ) and decreasing ( Type 3 ) range size with species age . Within the range of the clade age data , some quadratic models show marked intermediate peaks ( Type 4 ) or troughs ( Type 5 ) . However , where the vertex of the quadratic model is not central within the clade age data , quadratic models may describe predominantly increasing ( Type 2 ) and decreasing ( Type 3 ) range size with species age . We therefore assigned quadratic models to model Types 2–5 based on the location of the model vertex within terciles of the clade age data ( Figure S1; Table 1 ) . Our results are qualitatively similar if these classes of quadratic model are kept separate from linear models ( Figures S1 and S4; Table S1 ) . Data on the occupancy trajectories of extinct Cenozoic marine mollusks of New Zealand were obtained from Table S2 in Foote et al . [7] . The underlying data ( from the Fossil Record Electronic Database; http://www . fred . org . nz ) contains species records for many collection sites dated to geological stages: species range size was estimated as the proportion of collections within each geological stage that contained a species . To reduce sampling biases and to restrict analyses to extinct species , Foote et al . [7] restricted their dataset to species recorded from at least three contiguous stages and excluded all species with a Holocene record . In total , 140 species were used in the Foote et al . [7] analysis , and we refer the reader to their paper for further details on the methods . For each species we split its duration into thirds and calculated the mean occupancy across the stages occurring within each of these three time bins . Because occupancy is given in discrete stages , these cannot always be divided evenly amongst the larger time bins ( e . g . , when the species is present in four , five , or seven time stages ) . In these cases we sought the most equitable distribution of bin lengths and randomly assigned these to each bin . For instance , a species with a duration of four time stages could have bin lengths of ( 1 , 1 , 2 ) , ( 1 , 2 , 1 ) , or ( 2 , 1 , 1 ) . We repeated this procedure 1 , 000 times and calculated the mean percentage of species whose occupancy peaked in each time bin . | The changing distribution of species over the surface of the Earth , likened by Darwin to “a grand game of chess with the World for a board” , is a central theme in the evolution of life . Studies of the fossil record and molecular estimates of species ages appear to show that species follow a predictable trend of expansion and contraction of their geographic range over evolutionary time . However , using computer simulations we show that the observed correlation between the age of a species and the extent of its range is expected even if changes in range area have occurred randomly through time . Our results cast doubt on the evidence for directional trends in the long-term evolution of species ranges and suggest that the general rules governing this grand game may need to be re-evaluated . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"ecology",
"paleontology",
"biology",
"evolutionary",
"biology"
] | 2012 | Speciation and Extinction Drive the Appearance of Directional Range Size Evolution in Phylogenies and the Fossil Record |
In the city of Salvador , a large urban centre in Northeast Brazil , a city-wide sanitation intervention started in 1997 , aimed at improving the sewerage coverage of households from 26% to 80% . Our aim was to study the impact of the intervention on the prevalence and incidence of geohelminths in the school-aged population . The study comprised two comparable cohorts: the first assembled in 1997 , before the intervention , and the second assembled in 2003 , after the intervention . Both were sampled from 24 sentinel areas chosen to represent the different environmental conditions throughout the city . Copro-parasitological examinations were carried out on every individual from both cohorts , at baseline and nine months later . Demographic , socio-economic , and environmental data were collected using semi-structured questionnaires and environmental surveys . A hierarchical modelling approach fitting a sequence of Poisson multivariate linear models was undertaken to test the effect of the intervention variables on the prevalence and incidence rate ratios . 729 and 890 children aged 7–14 years ( mean = 10 . 4 y , SD = 0 . 05 y ) were analysed over the first and the second cohorts , respectively . The adjusted reductions of the prevalence and incidence rates at the second in relation to the first cohort were 27% and 34% , 25% and 32% , 33% and 26% , and 82% and 42% for geohelminths overall , Ascaris lumbricoides , Trichuris trichiura , and hookworm , respectively . Hierarchical modelling showed that a major part of each of these reductions was explained by the intervention . Our results show that a city-wide sanitation program may reduce significantly the prevalence and incidence of geohelminths .
Soil-transmitted helminths , or geohelminths , form one of the most important groups of infectious agents and are the cause of serious global health problems; more than a billion people are currently infected by at least one species of this group of pathogens [1] . At a global level , the most important geohelminths are roundworms ( Ascaris lumbricoides ) , whipworms ( Trichuris trichiura ) , and hookworms ( Necator americanus and Ancylostoma duodenale ) ; it is estimated that , respectively , these parasites have infected 1 . 2 billion , 800 million , and 740 million people [1] . In Brazil alone it is estimated that 41 . 7 million people are infected with A . lumbricoides , 18 . 9 million with T . trichiura , and 32 . 3 million with hookworms [2] . Geohelminths are more frequently found among children living in conditions of poor sanitation , and their impact on morbidity and mortality is more severe in malnourished populations [3] . Most studies suggest that approximately 70% of the worm population is hosted by 15% of the human host population . These few , heavily infected individuals are at a higher risk of disease and are also the prime source of environmental contamination [4] . Inadequate hygiene , poor health care systems and facilities , and social indifference make this situation worse . However , geohelminth control is often neglected , even in highly worm-infested countries . Geohelminths usually coinfect the host . Especially children living in deprived environments from less-developed areas may be chronically infected with more than one worm [5] , [6] . Such children have increased risks of malnutrition , stunted growth , mental retardation , and cognitive and learning deficiencies [1] . Large-scale environmental sanitation programs are complex , making interventions directly aimed at the transmission of geohelminths a challenge [7] . These interventions directly affect the transmission of several diseases in both the public and private domains [8] . Several factors should be considered for such an intervention to be successful . Amongst these are public investment in sewerage networks which must be matched by individual households' willingness to invest in a toilet and connect it to this network [7] . There are no studies of the health effects of sanitation intervention programs in large cities of developing countries . An extensive program of environmental sanitation was conducted in the Brazilian city of Salvador , Bahia aimed at expanding the city's sanitation network from 26% to 80% . A marked reduction in the rate of childhood diarrhoea has already been reported since the sanitation intervention program [7] . In this study , our aim was to report the impact of this environmental sanitation program on the prevalence and incidence of geohelminths among school aged 7–14 years .
The study was conducted in the city of Salvador , in the state of Bahia , Brazil . Bahia has an estimated population of 2 . 8 million inhabitants as of 2007 [9] . Before the intervention approximately 26% of the households was connected to the city sewage system , and it was thought that the remainder used alternative methods of sanitation ( such as septic tanks ) or simply disposed of their waste on the street . The program started in the mid-1990s with the objective to increase the sewerage coverage to 80% of households . The evaluation of the impact of the program , known as Bahia Azul ( Blue Bay ) , on the occurrence of intestinal parasites in children and teenagers of school age ( 7–14 years ) took place in two phases: the first , from 1997 to 1998 , involved the collection of pre-intervention data , and at the second , from 2003 to 2004 , involved the collection of post-intervention data by which time over 60% of residences were connected to the sewage network . In each case the studies were of a longitudinal design with two cohorts in 1997–1998 and 2003–2004 . The procedure for choosing the areas studied ( sentinel areas ) has been described in detail elsewhere [7] , [10] . These areas were selected to represent the poor , unsewered part of the city , which , before introduction of the sanitation programme in 1997 ( the intervention ) , represented about 75% of the population . Each sentinel area represented about 600 households . A sample of households with children and adolescents aged 7–14 years was randomly selected from a census of each sentinel area , and only one eligible child per household was randomly chosen to be enrolled in the investigation . The sample was stratified proportional to the number of school-aged children present in each of the sentinel areas . In both cohorts the populations used were in the same age range ( 7–14 years at baseline ) and the same methods were used for data collection . Demographic , socio-economic , environmental , and sanitary information was collected by appropriately trained field-workers using semi-structured questionnaires given to parents or guardians of the schoolchildren . Environmental inquiries conducted in 1997 and 2004 allowed for the definition of contextual variables in each sentinel area [11] . Samples of faeces were collected twice from each studied individual from both cohorts at baseline and approximately nine months later . The copro-parasitological methods of spontaneous sedimentation [12] were used to identify eggs , protozoan cysts , and the Kato Katz method [13] for quantification of helminth eggs . After each examination , the children were treated for any parasitic infection . Prevalence and incidence rates ( PR and IR ) were calculated for each geohelminth . To calculate the incidence in each cohort , those shown to be positive at the second examination but negative at the first were divided by the individuals negative at the first examination and re-examined at the second examination . Prevalence and incidence rate ratios were constructed by dividing the rate ( of prevalence or incidence ) of the second cohort ( 2003–4 ) by the corresponding rate in the first cohort ( 1997–8 ) . The analyses comprised univariate and multivariate analyses . A hierarchical modelling [14] approach involved fitting a sequence of hierarchical Poisson regression multivariable log-linear models to test the effect of the intervention variables on the prevalence and incidence rate ratios . The PR and IR were obtained from the hierarchical model as the Poisson regression coefficient comparing post- versus pre-intervention periods , with standard errors calculated from a robust covariance matrix . A conceptual model ( Figure 1 ) presents the confounding variables and the variables related to the intervention ( mediating factors ) . It is worth noting that the Bahia Azul program , besides its main intervention—sanitation—had complementary actions on water supply , garbage collection , and rainwater drainage . Based on this model , variables were selected and measured in the pre- and post-intervention studies . These variables were: the proportion of households connected to the Bahia Azul sewage system , the proportion of households with a regular water supply , the proportion of households connected to the drainage system , and the percentage of households without points of sewage . All four variables presented significant changes in between the two studies ( Table 1 ) . Confounding variables were included in the regression equations in stages to measure the effect of the intervention on geohelminths . Confounding variables not associated with the program that were included were: sex , age of the child , maternal educational level , street paving , number of children <5 years of age in household , and the presence of some type of sewage system in 1997 . Cluster adjustment to account for the 24 sentinel areas was also added to the analysis . Attributable fraction ( AF ) was also estimated; this is the proportion of reduction that can be attributed to the intervention variables . The data were analysed using the STATA ( ver . 9 . 0 ) statistical program . After each parasitological examination , the children found positive for the investigated geohelminths received appropriate treatment or were directed to a health service . Written informed consent to participate in the study was obtained from the children's parents or guardians , according to the study protocol approved by the Research Ethics Committee of the Federal University of Bahia ( UFBA ) .
A total 1 , 619 children were studied , 729 in the pre-intervention ( 1997–8 ) and 890 in the post-intervention cohort ( 2003–4; “baseline” ) . Respectively 390 and 356 tested positive for some geohelminth at baseline . The overall prevalence of all three studied geohelminths was 53 . 5% in the pre-intervention group and 40 . 0% in the post-intervention group ( p<0 . 001 ) ( Table 2 ) . Regarding the prevalence of each geohelminth , in 1997 , 42 . 9% were positive for T . trichiura , 33 . 1% for A . lumbricoides , and 9 . 9% for , hookworms . In 2003 , prevalences were 28 . 8% for T . trichiura , 25 . 5% A . lumbricoides , and 1 . 7% hookworms . All parasite infections decreased statistically significantly between the two periods . The overall incidence for all three geohelminths in the pre-intervention period ( 1997–8 ) was 36 . 0% and in the post-intervention period ( 2003–4 ) , 25 . 7% , a statistically significant difference . The parasites with the highest incidence both in 1997–8 and in 2003–4 were A . lumbricoides and T . trichiura , although only A . lumbricoides showed a statistically significant reduction after the intervention . The crude prevalence rate ratio ( PR ) observed for the three geohelminths together was 0 . 74 ( 95% confidence interval [CI] , 0 . 57–0 . 89 ) and the prevalence rate ratio adjusted for confounders ( age , sex , street paving , number of children <5 years in household , maternal educational level , and presence of sewage system in 1997 ) was 0 . 73 ( 95% CI , 0 . 66–0 . 81 ) . Therefore , a reduction of 27% was seen in the overall prevalence of geohelminths following the sanitation program ( Table 3 ) . The mediating variables of the intervention were introduced in stages ( Table 3 ) . The presence of a drainage system for rainwater and the absence of open refuse collection points had a small impact on the size of AF , showing variations of 3 . 7% and −3 . 7% , respectively . When the frequency of water supply variable was added , the AF was 14 . 8%; however , the greatest proportion of variation occurred with the introduction of the variable for the proportion of household connected to the Bahia Azul sewage system , with an AF of 85 . 2% . When all the variables , including both the confounders and the intervention variables , were added to the model , the decrease in the overall geohelminth prevalence was fully explained . Considering each geohelminth species , the adjusted prevalence rate ratio for A . lumbricoides , T . trichiura , and hookworms showed reductions of 25% , 33% , and 82% , respectively . When we analyse the intervention variables and the proportion of reduction attributable to the intervention variable ( AF ) , we can demonstrate that for A . lumbricoides the variable for frequency of water supply accounted for 24% of the observed reduction , and the proportion of households connected to the Bahia Azul sewage system variable accounted for 100% of the observed reduction . For T . trichiura the variable for frequency of water supply accounted for 12 . 1% of the observed reduction , and the proportion of household connections to the Bahia Azul sewage system accounted for 48 . 5% of the observed reduction . When all variables were combined , 100% of the reduction of T . trichiura was explained . For hookworms , the effect of the sewage was the lowest observed , given that the proportion of household connections to the Bahia Azul sewage system accounted for 6 . 1% of AF , and that all combined intervention variables achieved an AF of 23 . 2% ( Table 3 ) . For the three geohelminths , the crude incidence rate ratio ( IR ) between the periods analysed ( 2003–4 and 1997–8 ) was 0 . 71 ( 95% CI = 0 . 57–0 . 89 ) and the IR adjusted by the confounders was 0 . 66 ( 95% CI = 0 . 55–0 . 79 ) with a 34% reduction between the periods ( Table 4 ) . When the same model with mediating variables was applied to the incidence data ( Table 4 ) , the presence of rainwater drainage systems accounted for 5 . 9% of the reduction , frequency of water supply accounted for 14 . 7% , the proportion of household connections to the Bahia Azul sewage system accounted for 38 . 2% , and the combination of all intervention variables accounted for 100% of the reduction . The same model was also applied to each geohelminth species using incidence data which gave an adjusted IR for A . lumbricoides of 0 . 68 ( 95% CI = 0 . 51–0 . 90 ) signifying a reduction of 32% between the periods . The observed reduction in the variable presence of rainwater drainage accounted for 6 . 2% , the variable frequency of water supply accounted for 9 . 4% , the proportion of household connections to the Bahia Azul sewage system for 44% , and all the variables of the model accounted for 100% of the reduction . For T . trichiura a statistical difference in the crude or in the adjusted was not shown , however , in combination the AF observed for the intervention variables was 80 . 8% . For hookworms an adjusted IR of 0 . 58 ( 95% CI = 0 . 35–0 . 93 ) , meaning a reduction of 42% , was observed; when the AF of the intervention variables was analysed , the AF of the variable frequency of water supply was 14 . 3% and the variable proportion of household connections to the Bahia Azul sewage system accounted for 100% of the observed reduction . Because of the small number of cases the model with all intervention variables was not estimated ( occurrence of overlap ) .
Our results show that a city-wide sanitation program may significantly reduce the prevalence and incidence of geohelminths infections . After controlling for potential confounders , the observed reductions in the prevalence and incidence adjusted rates at the post-intervention cohort in relation to the pre-intervention one were 27% and 34% for all geohelminths , 25% and 32% for A . lumbricoides , 33% and 26% for T . trichiura , and 82% and 42% for hookworms . With the hierarchical modelling , it was observed that a major part of each of these reductions was explained by the sanitation intervention . High prevalences of A . lumbricoides and T . trichiura have been found in the Salvador population over several decades , showing that geohelminths are highly endemic in this city . Faria et al . [15] , in the 1960s , reported prevalence rates in schoolchildren from public schools of 76 . 5% , 97 . 8% , and 36 . 2% for A . lumbricoides , T . trichiura , and hookworms , respectively . Moraes et al . [16] , two decades later , in a population of the same age range in locations lacking any sanitary infrastructure , found prevalence rates for the same helminth species of 66 . 4% , 87 . 8% , and 25 . 2% . Similarly high prevalence of geohelminths has been found in children resident in periurban areas within other developing countries , all of which exhibit deficiencies in environmental sanitation [17]–[19] . In 1997 , our results of the pre-intervention period showed an overall prevalence for the three studied geohelminths of 53 . 5% , with 33 . 1% for A . lumbricoides , 42 . 9% for T . trichiura , and 9 , 9% for hookworms . A few years later , when the Bahia Azul sanitation intervention had taken place , the overall prevalence as well as the prevalence of each individual geohelminth showed significant reductions . In contrast to previous studies published in the literature , the present study's use of similar methodologies and the potential confounders to compare the results in the pre- and post-intervention periods permitted the inference that there were effective reductions in the prevalence and incidence of geohelminths in the course of the sanitation intervention . In order to analyse the extent to which the intervention contributed to these reductions , we used a hierarchical modelling strategy and AFs were estimated . For the three geohelminths together , 85 . 2% of the reduction was due to the rise in the proportion of households connected to the new sewage system that occurred in the period between the two surveys . It was also responsible for 100% , 48 . 5% , and 6 . 1% of the observed reduction in the prevalence rates of A . lumbricoides , T . trichiura , and hookworms , respectively . When this variable was combined with other variables related with the intervention ( the presence of drainage water system , absence of open refuse collection points , frequency of water supply ) and confounders , the observed reductions in prevalence rates were fully explained for geohelminths overall , A . lumbricoides , and T . trichiura , but not for hookworms ( only 23% of the prevalence reduction was explained ) ( model F; see legends of Tables 3 and 4 ) . In contrast to A . lumbricoides and T . trichiura , the mechanism of transmission of hookworms is centred in the peri-domestic environment . The efficiency of hookworm transmission increases when the infective stage ( L3 larva ) find a humid environment with high temperatures , substantial rainfall , and sandy soil . Other unmeasured environmental factors [3] , [20] and other interventions [21] could have interfered with the drastic reduction of prevalence of these parasites adequately estimate this specific type of transmission . The scarcity of data on geohelminth incidence makes difficult any comparison with the incidence rates found in pre- and post-intervention cohorts in Salvador . It is known that the level of prevalence of geohelminths is the cumulative effect of the level of incidence over time . Besides the environmental factors , prevalence could also be affected by non-environmental factors such as chemotherapy , widely used in their treatment and control [5] , [22] , [23] . However , the incidence , in contrast to the prevalence , is much more dependent on environmental factors . Consequently , the effect of an environmental intervention is best measured by its effect on the incidence . Between pre- and post-intervention periods there were important reductions in the incidence rates for geohelminths overall and for each specific geohelminth studied . In our models an important part of this reduction could be attributed to the rise in the proportion of households connected to the new sewage system . This was fully explained by this variable , and the other three variables related with the intervention ( model F ) . Our results are a clear demonstration that changes in the urban environment , particularly those associated with sewage sanitation , affect the population's health by reducing the prevalence and incidence rates of geohelminths infections . It has already been shown that the implementation of this sanitation program was followed by a reduction of 22% in the prevalence of diarrhoea in pre-school children , and this reduction was fully explained by the intervention [7] . An adequate sewage system is a sustainable strategy of disease control , bringing various benefits to the health of the population [24] . Public investments in sanitation are essential to protect individuals from open air effluents or effluents running in the streets and to control geohelminths and other sanitation-related infectious diseases [7] , [25] . The importance of basic sanitation is incontestable . According to UNICEF and WHO [26] , basic sanitation can prevent 1 . 5 billion children from dying of diarrhoea-related diseases and protect the health of millions of people . Even today , throughout the world , 2 . 6 billion people ( including one billion children ) do not have access to sanitation , meaning that only 62% of the world's population has access to sanitation infrastructure that allows for the adequate disposal of human excrement . | In the city of Salvador , a large urban centre in Northeast Brazil , a city-wide sanitation intervention started in 1997 , aiming to improve the sewerage coverage of households from 26% to 80% . Our aim was to study the impact of the intervention on the prevalence and incidence of geohelminths in the school-aged population . The longitudinal study comprised two cohorts: from the beginning of 1997 to 1998 , where data was collected before the intervention , and at the end of 2003 to 2004 , after the intervention . Copro-parasitological examinations were carried out on every individual from both cohorts , at the start and nine months later . Demographic , socio-economic , and environmental data were collected using semi-structured questionnaires . The variables utilized to demonstrate the effects of intervention , when utilized together in a multivariate model , accounted for a 100% observed reduction in the prevalence ratio ( PR ) and incidence ratio ( IR ) . As well as proving that the variables associated with the effect of the program intervention were mediators in this reduction , the reduction in the PR and IR between these periods demonstrated that modifications to the urban environment , particularly those associated with sanitary sewage systems , affected the health of the population , significantly reducing the prevalence of geohelminths . | [
"Abstract",
"Introduction",
"Methodology",
"Results",
"Discussion"
] | [
"infectious",
"diseases/epidemiology",
"and",
"control",
"of",
"infectious",
"diseases"
] | 2010 | Reductions in the Prevalence and Incidence of Geohelminth Infections following a City-wide Sanitation Program in a Brazilian Urban Centre |
Implementation of ecosystem-based fisheries management ( EBFM ) requires a clear conceptual and quantitative framework for assessing how different harvest options can modify benefits to ecosystem and human beneficiaries . We address this social-ecological need for Pacific salmon fisheries , which are economically valuable but intercept much of the annual pulse of nutrient subsidies that salmon provide to terrestrial and aquatic food webs . We used grizzly bears , vectors of salmon nutrients and animals with densities strongly coupled to salmon abundance , as surrogates for “salmon ecosystem” function . Combining salmon biomass and stock-recruitment data with stable isotope analysis , we assess potential tradeoffs between fishery yields and bear population densities for six sockeye salmon stocks in Bristol Bay , Alaska , and British Columbia ( BC ) , Canada . For the coastal stocks , we find that both bear densities and fishery yields would increase substantially if ecosystem allocations of salmon increase from currently applied lower to upper goals and beyond . This aligning of benefits comes at a potential cost , however , with the possibility of forgoing harvests in low productivity years . In contrast , we detect acute tradeoffs between bear densities and fishery yields in interior stocks within the Fraser River , BC , where biomass from other salmon species is low . There , increasing salmon allocations to ecosystems would benefit threatened bear populations at the cost of reduced long-term yields . To resolve this conflict , we propose an EBFM goal that values fisheries and bears ( and by extension , the ecosystem ) equally . At such targets , ecosystem benefits are unexpectedly large compared with losses in fishery yields . To explore other management options , we generate tradeoff curves that provide stock-specific accounting of the expected loss to fishers and gain to bears as more salmon escape the fishery . Our approach , modified to suit multiple scenarios , provides a generalizable method to resolve conflicts over shared resources in other systems .
Due to the impacts of fisheries on non-target species and ecological processes , there is growing pressure to apply ecosystem-based fisheries management ( EBFM ) [1]–[4] . Guiding principles exist , but EBFM cannot be implemented without quantitative methods that can guide policy . Additionally , designing EBFM approaches requires an assessment of the tradeoffs inherent to balancing ecosystem protection and economic costs . This is because any EBFM plan , however technically robust , requires political will . Confronting these challenges requires a new focus on case studies that account for the unique biology of each fishery and from which general guidance might emerge for other systems . Pacific salmon ( Oncorhynchus spp . ) are economically , socio-culturally , and ecologically important . Alaskan landings alone surpass 300 , 000 metric tons and ex-vessel values exceed US$260 million annually [5] . Many cultures , aboriginal and otherwise , are also tied to salmon [6] . Transcending value to humans , adult wild salmon are critical to aquatic , terrestrial , and marine ecosystem function . They are the dominant prey of a number of marine and terrestrial predators such as orcas [7] , salmon sharks [8] , pinnipeds [9] , and grizzly bears [10] . Salmon carcasses , distributed primarily by bears during spawning events , contribute annual pulses of marine-derived nutrients to freshwater systems that propagate through food webs and influence primary producers , invertebrates , fish , and wildlife [11] . The inherent conflict between the socio-economic value of salmon and their critical role in ecosystem function has led to calls for a change from current single-species management to EBFM [12] . However , such challenges have yet to lead to scientifically grounded and quantitative policy recommendations that can inform managers and fishery certifiers such as the Marine Stewardship Council ( MSC ) . One of the MSC's guiding principles is that fisheries must minimize ecosystem impacts , but it remains unclear how to quantify i ) the impact that competition with fisheries has on wildlife , ii ) the influence of modifying harvest levels on the ecosystem , iii ) or the economic costs of various management options . Selecting which organisms to monitor is also a consistent problem in the implementation of EBFM because knowledge of the relationships between biomass availability of the central resource and population responses of non-human consumers are often limited [13]–[15] . Here we cross ecosystem boundaries to use a terrestrial animal , the grizzly bear ( Ursus arctos horribilis ) , as a focal species to develop a quantitative framework that evaluates the tradeoffs between fisheries yields and an ecosystem response to salmon ( i . e . , grizzly bear densities ) . We chose grizzly bears , which are also called brown bears in coastal systems , as a surrogate of salmon-influenced ecosystem function because 1 ) bear population dynamics are strongly linked to salmon abundance [10]; 2 ) bears are the terminal predator , consuming salmon in their final life history phase; thus , if there are enough salmon to sustain healthy bear densities , we reason that there should be sufficient salmon numbers to sustain populations of earlier salmon-life-history predators such as seabirds , pinnipeds , and sharks ( Figure 1A and 1B ) ; and 3 ) bears are the dominant species mediating the flow of salmon-derived nutrients from the ocean to the terrestrial ecosystem ( Figure 1B ) [16] . After capturing salmon in estuaries and streams , grizzly bears typically move to land to consume each fish , distributing carcass remains to vertebrate and invertebrate scavengers up to several hundred meters from waterways [17] , [18] . Carcass remains ( nutrients and energy ) can influence all trophic levels from primary producers to large carnivores in both terrestrial and aquatic ecosystems [16] , [19] , [20] . Described as a “keystone interaction” , this coupled grizzly-salmon association ( at high bear densities ) can provide up to a quarter of the nitrogen budget to plant communities in riparian areas adjacent to spawning grounds [19] . Additional benefits provided by a focus on grizzly bears are their charismatic appeal to the public and their status as a large carnivore commonly of conservation concern . The fundamental challenge with implementing EBFM in this bear-salmon-human system ( and others ) is to determine how much of the fished resource to allocate to fisheries versus the ecosystem . Currently , under single-species management , fisheries commonly intercept more than 50% of inbound salmon that would otherwise be available to bears and the terrestrial and aquatic ecosystems they support [6] . Managers , typically focused exclusively on prioritizing allocation to fisheries , determine an optimum number of the total salmon run to allocate to spawning , or “escapement” . The goal is generally to achieve maximum sustainable yield ( MSY ) , but the political process , uncertainty in the relationship between spawning stock ( escapement ) and recruitment , and multiple management objectives can result in escapement goals below an estimated MSY level ( see below ) . For fisheries like this , managed below MSY , both yield and bear density would increase with greater escapement , but the potential responses have not been explored quantitatively . For those managed at MSY , increased escapement would benefit grizzly bears ( and the ecosystem ) , but costs would be borne by fishers via losses in yield . The precise tradeoffs , however , require a detailed quantitative assessment over a range of managed escapements to be of maximum value to decision-makers faced with this potentially contentious change to salmon management . To evaluate the effects of different management options , we modeled how bear population densities and fisheries yields would respond to increased escapement . This involved first estimating a relationship between salmon biomass availability and salmon consumption by bears from 18 grizzly bear populations across British Columbia ( BC ) , Canada ( Figure 1C and 1D ) . We linked this relationship to a known positive relationship between meat ( i . e . , salmon ) consumption by grizzlies and grizzly densities [10] , [21] . We then used stock-recruitment models , specific to sockeye salmon ( O . nerka ) stocks that spawn in Bristol Bay , Alaska , and BC ( Figure 2 ) , to estimate fisheries yields as a function of escapement , and the expected abundance of salmon in the absence of the fishery ( Figure 3 ) . For stocks managed below a MSY escapement , we assessed how departures from status quo management would increase bear densities and fisheries yields . For stocks managed at MSY , we scaled bear density and fishery yield by their system-specific maxima to create dimensionless and commensurate values that could be compared . In all assessments , we focused on sockeye while holding other salmonids at their management escapement targets , or mean escapement levels , because sockeye i ) are often dominant runs , ii ) migrate deep into interior regions , iii ) are the most commercially valuable species [6] , and iv ) are species for which high quality stock-recruitment data exist . While this work aims to develop a new conceptual and quantitative framework applicable to other resource management contexts , we also seek to inform contemporary bear and salmon management in BC and Alaska . First , we model potential population responses by grizzly bears in the Fraser River watershed , where bears are provincially threatened in the Chilko and partially extirpated in the Quesnel system ( Figure 2 ) . Second , we assess whether competition with the salmon fishery has the potential to significantly constrain grizzly bear productivity . This is particularly relevant because both the Fraser River and Bristol Bay stocks are certified by the MSC , having satisfied the minimal ecosystem impact principle .
In all systems , bear diets would respond considerably to increases in salmon abundance ( i . e . , escapement ) . Despite the myriad potential errors in estimating both variables across such large spatial scales , we found that salmon biomass availability alone explained nearly 50% of the variation in bear diets ( % salmon in diet ) , which followed a saturating trend ( Figure 1C ) . The relative accessibility of salmon that spawn in varied habitats , from small streams to rivers to lakeshores , likely explains some of the additional variability . Statistically fitting this relationship to 18 grizzly bear populations accounted for errors to produce a robust estimate of the relationship between salmon availability and salmon in bear diets . We estimated that the salmon biomass density necessary for salmon to constitute roughly 45% of bear diets ( half of the recorded maximum salmon consumption by bears; see Materials and Methods ) is 80 . 08 kg/km2 , with a 95% confidence interval from 50 . 9 to 128 . 4 kg/km2 . This population scale model was robust at other scales , accurately predicting bear diets at the watershed scale for three systems with known salmon biomass ( Figure 1D; Table S1 ) . This model , which predicts how percent salmon in bear diets responds to increased salmon escapements , helps explain corresponding increases in bear densities ( see below; Materials and Methods ) . Increased escapements relative to current management levels would also affect long-term fisheries yields , though patterns differ among systems . By fitting stock-recruitment relationships for each fishery , we identified three qualitatively distinct types of sockeye management dynamics ( Figure 3 ) . The Chilko and Quesnel stocks ( Fraser River ) exhibit clear overcompensating density dependence ( when recruitment declines as the number of spawners increases ) . For these stocks , both the escapement that produces MSY , EMSY , and the escapement in the absence of a fishery , Em , could be reasonably estimated . These fisheries are currently managed at MSY ( Figure 3 ) . The Ugashik and Nushagak stocks are data poor in the upper regions of escapement , making Em difficult to estimate , but reasonable estimates of EMSY are possible . These systems are managed for lower and upper escapement goals , which are both below an estimated EMSY . Finally , the Egegik and Rivers Inlet stocks have the highest uncertainty because it is unclear if the stock-recruitment relationship is even appropriate to characterize the data . Recruitment in the Egegik stock does not saturate over the observed range of escapement , which is strong evidence that escapement goals could increase to reach EMSY . Similarly , management here occurs with lower and upper escapement goals , both below predicted EMSY . Rivers Inlet is uncertain because after a period of high productivity , the stock has collapsed and is slowly rebuilding , which raises the possibility that unobserved factors ( e . g . , changing productivity due to a regime shift ) are driving recruitment dynamics [22] . Rather than consider upper and lower escapement goals for this stock in our analyses , we consider the escapement above which fishing is currently allowed and the optimal ( and higher ) escapement target estimated from a lake productivity model [22] . Although fishery yields are difficult to assess when there is high uncertainty in the stock-recruitment relationship , the impact of increasing escapement on bear densities can still be assessed . We found that the presence and degree of conflict between fisheries yields and bear densities is stock-specific . Increasing escapement from lower to upper management targets in Rivers Inlet and the Alaskan systems would increase not only bear densities , but also fisheries yields ( Figure 4A and 4B ) . Compared with the lower goals , the upper escapement goals of Ugashik , Egegik , Nushagak , and Rivers Inlet are expected to provide for roughly 22% , 8% , 8% , and 28% increases in bear density , respectively; if escapements were to increase from the lower goals to the estimated EMSY levels , bear density would increase by roughly 34% , 19% , 8% , and 44% , respectively ( Figure 4B ) . Notably , expected increases in yield are proportionately much greater than increases in bear densities ( Figure 4B ) . For stocks with predictable stock-recruitment relationships and overcompensating density dependence ( Chilko and Quesnel ) , we detect conflict between benefits to bears and benefits to fisheries . Across a range of escapements , expected fishery yields increase until escapements produce MSY and decline thereafter ( relative fisheries yield [RFY] line in Figure 5A ) . In contrast , predicted bear densities increase monotonically and saturate as escapements increase ( relative bear density [RBD] line in Figure 5A ) . In these interior systems of the Fraser River , where species other than sockeye contribute relatively little to total available salmon biomass , realizable bear densities are highly dependent on sockeye escapement ( y-intercept of RBD in Figure 5A ) . Increasing escapement beyond EMSY leads to conflict between fishery yields and bear density , with the former decreasing and the latter increasing . To aid in resolving such conflict in these systems and others , we provide here a straightforward EBFM decision-making framework . By scaling yields and bear densities relative to their maxima ( that occur at EMSY and in the absence of fishing , respectively ) , we compare the dimensionless and commensurate values of RFY [15] and RBD . When RFY and RBD are equal , which is visualized at the intersection of RFY and RBD when plotted together ( Figure 5A ) , equal relative costs are imposed on bears and fishers . We propose that this escapement level , which places equal social value to fisheries and the ecosystem , be termed “ecosystem-based management escapement” , or EEBM . Managing at EEBM , rather than at EMSY , would impose considerable costs to fisheries . Losses in long-term yield are about 12% and 23% in the Quesnel and Chilko systems , respectively ( Figure 5B and 5C ) . Based on 10-year average ex-vessel prices , lost revenues would be approximately C$680 , 000 and C$480 , 000 annually . These losses in yield would correspond to proportionally greater increases in escapement , however , nearing 50% in the Quesnel system and 80% in the Chilko run ( Figure 5C ) . These EEBM escapement levels , however , represent only one option within a continuum of ecosystem-harvest tradeoffs . We quantified these tradeoffs to assess losses in yield associated with increased bear densities as escapement varies above EMSY ( Figure 5B ) . Costs to fisheries for increasing bear densities accrue slowly at first ( low initial slope ) and then accelerate .
Our goal here was to assess quantitatively the expected impact to fisheries and grizzly bears—a surrogate for salmon ecosystem function—if status quo management was adjusted to increase escapement across a range of contexts . We present a general framework that is flexible enough to address salmon management in systems that vary in escapement targets that themselves vary as a function of certainty in stock-recruitment relationships . In low certainty systems , managed at targets below estimated EMSY , the benefits to bears ( and fisheries ) of increased escapements can be assessed , but fishery yields are too uncertain beyond this level to assess accurately the tradeoffs . In relatively high certainty systems managed for MSY , we were able to evaluate the system-specific tradeoffs between the costs to humans in lost yield and the benefits of salmon escapement to bears ( and the ecosystem ) if escapements were to increase . Any departure from current management would necessarily involve conflict between multiple competing objectives . Whereas forgoing yield for increased bear densities with escapements beyond EMSY in the Chilko and Quesnel systems represents obvious tradeoffs , others are more complex . For example , the expected increase in both bear density and fishery yield in the other four systems results in an apparent win-win situation where both the ecosystem and fisheries benefit from increasing escapement . However , high annual variability in recruitment could sometimes lead to a fishery closure if higher escapement targets committed to cannot be met . One way to avoid this is to increase upper escapement goals while retaining lower goals , which would continue to allow some fishing in low return years ( as long as lower escapement goals are met ) while allowing for increased escapement in other years . Retaining lower escapement goals may benefit subsistence fishers , who must harvest some fish each year but face restrictions if escapements are perceived to be too low . Finally , although we argue that the grizzly bear offers a sensible and attractive surrogate for salmon ecosystem function , additional ecosystem responses to different management options might instead be considered . For example , increasing net nutrient input into systems ( e . g . , [23] ) or trophic ( egg ) subsidies to resident fishes ( e . g . , [24] ) might also form reasonable and important ecosystem objectives . Similarly , minimizing the probability of years without harvests might form a desirable management objective; a quantitative evaluation of these tradeoffs might lead to very different escapement targets . In our system and others , multiple competing objectives like these increase complexity for managers , though relevant methods have been developed for decision-making ( e . g . , [25] , [26] ) . One utility of our approach is that it offers a quantitative method to evaluate how well various harvest options satisfy the MSC ecosystem criterion while accounting simultaneously for the potential economic costs to fishers . Our results suggest that low sockeye escapement is most detrimental to bears in systems where there is little biomass available from other salmon species . For example , because Nushagak has large runs of all five Pacific salmon species , salmon are expected to represent roughly 63% of bear diets even in the absence of sockeye ( Figure 4A ) . In contrast , nearly no salmon other than sockeye is available in the Quesnel run . This makes consideration of ecosystem needs in salmon management particularly important for inland stocks , where abundant runs of pink ( O . gorbuscha ) and chum ( O . keta ) salmon are absent . Moreover , in all six systems , which have received MSC certification , the observation that bear densities can increase substantially with increased escapement from current management levels implies that fisheries compete with bears and other ecosystem recipients . This suggests that the “minimal ecosystem impact” criterion , currently satisfied with certification , might in fact require increased scrutiny . This might be particularly the case with the newly certified Fraser River sockeye; grizzly bears are provincially threatened in the Chilko and partially extirpated in the Quesnel system ( Figure 2; [27] ) . Thus , the significant restrictions to bear population productivity we document as a result of conflict with fisheries are relevant to bear conservation . Another utility of our approach , particularly when applied to systems with high certainty managed at MSY , is that it offers a novel conceptual and philosophical framework of conservation value . Although arbitrary , the escapement that imposes equal costs on bears and fisheries , EEBM , can serve as a starting point to guide what are likely to be contentious management decisions . Although provocative , we highlight that this target would provide greater benefit than expected; the additional sockeye escapement to bears ( and the ecosystem ) at EEBM relative to EMSY is greater than the penalty to fishers might suggest ( Figure 5C ) . Such unexpectedly large contributions of salmon carcasses to broader ecosystem beneficiaries might form a good conservation investment . Compelling support for an “abundance matters” hypothesis is now emerging [28]; that is , while often site-specific , evidence is accumulating that suggests increased spawning density is associated with positive ecological responses across a broad array of taxa , including aquatic primary productivity [29] , terrestrial vegetation growth [30] , [31] , invertebrate density [31] , songbird density [32] , and growth rates of resident fish ( including juvenile salmon [33] ) , as well as other aquatic and terrestrial ecological processes [34] . Higher salmon escapement might also provide increased opportunities for salmon-based eco-tourism [28] . Adopting EEBM escapement goals using bears as an ecosystem surrogate has several additional desirable properties . First , implementing EEBM might be more politically robust than increasing escapements above EMSY by some arbitrary amount . Due to the saturating relationship between salmon biomass and bear density , harvests are not sacrificed in systems where bears can maintain high densities . Second , EEBM is environmentally robust . In systems with lower relative bear densities , moderate reductions in yield can translate to substantial gains for bears and ecosystems ( Figure 5C ) . Third , this model , which makes tractable the complex cross-boundary interactions between salmon nutrients and multiple beneficiaries , reflects a quantifiable ecosystem approach to management . Implementation of this method by managers can be refined with a site-specific approach relating bear diets to salmon availability across years from focal populations , rather than across populations as we have done . Finally , recognizing that EEBM might not be socio-politically possible , our tradeoff curve approach ( Figure 5B ) allows estimation of costs and benefits associated with adjustments to escapement in either direction . Applying our framework to other fisheries requires the following consideration . First , critical knowledge sets for focal non-target species should include not only their estimated population responses across a range of fish biomass , but also some distinguishing role the candidate species serves in the ecosystem ( e . g . , keystone function ) . Additionally , estimates of the costs to fisheries across a range of management options that depart from the status quo are critical . Moreover , selecting focal species of conservation concern to resource managers and the public might extend greater political will to any EBFM recommendation ( see also [35] ) . Finally , we note that the principles of single-species fisheries management and EBFM depart conceptually and practically . The former focuses narrowly and almost exclusively on the exploitation of natural resources for humans , whereas EBFM is inclusive of all biodiversity , including humans . Our proposed EBFM targets , in which costs are equally borne by fisheries and bears ( and by extension , the ecosystem ) , closely match the spirit of EBFM .
We used estimates of the proportion of salmon ( including Kokanee ) in the diet of bears from 18 grizzly bear population units ( GBPUs ) in BC , Canada , that were derived from stable isotope analysis [36] . These estimates were derived from hair , which grows throughout most of the annual activity period of bears . For these same GBPUs across the same period ( 1995–2003 ) , we estimated the mean annual salmon biomass potentially available to bears ( after interception by fisheries; the “escapement” ) . This involved using spatially explicit escapement data for all five species ( pink , chum , coho , sockeye , and Chinook ) to estimate the salmon returns in each of the watersheds captured by GBPUs ( Figure 2C ) . We assigned a portion of these estimates to GBPUs based on the fraction of each watershed that intersects each GBPU . We converted salmon numbers to biomass , using average masses of each species and sex [37] , assuming a 50∶50 ratio between sexes . To determine how the availability of salmon biomass , S ( kg/km2 ) , influenced the proportion of salmon in grizzly bear diets , D ( S ) , we fit a saturation curve using nonlinear least squares ( Equation 1 ) . Stable isotope data from grizzly bear hair sampled in the Columbia River basin , United States , during the late 1800s , when salmon were much more abundant , indicate that salmon can represent up to 90% of bear diets [38] . Several current bear populations consume more than 80% salmon [36] , but—logically—we constrained consumption to values less than 100% . Accordingly , we fixed the asymptotic maximum consumption ( i . e . , the consumption when there are infinite salmon on the landscape ) at 90% and used the data to fit the half-saturation parameter of the saturation curve . Robust estimation of the half-saturation parameter , and its confidence interval , is key because the 90% assumption will cancel in our analysis . Percent salmon in diet , D ( S ) , as a function of salmon biomass density , S ( kg/km2 ) , is given by ( 1 ) where h is the half-saturation parameter that determines how quickly bear diets respond to salmon availability . We tested the derived relationship ( Equation 1 ) at the watershed level ( as opposed to population [i . e . , GBPU] level ) using escapement data from Rivers Inlet and Quesnel ( BC ) [22] and Ugashik and Egegik ( Alaska ) [39] , [40] to estimate salmon consumption by bears ( see Tables S1 and S2 ) . The Rivers Inlet escapement and stable isotope data are from 1998 and 1999 , when salmon were relatively rare due to an extremely poor sockeye run ( Table S1 ) . Note that we estimated biomass density by summing over escapements of all salmon species . We grouped Egegik and Ugashik watersheds ( Figure 2 ) and compared predicted dietary salmon ( Ugashik: 67 . 2% , Egegik: 76 . 6% , average: 71 . 9% ) with the average estimates from stable isotope data , also from hair , collected in the associated Alaska Game Management Units 9B , 9C , and 9D ( 71% , 73% , and 73% dietary salmon , respectively , average of 72 . 3% ) [36] . Because the Quesnel sockeye run is cyclic , we used the median , rather than mean , escapement since it is a more robust approximation of inter-annual biomass availability . We determined the expected salmon harvest ( run size minus escapement ) using standard Ricker stock-recruitment models ( Figure 3 ) , which are well suited to characterize overcompensating density dependence [39] . They are also conservative in favor of fisheries because yields decline more quickly with increased escapement than if Beaverton-Holt dynamics are assumed . The size of the recruited salmon population R , when the spawning population is E , is given by ( 2 ) and yield is simply recruitment minus escapement . The escapement that maximizes long-term sustainable yield is EMSY , which we determined graphically based on the best-fit parameters . However , it is often difficult to estimate EMSY because many stock-recruitment relationships are fraught with uncertainty in parameter estimates and even uncertainty over whether the stock-recruitment relationship is appropriate to describe the dynamics of the fishery . As a result , fisheries with adequate stock-recruitment data can be managed by targeting a biologically based escapement of EMSY ( called a “biological escapement goal” ) . Other fisheries are managed between lower and upper target escapements that have provided adequate yield in the past ( called a “sustainable escapement goal” ) , but this escapement range is not necessarily optimal ( i . e . , maximizing long-term yield ) . Because our goal was to determine how departures from status quo management impact bears and ecosystems , we conducted distinct analyses for stocks managed at EMSY and those managed for a range of target escapements that were generally below estimates of EMSY as determined by stock-recruitment relationships . For fisheries managed at EMSY , the relative fishery yield ( RFY ) achieved with escapement E relative to the maximum yield is ( 3 ) which is a measure of the proportion of yield achieved by the fishery when escapement is E compared with when yields are maximized at EMSY . For fisheries managed for a range of target escapements , we used the same functional form but with the lower target escapement as our management baseline rather than EMSY ( Figure 4B ) . We consider the bear density at a particular escapement relative to the bear density at the stock-specific maximum escapement ( i . e . , no fishery ) . The escapement in the absence of the fishery , Em , is the escapement at the steady state ( i . e . , where recruitment and escapement are equal ) of the Ricker stock-recruitment model , ( 4 ) However , for fisheries without adequate certainty in stock recruitment data to estimate Em , we use the maximum observed escapement ( Figure 3 ) instead . The maximum observed escapement in these stocks is well below estimates of Em from stock-recruitment relationships , which suggests that our projections of impacts of fisheries on bear populations are conservative . We estimated the expected bear density for a given level of escapement relative to the expected bear density with the maximum escapement ( Em; Equation 4 or maximum observed escapement ) . Bear density , B , was estimated by linking Equation 1 with a known linear relationship between percent meat in diet and bear density [10] , but we assumed a zero intercept , which is conservative in favor of fisheries because some meat is likely necessary to sustain even the smallest bear density . Note that we assumed all meat consumed in coastal populations was derived from salmon , a reasonable assumption based on data from multiple populations [36] . The bear density for a given escapement is thus ( 5 ) where b0 determines how quickly bear densities increase with dietary salmon . Because bear densities increase linearly , b0 cancels when determining relative bear density so that our results depend only on the assumption of linearity and are not dependent on any particular slope from Equation 5 . Although in practice bear densities are limited by bottom-up ( i . e . , salmon ) and top-down ( i . e . , hunting ) forces , bottom-up forces influence population productivity and potential bear densities in the absence of killing by humans [41] . Percent salmon in diet ( Equation 1 ) saturates with salmon availability , ( 6 ) where ms is the mean mass ( kg ) of an individual sockeye , M is an estimate of the biomass of all other salmon species present in each system , and A is the area ( km2 ) of the watersheds that contains each salmon stock ( Table S2 ) . To estimate M , we used target escapement goals when they existed ( mean of lower and upper goal ) [39]; if not , we used average escapements from 1999 to 2008 . For runs with neither escapement targets nor data , we used harvest to approximate escapement by assuming a 50% harvest rate [40] ( see Table S2 for stock- and species-specific data sources ) . The relative bear density , RBD , can be written by combining Equations 5 and 6 as ( 7 ) Plugging Equation 6 into Equation 7 and simplifying , RBD becomes ( 8 ) The relative fisheries yield , RFY , and the relative bear density , RBD , are now both dimensionless and commensurate values that can be directly compared . For stocks with high uncertainty , EMSY and Em could not be reliably estimated . Moreover , current management practice in these systems targets a range of escapements , bounded by lower and upper goals , rather than EMSY level escapements . For these stocks we calculated percent changes in bear densities and fisheries yields when increasing from lower escapement goals to upper goals and to EMSY ( Figure 4B ) . To do this , we followed the same functional form as for RFY and RBD ( Equations 3 and 8 ) , but used the lower escapement goal as our baseline rather than Em and EMSY . Thus , rather than assess how bear densities and fishery yields compare to their system-specific maxima , we assessed how they are expected to respond to variation in the current management regime ( i . e . , from current lower to upper escapement goals ) , as well as how they are expected to respond when moving from lower escapement goals to predicted EMSY . | Commercial fisheries that harvest salmon for human consumption can end up diverting nutrients that would normally be directed to terrestrial and aquatic ecosystems . We examined this problem for Pacific salmon fisheries by using grizzly bears as indicators of salmon ecosystem function . Bear densities vary enormously depending on salmon availability , and by leaving uneaten salmon carcass remains beside spawning streams , bears play an important role in dispersing marine nutrients to plants , invertebrates , and other wildlife . By relating the number of spawning fish to bear diet and density , we developed a model to quantify “ecosystem-harvest” tradeoffs; i . e . , how bear density changes with the amount of fish harvested ( fishery yields ) . We estimated this tradeoff between yields and bear density for six sockeye salmon stocks in Alaska and British Columbia ( BC ) across a range of management options that varied the number of salmon allowed to escape from the fishery . Our model shows that bear densities will increase substantially with more spawning fish at all sites . Notably , in most study systems , fishery yields are also expected to increase as the number of spawning fish increases . There is one exception , however , in the Fraser River ( BC ) , where bears are threatened and sockeye salmon are nearly the only species of salmon available . Here , releasing more salmon to spawn would result in lower fishery yields . To resolve such conflicts in this and other systems , we propose a generalizable ecosystem-based fisheries management framework , which allows decision-makers ( such as fisheries managers and conservation scientists ) to evaluate different allocation options between fisheries and other ecosystem recipients . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"ecology",
"marine",
"and",
"aquatic",
"sciences",
"biology",
"population",
"biology",
"marine",
"biology"
] | 2012 | Using Grizzly Bears to Assess Harvest-Ecosystem Tradeoffs in Salmon Fisheries |
Botulinum neurotoxin serotype C ( BoNT/C ) is a neuroparalytic toxin associated with outbreaks of animal botulism , particularly in birds , and is the only BoNT known to cleave two different SNARE proteins , SNAP-25 and syntaxin . BoNT/C was shown to be a good substitute for BoNT/A1 in human dystonia therapy because of its long lasting effects and absence of neuromuscular damage . Two triple mutants of BoNT/C , namely BoNT/C S51T/R52N/N53P ( BoNT/C α-51 ) and BoNT/C L200W/M221W/I226W ( BoNT/C α-3W ) , were recently reported to selectively cleave syntaxin and have been used here to evaluate the individual contribution of SNAP-25 and syntaxin cleavage to the effect of BoNT/C in vivo . Although BoNT/C α-51 and BoNT/C α-3W toxins cleave syntaxin with similar efficiency , we unexpectedly found also cleavage of SNAP-25 , although to a lesser extent than wild type BoNT/C . Interestingly , the BoNT/C mutants exhibit reduced lethality compared to wild type toxin , a result that correlated with their residual activity against SNAP-25 . In spite of this , a local injection of BoNT/C α-51 persistently impairs neuromuscular junction activity . This is due to an initial phase in which SNAP-25 cleavage causes a complete blockade of neurotransmission , and to a second phase of incomplete impairment ascribable to syntaxin cleavage . Together , these results indicate that neuroparalysis of BoNT/C at the neuromuscular junction is due to SNAP-25 cleavage , while the proteolysis of syntaxin provides a substantial , but incomplete , neuromuscular impairment . In light of this evidence , we discuss a possible clinical use of BoNT/C α-51 as a botulinum neurotoxin endowed with a wide safety margin and a long lasting effect .
Few species of the bacterial genus Clostridium produce botulinum neurotoxins ( BoNTs ) , which cause the flaccid paralysis of botulism [1] . BoNTs are divided into at least seven different serotypes ( BoNT/A to G ) that comprise an increasing number of subtypes [1–3] . BoNTs are the most poisonous toxins known to date and display lethal doses in the low ng/kg range [4 , 5] . This remarkable potency is due to their selective action within the peripheral nervous system , most notably at the neuromuscular junction ( NMJ ) , where BoNTs inactivate the machinery responsible for neurotransmitter release , causing muscle paralysis and blockade of autonomic innervations [6] . Therefore , BoNTs are used to treat human diseases characterized by hyperactivity of peripheral nerve terminals of both the motor and autonomic nervous system [7] . This clinical use is almost exclusively restricted to BoNT/A1 as it produces the longest effect , and in very few circumstances to BoNT/B1 , mainly to overcome BoNT/A1 resistance [5 , 8] . The BoNT structure is composed of three domains that perform different functions [1]: a ) the C-terminal part harbors two binding sites for two different receptors that mediate toxin anchoring and internalization within nerve terminals [9 , 10]; b ) an intermediate domain responsible for the translocation of the catalytic domain into the cytosol of nerve terminals [11 , 12]; and c ) the N-terminal catalytic domain , termed light chain ( LC ) , which is a metalloprotease cleaving one of the three SNARE ( Soluble NSF Attachment Protein Receptors ) proteins , namely VAMP-1/2 ( vesicle-associated membrane protein 1/2 , also known as synaptobrevin-1/2 ) , SNAP-25 ( synaptosomal-associated protein of 25 kDa ) and syntaxin-1A/1B ( Stx ) [13 , 14] . These three proteins assemble into a complex , i . e . the SNARE complex , which mediates the fusion of synaptic vesicles with the presynaptic membrane [15] , and their proteolysis is directly responsible for the pathogenicity of BoNTs [1 , 6] . BoNT/B , /D , /F and /G cleave the vesicular SNARE protein VAMP-1/2 [16–19] , whereas BoNT/A , and /E cleave the plasma membrane protein SNAP-25 [20 , 21] . BoNT/C is the only toxin known to cleave two SNARE substrates , SNAP-25 and syntaxin-1A/1B , in vitro [22–24] . Each toxin cleaves its SNARE at a unique site thereby removing different portions of the respective substrates [13 , 14] . Interestingly , while BoNTs cleaving VAMP cause a paralysis of intermediate duration , the three serotypes that cleave SNAP-25 provide the shortest and the longest persistence of action [25–28]: BoNT/E removes 26 amino acids from SNAP-25 C-terminus and produces a muscle paralysis of a few days . BoNT/A and BoNT/C remove only nine and eight amino acids , respectively , and cause a paralysis that lasts for months in humans [29–31] . However , it is currently unknown whether BoNT/C cleaves syntaxins at the NMJ and to what extent this cleavage contributes to its long lasting paralysis . Recently , two BoNT/C LC mutants were reported to display selective protease activity against syntaxins [32] . These mutants offer the unique opportunity of dissecting the contribution of syntaxin and SNAP-25 cleavage to BoNT/C-induced paralysis and duration of action . Therefore , we synthesized the respective full-length BoNT/C mutants and tested their potency in vitro and in vivo . Surprisingly , we found that the two mutant toxins are much less toxic than wild type BoNT/C and their respective toxicity correlates with an unexpected residual activity against SNAP-25 . Our findings suggest that BoNT/C lethality is mainly due to SNAP-25 cleavage , while the proteolysis of syntaxin accounts for a prolonged and substantial , albeit incomplete , impairment of neuromuscular transmission .
Based on the work of Wang et al . ( 2011 ) [32] , we produced the full-length triple mutants BoNT/C S51T/R52N/N53P ( hereafter referred to as BoNT/C α-51 ) and BoNT/C L200W/M221W/I226W ( BoNT/C α-3W ) in Escherichia coli , along with wild type BoNT/C ( BoNT/C-wt ) . Amino acid substitutions are mapped in the crystal structure of BoNT/C-wt LC ( PDB 2QN0 ) and shown in S1 Fig . Mutations are concentrated either in the S1’ pocket of the LC ( L200W/M221W/I226W , red spot ) , thus likely impinging on LC-substrate recognition around the active site ( blue spot ) , or on a region outside the active site ( S51T/R52N/N53P , green spot ) , which is possibly involved in LC-SNAREs interaction [32] . Recombinant toxins were expressed in E . coli , activated into di-chain toxins by host proteases , and one-step affinity-purified using StrepTactin-sepharose matrix . The level of purification was suitable for biochemical and toxicological characterization ( S2 Fig ) . We first tested the overall functionality of recombinant wild type and mutant BoNT/C in cerebellar granular neurons ( CGNs ) , which are highly sensitive to BoNTs and provide a rapid and reliable method to assay the cleavage of SNARE proteins by western blot [28 , 33] . We used two specific antibodies that recognize both the intact and the truncated form of SNAP-25 and syntaxin-1A/1B . After 12 hours of incubation , BoNT/C-wt cleaved both syntaxin-1A/1B ( EC50 ~ 0 . 25 nM ) and SNAP-25 ( EC50 ~ 0 . 05 nM ) , the latter more efficiently ( Fig 1A ) . Importantly , the extent of cleavage was similar to that of a “natural BoNT/C” purified from Clostridium botulinum , implying that production in E . coli provides BoNT/C-wt with identical biological properties , as previously reported for other recombinant toxins [28 , 34–36] . The two mutant BoNT/C toxins cleaved syntaxin-1A/1B with similar efficiency ( EC50 ~ 0 . 5 nM ) , which was only slightly lower compared to BoNT/C-wt , indicating that the mutations do not alter the mutant’s capacity to enter neurons . Contrary to what was reported in the original paper by Wang et al . [32] , we also detected cleavage of SNAP-25 ( Fig 1A , middle and bottom panels ) , with BoNT/C α-3W being more active ( EC50 ~ 2 . 5 nM ) than BoNT/C α-51 ( EC50 > 5 nM ) indicating they are not specific for syntaxins . Notably , mutant BoNT/C toxins retained a 50-fold and >100-fold lower activity for SNAP-25 cleavage compared to BoNT/C-wt . This unexpected result may be due to the different methods used , i . e . Wang et al . virally transduced the gene encoding for mutant LCs while we exogenously added full-length toxins to the neuron culture medium allowing uptake of physiological amounts . We also noticed that a certain amount of syntaxin-1A/1B appears to be inaccessible to the three BoNT/Cs , even when toxins were used at high concentrations . This became particularly evident when the incubation time was extended to 24 hours ( S3 Fig ) . Moreover , unlike BoNT/C-wt , the amount of SNAP-25 cleaved by BoNT/C α-51 and BoNT/C α-3W did not increase significantly from 12 to 24 hours ( Fig 1A and S3 Fig middle and bottom panels ) . To further characterize the enzymatic properties of the mutated toxins , we assayed the proteolytic activity of their LCs in vitro against recombinant SNAP-25 and syntaxin1A . When applied at equal concentrations , LC/C-wt , LC/C α-51 and LC/C α-3W cleaved syntaxin to a similar extent ( Fig 1B , upper panel ) , suggesting an overall similar enzymatic efficiency . On the other hand , the activity of LC/C α-51 and LC/C α-3W against SNAP-25 ( bottom panel ) was much lower ( 30-fold and 10-fold , respectively ) than that of LC/C wt . Collectively , these experiments demonstrate that mutant BoNT/C toxins are not specific for syntaxin-1A/1B , but maintain a residual activity against SNAP-25 which is higher for BoNT/C α-3W compared to BoNT/C α-51 . To provide additional evidence on the proteolytic activity of BoNT/C mutants on SNAP-25 , we used an antibody that recognizes SNAP-25 segment 185–197 , which corresponds to the newly formed C-terminus generated upon cleavage by BoNT/A . This antibody recognizes BoNT/A-cleaved , and not intact , SNAP-25 both in vitro and in vivo ( S4 Fig ) [27 , 37 , 38] . Since BoNT/C cleaves SNAP-25 one amino acid downstream of the BoNT/A cleavage site [13 , 31] , we asked whether this antiserum would also recognize BoNT/C-cleaved SNAP-25 . The antibody recognized SNAP-25 cleaved by BoNT/C-wt as it caused an accumulation of staining like that generated by BoNT/A1 ( S4 Fig ) . Similar results were obtained upon treatment of CGNs with BoNT/C mutants ( 12 hours , 5 nM ) ( Fig 2 ) . In agreement with the western blot analysis , neurons treated with BoNT/C α-3W displayed a more prominent staining compared to those neurons treated with BoNT/C α-51 , yet less intense than that arising from BoNT/C-wt treatment ( 12 hours , 0 . 5 nM ) . Moreover , prolonging the incubation time did not significantly increase the amount of cleaved SNAP-25 ( S5 Fig ) . BoNT/C is long known to cause neurodegeneration in neuronal cultures [39–41] , and this effect can be monitored by staining neurofilament-200 ( NF200 ) , a major component of axon cytoskeleton . After 12 hours of incubation , BoNT/C-wt caused a substantial loss of NF200 and the formation of varicosities , also filled with cleaved SNAP-25 , all along the neurites ( Fig 2 ) . BoNT/C α-51 induced a very similar phenotype and caused an evident degeneration of neurons . Intriguingly , neurons treated with BoNT/C α-3W neither displayed significant signs of degeneration nor loss of NF200 . This was even more clear after 24 hours of intoxication when the detrimental effects of BoNT/C-wt and BoNT/C α-51 completely degenerated neurons ( S5 Fig ) . Together with the western blotting analyses , these results suggest that the three BoNT/C variants have different neurodegenerative activity on CGNs and that this effect may not depend on the proteolysis of SNAP-25 and syntaxin-1A/1B . Since the three toxins tested here have a similar activity against syntaxin-1A/1B and vary only for the extent of SNAP-25 cleavage , we evaluated the contribution of SNAP-25 cleavage to BoNT/C neuroparalysis . We first assessed the potency of the three toxins using the mouse phrenic nerve hemidiaphragm ( MPN ) assay . This ex vivo assay mimics the respiratory failure of botulism by intoxicating an explanted hemidiaphragm and allows for the recording of muscle contraction capacity elicited by phrenic nerve stimulation [35 , 42–45] . The addition of BoNTs to the organ bath impairs nerve-muscle transmission and causes progressive muscle neuroparalysis . The time needed to halve the initial twitch amplitude at a given concentration ( T50 ) is proportional to toxin potency [43] and can be used to provide very accurate comparisons of the activity of different BoNT preparations , including mutant toxins [35 , 46] . The black trace of Fig 3A shows the T50 values obtained for different bath concentrations of a standard BoNT/C , and provides a dose-response calibration curve ( y = 148 . 95x-0 . 2089; R2 = 0 . 9806 ) [45] which was used to compare the BoNT/C-wt and the two mutant toxins . BoNT/C-wt ( 100 pM; black diamond ) displayed a potency similar to standard BoNT/C ( Fig 3A ) , whereas to achieve a T50 value within the calibration curve , BoNT/C α-3W ( empty square ) and BoNT/C α-51 ( filled square ) were used at much larger concentrations ( 300 pM and 3000 pM , respectively ) . Their potency was calculated as 12 . 1% and 3 . 4% of the BoNT/C-wt , respectively ( Fig 3B ) . Hemidiaphragms were then stained for cleaved SNAP-25 ( Fig 3C ) . Cleaved SNAP-25 was detected in all muscles analyzed , suggesting that mutant toxins cleave SNAP-25 also at the NMJ . At the same time , we detected minute levels in hemidiaphragms treated with a concentration of BoNT/C α-51 ( 300 pM ) which caused a very slow decline in muscle strength compared to the higher concentration . These experiments suggest a strict correlation between the potency of BoNT/C variants and their capacity to cleave SNAP-25 . We next assessed the potency of wild type and mutant BoNT/C toxins in vivo using the mouse bioassay [47] , in which mice are injected intraperitoneally with different amounts of BoNTs and the dose causing death in the 50% of animals ( i . e . 1 LD50 ) is used as a parameter to estimate toxin lethality . BoNT/C-wt resulted in a LD50 = 0 . 75 ng/kg , whereas BoNT/C α-3W and BoNT/C α-51 displayed LD50 = 150 ng/kg and LD50 = 750 ng/kg , respectively ( Fig 4A and Table 1 ) . Similarly to BoNT/C-wt , both mutants induced the classical symptoms of botulism , i . e . the progressive collapse of flank musculature , ruffled fur , eye dryness , and labored breath until respiratory failure . Relative to BoNT/C-wt lethality ( Fig 4B , top panel ) , botulism onset due to BoNT/C α-3W was very rapid as mice died very quickly ( Fig 4B , middle panel ) . Conversely , BoNT/C α-51 induced botulism with a slower progression , with lethality occurring significantly later ( Fig 4B , bottom panel ) . Taken together , these observations suggest that the lethality/potency of the three toxins are linked to their capacity to cleave SNAP-25 . A fundamental feature of BoNT/C is the long lasting paralysis induced following local injection in sub-lethal amounts . Among the toxins tested to date , BoNT/A1 and BoNT/C have the longest persistence both in human and in mice [26 , 48] . Accordingly , we tested the duration of paralysis by mutant toxins upon injection in the hind limb with the Digit Abduction Score ( DAS ) [49] . In this assay , animals are scored from 0 to 4 , where 0 corresponds to normal mobility and 4 to complete block of digit abduction . One LD50 of BoNT/C-wt induced a severe paralysis within the first few days after administration , which was then progressively recovered ( Fig 5A ) . Surprisingly , one LD50 of BoNT/C α-3W induced a very quick onset followed by a complete recovery within 48 hours . Even more surprisingly , one LD50 of BoNT/C α-51 had a very long persistence , also exceeding that of BoNT/C-wt . Interestingly , we detected an acute phase ( days 1–5 ) of severe paralysis followed by a longer period of time characterized by a progressive , yet slow , recovery of function . To obtain a more quantitative time course of the paralyses , we measured the “evoked junction potential” ( EJP ) . This electrophysiological analysis allows an accurate estimation of neurotransmitter release at the NMJ and can be used to monitor neurotransmission recovery in a quantitative way [27] . Muscles treated with BoNT/C α-3W were fully paralyzed 24 hours after injection , yet recovery was fast and nerve-muscle transmission reached control levels within two weeks ( Fig 5B ) . Unlike BoNT/C α-3W , BoNT/C-wt and BoNT/C α-51 completely abrogated neurotransmission for at least one week and recovery was slow , being incomplete even after 5 weeks . Altogether , these experiments indicate that the duration of the neuroparalytic effects of BoNT/C mutants does not correlate with their respective potency . We hypothesized that lack of correlation between duration of paralysis and the relative potency of BoNT/C variants was due to different cleavage of SNAP-25 and syntaxin . To test this hypothesis , we stained soleus muscles of mice treated with wild type and mutant BoNT/C toxins for cleaved SNAP-25 . BoNT/C-wt induced long-lasting SNAP-25 cleavage , which was reduced only at a later stage , when neurotransmission has significantly recovered ( Fig 6A ) . Two weeks after toxin treatment , we also observed NMJ fragmentation , extensive nerve terminal sprouting , and synaptic remodeling , as previously reported [50] . Interestingly , cleaved SNAP-25 was detected all along sprout membranes and within the presynaptic side of newly formed nerve-muscle contacts . Following BoNT/C α-3W treatment , cleaved SNAP-25 was clearly detectable 24 hours after injection , when neurotransmission is completely blocked , but disappeared within four days , indicating that SNAP-25 1–198 has a half-life of <4 days ( Fig 7A ) . NMJs displayed neither signs of postsynaptic remodeling nor nerve sprouting . These results suggest that SNAP-25 cleavage is associated with a nerve terminal blockage and that muscle paralysis persists as long as SNAP-25 is cleaved . On the other hand , Fig 8A shows that in muscles treated with BoNT/C α-51 , cleaved SNAP-25 is evident only at 24 hours after injection and barely detectable at one and two weeks after . This indicates that SNAP-25 proteolysis occurs only at the very beginning of the time course and , consequently , the long lasting neurotransmission impairment induced by BoNT/C α-51 cannot be ascribed to SNAP-25 cleavage . Therefore , we stained injected muscles for syntaxin-1A/1B . Control NMJs displayed an intense and widespread staining characterized by large puncta of syntaxin-1B , the isoform prevalently expressed at the NMJ [51] ( Figs 6B , 7B and 8B ) . BoNT/C and mutant toxins caused a large , though incomplete , loss of signal and the complete disappearance of syntaxin-1B clusters . This effect , which we interpreted as syntaxin-1B cleavage , was prolonged for BoNT/C-wt and it was not recovered even five weeks after injection ( Fig 6B ) . In the case of BoNT/C α-3W , loss of syntaxin-1B was obvious only 24 hours after toxin injection with return to control levels between 7 to 14 days , indicating re-synthesis of syntaxin within 4 days and absence of LC/C α-3W activity ( Fig 7B ) . BoNT/C α-51 caused a loss of syntaxin-1B similar to BoNT/C-wt and even 35 days post-injection there were no signs of recovery ( Fig 8B ) . Importantly , nerve terminals underwent remodeling with a time course and an intensity comparable to the BoNT/C-wt-treated muscles , indicating a long-term impairment of NMJs . Considering that syntaxin re-synthesis is rapid ( Fig 7B ) , these results suggest that the long lasting effect of BoNT/C α-51 is due to a persistent cleavage of syntaxin-1B and that LC/C α-3W activity has a very short half-life in vivo . Moreover , it is interesting to note that syntaxin proteolysis by LC/C α-51 also occurs in a phase in which SNAP-25 is not cleaved anymore ( Fig 8B ) which might be ascribed to the >10-fold higher EC50 for SNAP-25 ( EC50Stx 0 . 5 nM vs EC50SNAP-25 >5 nM ) . Altogether , electrophysiological measurements and nerve terminal analyses indicate that the cleavage of SNAP-25 is necessary for a complete neuromuscular block and muscle paralysis , while the proteolysis of syntaxin-1B induces a significant , though incomplete , impairment of neurotransmission . Since the previous experiments showed that SNAP-25 cleavage occurs only with large amounts of mutant toxins , we reasoned that a low dose might induce exclusive proteolysis of syntaxin-1B . In this way , motor nerve activity should be attenuated without complete abrogation of neurotransmission . To test this hypothesis , we opted to use BoNT/C α-51 because of its low activity against SNAP-25 and its long half-life . Toxin was injected in the hind-limb at a dose of 10 ng/kg , which corresponds to about 1/75 of its LD50 and 13-fold LD50 of BoNT/C-wt . Interestingly , this “low dose” of BoNT/C α-51 did not result in a visible neuroparalysis ( i . e . , DAS score 0 ) . Rather BoNT/C α-51 caused a substantial decrease of EJP amplitude in injected muscles , which lasted for almost one month before returning to control levels ( Fig 9A ) . Importantly , no cleavage of SNAP-25 was detected during the entire time course ( Fig 9B ) , while a significant loss of syntaxin-1B staining occurred , especially within the first two weeks after injection ( Fig 9C ) . Thereafter , syntaxin-1B expression recovered together with neurotransmission , suggesting that nerve terminal activity ( EJP amplitude ) is proportional to the amount of syntaxin-1B . In addition , we found a reduced atrophy of muscles injected with the low dose of BoNT/C α-51 relative to muscles injected with either 1 LD50 of BoNT/C α-51 or 1 LD50 of BoNT/C-wt . These results indicate that BoNT/C α-51 can persistently modulate nerve terminal activity without compromising the overall activity of muscles .
Among the many botulinum neurotoxins characterized so far , BoNT/C is unique in that it cleaves two neuronal SNARE proteins , i . e . SNAP-25 and syntaxins . Although this parallel activity was demonstrated over 20 years ago in cultured neurons [24] , it has never been reported at the NMJs nor was it clear which proteolytic event by BoNT/C causes neuroparalysis . Our results show that BoNT/C cleaves both substrates at the NMJ , and that the key determinant of potency and lethality is the proteolysis of SNAP-25 rather than syntaxin . We also report that the cleavage of syntaxin-1B ( syntaxin-1A is not expressed at the NMJ [51] ) does not cause complete block of the NMJ , although it accounts for a substantial impairment of neurotransmission efficiency . Such a result is surprising considering the current view of the SNARE-mediated mechanism of neuroexocytosis [52] , but it is supported by previous reports showing that: i ) BoNT/C paralysis at NMJ can be reversed by 3 , 4-diaminopyridine [50] , consistent with a neuroparalytic effect lying on SNAP-25 cleavage rather than syntaxin , like for BoNT/A1 [53]; ii ) Syntaxin-1B knock out mice have a very limited life span , yet they survive for a couple of weeks after birth , implying that neuromuscular transmission is viable [54 , 55]; iii ) Syntaxin-1B deficiency reduces but does not abolish NMJ capacity of neurotransmitter release [55] . A likely explanation is the compensation of syntaxin-1B knock out/proteolysis at the NMJ by other syntaxin isoforms . In knock out mice a minor expression of syntaxin-1A was reported , which might occur as a compensatory mechanism [55] . In the present study , this possibility has to be discarded as syntaxin-1A would also be substrate of the toxins . On the other hand , many non-cleavable syntaxins exist [5] , raising the possibility that a cognate isoform compensates for syntaxin-1 biochemical knock down , leading to a largely inefficient , yet functional neurotransmitter release . As a general conclusion , syntaxin-1B proteolysis does not seem to be critical for the acute neuroparalytic action of BoNT/C , which instead relies on SNAP-25 cleavage . Nonetheless , syntaxin cleavage may contribute to delaying the recovery process . In fact , fusion of synaptic vesicles with the plasma membrane depends on the incorporation of multiple SNARE complexes into SNARE super-complexes [56] . Since cleavage by BoNT/C occurs at the very C-terminus and frees syntaxin from its transmembrane domain [5 , 13 , 14] , it may be speculated that the SNARE motif of cleaved-syntaxins are incorporated within SNARE super-complexes , and negatively modulate vesicle fusion . This effect may be added to the long known effect of BoNT/C ( and BoNT/A ) cleaved-SNAP-25 in modulating neuroexocytosis [5 , 56] . We were surprised to find that BoNT/C mutants exert different neurodegenerative effects in cultured neurons . This effect was previously attributed to cleavage of SNAP-25 and syntaxin-1A/1B , and to the complete elimination of one of them [40 , 41] . However , we found that in vitro neurodegeneration triggered by BoNT/C-wt and BoNT/C α-51 occurs even if a small portion of syntaxin-1A/1B is resistant to cleavage . Moreover , BoNT/C α-3W is not cytotoxic even if it displays proteolytic activity against SNAP-25 and syntaxin-1A/1B to an extent equal , if not superior , to BoNT/C α-51 . These results indicate that SNARE cleavage may be not directly implicated in BoNT/C-mediated neurodegeneration , at least in CGNs . An intriguing alternative explanation may be the proteolysis of an additional substrate , still recognized by BoNT/C-wt and BoNT/C α-51 , but not by BoNT/C α-3W . BoNT/C neurodegeneration does not occur at the NMJ in vivo in mice [50] and humans [48 , 57 , 58] . A plausible explanation as to why neurodegeneration occurs only in cell culture may be found in the different mode of entry of BoNT/C at the NMJ and in cultured neurons . In vivo , like the other BoNTs , BoNT/C entry is restricted to unmyelinated areas of the nerve terminals as axons are covered and protected by a tight nerve-blood-barrier . Neurons in culture are instead not myelinated and fully exposed to the action of all toxins residing in the culture medium . Accordingly , BoNT/C may affect neuronal compartments that are not accessible in vivo . In any case , we show here that poisoned nerve terminals do not degenerate at any time after toxin injection , a relevant finding considering that very high amounts of BoNT/C α-51 had been locally injected ( 1 LD50 ≥ 15 ng/mouse ) . Rather , poisoned nerve terminals activate and set in motion a profound remodeling of the NMJ as an attempt to bypass the functional block of the synapse , similarly to what is observed upon BoNT/A1 and BoNT/B1 intoxication [25 , 27 , 50 , 59] . Therefore , our results suggest that the use of BoNT/C in humans would be safe , and that BoNT/C α-51 may be a suitable candidate for this purpose . In fact , used at a dose comparable to BoNT/C-wt , this toxin provides a persistent modulation of nerve terminal activity without causing the complete paralysis of the muscle , a relevant finding for therapeutic and cosmetic applications . BoNT/C α-51 would likely be ideal for applications characterized by a narrow therapeutic window , when an optimal modulation of nerve terminal hyperactivity is usually difficult to achieve without causing significant muscle weakening , as in focal dystonia [5 , 60] . Moreover , considering the low systemic toxicity in mice , this toxin may be suitable in clinical conditions requiring considerable amounts of BoNT , like the treatment of large muscles in post-stroke spasticity [5 , 61 , 62] . As for other BoNTs [29 , 57 , 58 , 63] , electrophysiological testing on human volunteers can be used to assess the time course of action and the susceptibility of human muscles to BoNT/C α-51 ( or syntaxin-specific BoNTs ) . These preliminary analyses would reveal a dose-response window and may be essential in evaluating the therapeutic potential of the toxin as well as its safety margin and immunogenicity [5] . In conclusion , this work highlights that minimal changes can functionally impact BoNT biological activity , and suggests that inspection of structure-activity relationships may be used to generate tailor-made toxins with ad hoc pharmacological properties to improve the present applications and expand the clinical landscape of BoNT pharmacotherapy [41 , 46 , 64 , 65] .
Native BoNT/C and BoNT/A1 were purified as previously described [66 , 67] . Cytosine β-D-arabinofuranoside hydrochloride ( C6645 ) , DNAse I from bovine pancreas ( DN25 ) , poly-L-lysine hydrobromide ( P1274 ) and trypsin ( T4799 ) were from Sigma Aldrich . μ-Conotoxin GIIIB is from Alomone , Jerusalem , Israel . Primary antibodies: anti-SNAP-25 ( SMI81 , ab24737 ) was from Abcam . Anti-SNAP-25 ( cleaved ) and syntaxin-1A/1B polyclonal antibodies were produced in our laboratory and previously characterized [37 , 68] . Secondary antibodies conjugated to HRP were from Calbiochem; secondary antibodies for immunofluorescence conjugated to Alexa Fluorophores 488 or 555 and α-Bungarotoxin conjugated to Alexa 647 were from Thermo Scientific , Waltham , MA , USA . Full-length BoNT/C ( GenBank: X53751 . 1 ) and BoNT/C LC ( aa 1–430 ) as well as the mutants thereof were produced , the former under biosafety level 2 containment ( project number GAA A/Z 40654/3/57 ) , in E . coli strain M15pREP4 ( Qiagen , Hilden , Germany ) during 15 h of induction at 21°C and purified on StrepTactin-sepharose matrix ( IBA GmbH , Göttingen , Germany ) and Ni2+-nitrilotriacetic acid-agarose matrix ( Qiagen ) , respectively , according to the manufacturers’ instructions . Aliquots of BoNT/C derivatives ( in 100 mM Tris , pH 8 . 0 ) and of BoNT/C LC derivatives ( dialyzed against toxin assay buffer ( 150 mM potassium glutamate , 10 mM HEPES-KOH , pH 7 . 2 ) , were frozen in liquid nitrogen , and kept at -70°C . Recombinant substrate proteins , rat SNAP-25His6 [69] and a syntaxin fusion protein comprising an N-terminal His6-tag followed by the Halo-tag , rat syntaxin 1A aa 183–259 , luciferase , and a C-terminal Strep-tag , were produced using the E . coli strains M15 pREP4 and BL21-DE3 ( Stratagene Europe , Ebsdorfergrund , Germany ) , respectively , purified via His6- or His6- and Strep-tag , dialyzed against toxin assay buffer or PBS , pH 7 . 4 , supplemented with 7% ( w/v ) sucrose , and finally frozen in liquid nitrogen . Radiolabeled substrates were generated by in vitro transcription/translation using the plasmids pSNAP-25his6 and pET29-HASyn ( 183–259 ) LS , the SP6/T7 coupled TNT reticulocyte lysate system ( Promega ) , and [35S]methionine ( 370 KBq/μL , >37 TBq/mmol , Hartmann Analytic , Braunschweig , Germany ) according to the manufacturer´s instructions . Concentrations of E . coli expressed proteins were determined subsequent to SDS-PAGE and Coomassie blue staining by using a LAS-3000 imaging system ( Fuji Photo Film ) , the AIDA 3 . 51 program , and various known concentrations of BSA . The extent of hydrolytic activation of full-length BoNT/C by E . coli proteases was 81% ( wild type ) , 73% ( α-51 ) , and 79% ( α-3W ) . Cleavage assays were conducted using 10 μM SNAP-25 or 1 μM syntaxin fusion protein , respectively , each 1 μL of transcription/translation mixture of the respective substrate as [35S]-methionine-labeled protein , and purified LC/C derivative at 1 to 3 μM final concentrations in 10 μL . Incubation was done for 60 min at 37°C in toxin assay buffer . Reactions were stopped by the addition of an equal volume of double-concentrated sample buffer ( 120 mM Tris-HCl , pH 6 . 75 , 10% ( v/v ) β-mercaptoethanol , 4% ( w/v ) SDS , 20% ( w/v ) glycerol , 0 . 014% ( w/v ) bromphenol blue ) and then subjected to SDS-PAGE using 10% or 15% tris/glycine gels ( the latter using acrylamide/bis-acrylamide in 73 . 5:1 ratio ) . Subsequently , gels were dried and radiolabeled proteins were visualized employing a FLA-9000 phosphorimager ( Fuji Photo Film , Co . , Ltd . , Tokyo , Japan ) . Quantification of cleavage was done by means of the radiolabeled substrates by phosphorimaging using the Multigauge 3 . 2 software ( Fuji Photo Film ) . Primary cultures of rat cerebellar granule neurons ( CGNs ) were prepared from 6- to 8-day-old rats as previously described [26] . Briefly , cerebella were isolated , mechanically disrupted and trypsinized in the presence of DNase I . Cells were then collected and plated into 24 well plates pre-coated with poly-L-lysine ( 50 μg/ml ) at a cell density of 4x105 cells per well . Cultures were maintained at 37°C , 5% CO2 , 95% humidity in BME ( Basal Medium Eagle ) supplemented with 10% fetal bovine serum , 25 mM KCl , 2 mM glutamine and 50 μg/ml gentamicin ( hereafter indicated as complete culture medium ) . To arrest growth of non-neuronal cells , cytosine arabinoside ( 10 μM ) was added to the complete culture medium 18–24 h after plating . CGNs at 6–8 days in vitro ( DIV ) were incubated with increasing concentrations ( from 0 . 01 nM to 5 nM ) of the indicated BoNT/C in complete culture medium for 12 or 24 hours at 37°C . The specific proteolytic activity against SNAP-25 and syntaxin-1A/1B was evaluated via immunoblotting with antibodies that recognize both the intact and the truncated form of the two proteins . Cells were directly lysed with Laemmli sample buffer containing protease inhibitors ( Roche ) . Cell lysates were loaded onto NuPage 12% Bis-Tris gels ( Life technologies ) and separated by electrophoresis in MOPS buffer ( Life technologies ) . Proteins were transferred onto Protran nitrocellulose membranes ( Whatman ) and saturated for 1 h in PBS-T ( PBS , 0 . 1% Tween 20 ) supplemented with 5% non-fatty milk . Incubation with primary antibodies was performed overnight at 4°C . The membranes were then washed three times with PBS-T and incubated with appropriate secondary antibodies for 1 h . Membranes were washed three times with PBS and proteins revealed either with an Odyssey imaging system ( LI-COR Bioscience ) or with an Uvitec gel doc system ( Uvitec Cambridge ) . CGNs were seeded onto 13 mm round glasses in 24-well plates at a cell density of 4x105 cells per well . CGNs at 6–8 DIV were incubated for the indicated time and concentration of toxin in complete culture medium at 37°C . After treatment , neurons were fixed for 10 min with 4% ( w/v ) paraformaldehyde in PBS and stained with an antibody against cleaved SNAP-25 and an antibody against neurofilament-200 ( NF200 ) . Coverslips were mounted using Fluorescent Mounting Medium ( Dako ) and examined with a Leica SP5 confocal microscope ( Leica Microsystems , Wetzlar , Germany ) equipped with 100X HCX PL APO NA 1 . 4 objective . The MPN assay was performed as described previously [43 , 45] . To limit the consumption of mice , the left and right phrenic nerve hemidiaphragms were excised from female mice of strain RjHan:NMRI ( 18–25 g , Janvier , St Berthevin Cedex , France ) and placed in an organ bath containing 4 ml of Earle's Balanced Salt Solution . The pH was adjusted to 7 . 4 , and oxygen saturation was achieved by gassing with 95% O2 and 5% CO2 . The phrenic nerve was continuously electro-stimulated at a frequency of 1 Hz with a pulse duration of 0 . 1 ms and a current of 25 mA to achieve maximal contraction amplitudes . Isometric contractions were recorded with a force transducer ( Scaime , Annemasse , France ) and the software VitroDat ( Föhr Medical Instruments GmbH ( FMI ) , Seeheim , Germany ) . The resting tension of the hemidiaphragm was approximately 10 mN . In each experiment , the preparation was first allowed to equilibrate for 15 min under control conditions . Then , the buffer was exchanged to 4 ml of Earle's Balanced Salt Solution supplemented with 0 . 1% BSA and varying BoNT/C dilutions . The previously reported calibration curve determined for recombinant , E . coli host activated BoNT/C ( y ( BoNT/C; 15 , 50 , 70 , 100 and 233 pM ) = 148 . 95x-0 . 2089; R2 = 0 . 9806 ) was used to calculate the residual potency of BoNT/C mutants . The resulting paralytic half-times of BoNT/C mutants were converted to the corresponding concentrations of wild type BoNT/C , using the equation mentioned above . The toxicities were finally expressed relative to wild type BoNT/C . Swiss-Webster adult female CD1 mice ( 20–24 grams ) were housed under controlled light/dark conditions , and food and water were provided ad libitum . All experiments were performed in accordance with the European Community Council Directive n° 2010/63/UE and approved by the Italian Ministry of Health . LD50 were determined by injecting different doses of BoNT/C-wt or BoNT/C α-3W or BoNT/C α-51 diluted in 0 . 9% NaCl 0 . 2% gelatin . Toxins were prepared at a given concentration and mice were injected intraperitoneally with different volumes according to their body weight in order to reach the indicated doses . Mice were monitored every 4 hours for 96 hours , when the experiment was considered to be concluded . Swiss-Webster adult female CD1 mice weighing 20–24 g were injected in the left hind limb with 1xLD50 of BoNT/C-wt ( 0 . 75ng/kg ) or BoNT/C α-3W ( 150 ng/kg ) or BoNT/C α-51 ( 750 ng/kg ) diluted in 0 . 9% NaCl with 0 . 2% gelatin . Neuroparalysis was assessed daily according to the Digit Abduction Score ( DAS ) assay scale , as previously reported [49] . Mice were injected in the left hind limb as described for the DAS assay with indicated doses . At scheduled times , mice were sacrificed by anesthetic overdose coupled to cervical dislocation and the soleus muscle dissected . Electrophysiological recordings were performed in oxygenated Krebs-Ringer solution , using intracellular glass microelectrodes ( WPI ) filled with 1 M KCl and 2 M CH3COOK . Evoked junction potentials ( EJP ) were recorded in current-clamp mode , starting from resting membrane potential of -70 mV , adjusted with direct current injection if needed . EJPs were elicited by supramaximal nerve stimulation at 0 . 5 Hz , using a suction microelectrode connected to a S88 stimulator ( Grass , Warwick , RI , USA ) . Muscle contraction was prevented by 1 μM μ-Conotoxin GIIIB ( Alomone , Jerusalem , Israel ) . Signals were amplified with intracellular bridge mode amplifier ( BA-01X; NPI , Tamm , Germany ) , sampled using a digital interface ( NI PCI-6221; National Instruments , Austin , TX , USA ) and recorded by means of electrophysiological software ( WinEDR; Strathclyde University , Glasgow , Scotland , UK ) . EJPs measurements were carried out with Clampfit software ( Molecular Devices , Sunnyvale , CA , USA ) . EJPs represent the average value obtained analyzing at least three muscles ( 15 fibers/muscle ) for each condition at each time-point and reported as a percentage with respect to control muscles . Immediately after electrophysiological recording , soleus muscles were fixed in 4% paraformaldehyde in PBS for 10 min at RT . Each muscle was then separated in bundles of about 20–40 fibers to facilitate the following steps , in particular antibody penetration . In the case of explanted and intoxicated hemidiaphragms , upon completion of paralysis , muscles were fixed for 30 minutes . Samples were quenched in 50 mM NH4Cl in PBS and treated for 2 h with a blocking solution ( 15% v/v goat serum , 2% w/v BSA , 0 . 25% w/v gelatin , 0 . 2% w/v glycine in PBS , 0 . 5% Triton X-100 ) to saturate and permeabilize nerve terminals . Thereafter , incubation with the following primary antibodies was carried out for at least 48 h in blocking solution with either anti-cleaved SNAP-25 or anti-Syntaxin-1A/1B . Muscles were then extensively washed and incubated with a secondary antibody conjugated with Alexa-555 diluted in blocking solution supplemented with α-Bungarotoxin conjugated to Alexa 647 to counterstain post-synaptic nicotinic acetylcholine receptors . Images were collected with a Leica SP5 confocal microscope ( Leica Microsystems , Wetzlar , Germany ) equipped with 100X HCX PL APO NA 1 . 4 objective . Laser excitation line , power intensity , and emission range were chosen according to each fluorophore in different samples to minimize bleed-through . All experiments were performed in accordance with the Italian laws and policies ( D . L . n°26 14th March 2014 ) and with the guidelines established by the European Community Council Directive n° 2010/63/UE and approved by the veterinary services of the University of Padova ( O . P . B . A . —Organismo Preposto al Benessere degli Animali ) ( protocol 359/2015 ) . | The seven established Botulinum Neurotoxins serotypes ( BoNT/A to G ) and the many BoNT subtypes , the causative agents of botulism , are the most poisonous substances known ( lethal doses in the low ng/kg range ) . Due to their toxicological properties , BoNTs are Janus-faced toxins: potent pathogenic factors and potential bioterrorism agents as well as safe and efficacious therapeutics . BoNTs exert their neuroparalytic action by cleaving SNARE proteins , either SNAP-25 or synaptobrevin/VAMP , which mediate neurotransmitter release at the neuromuscular junction; BoNT/C is the only serotype shown to cleave SNAP-25 and syntaxin-1 in vitro . Our study shows for the first time that this parallel cleavage also occurs in vivo . By using mutated toxins reported to be syntaxin-selective , we found that SNAP-25 proteolysis at the neuromuscular junction is the key determinant of BoNT/C lethality as it completely blocks nerve-muscle transmission . Conversely , syntaxin-1 cleavage only attenuates nerve terminal activity without inactivating the synapse , leading to only a partial decrease of neuromuscular functionality . As a result , the BoNT/C mutants have dramatically reduced lethality , but still modulate neuromuscular junction activity upon intramuscular injection . This aspect is particularly relevant considering the possible use of syntaxin-specific BoNT/C derivatives to improve the present clinical utilization of BoNTs . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"soleus",
"muscles",
"medicine",
"and",
"health",
"sciences",
"toxins",
"pathology",
"and",
"laboratory",
"medicine",
"metabolic",
"processes",
"neuroscience",
"toxicology",
"toxic",
"agents",
"paralysis",
"signs",
"and",
"symptoms",
"proteolysis",
"bioassays",
"and",
"physiological",
"analysis",
"muscle",
"electrophysiology",
"research",
"and",
"analysis",
"methods",
"musculoskeletal",
"system",
"bacterial",
"toxins",
"animal",
"cells",
"proteins",
"muscles",
"electrophysiological",
"techniques",
"biochemistry",
"cellular",
"neuroscience",
"diagnostic",
"medicine",
"cell",
"biology",
"anatomy",
"neurotransmission",
"neurons",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"metabolism",
"botulinum",
"toxin"
] | 2017 | Botulinum neurotoxin C mutants reveal different effects of syntaxin or SNAP-25 proteolysis on neuromuscular transmission |
TSSA ( Trypomastigote Small Surface Antigen ) is an antigenic , adhesion molecule displayed on the surface of Trypanosoma cruzi trypomastigotes . TSSA displays substantial sequence identity to members of the TcMUC gene family , which code for the trypomastigote mucins ( tGPI-mucins ) . In addition , TSSA bears sequence polymorphisms among parasite strains; and two TSSA variants expressed as recombinant molecules ( termed TSSA-CL and TSSA-Sy ) were shown to exhibit contrasting features in their host cell binding and signaling properties . Here we used a variety of approaches to get insights into TSSA structure/function . We show that at variance with tGPI-mucins , which rely on their extensive O-glycoslylation to achieve their protective function , TSSA seems to be displayed on the trypomastigote coat as a hypo-glycosylated molecule . This has a functional correlate , as further deletion mapping experiments and cell binding assays indicated that exposition of at least two peptidic motifs is critical for the engagement of the ‘adhesive’ TSSA variant ( TSSA-CL ) with host cell surface receptor ( s ) prior to trypomastigote internalization . These motifs are not conserved in the ‘non-adhesive’ TSSA-Sy variant . We next developed transgenic lines over-expressing either TSSA variant in different parasite backgrounds . In strict accordance to recombinant protein binding data , trypomastigotes over-expressing TSSA-CL displayed improved adhesion and infectivity towards non-macrophagic cell lines as compared to those over-expressing TSSA-Sy or parental lines . These phenotypes could be specifically counteracted by exogenous addition of peptides spanning the TSSA-CL adhesion motifs . In addition , and irrespective of the TSSA variant , over-expression of this molecule leads to an enhanced trypomastigote-to-amastigote conversion , indicating a possible role of TSSA also in parasite differentiation . In this study we provided novel evidence indicating that TSSA plays an important role not only on the infectivity and differentiation of T . cruzi trypomastigotes but also on the phenotypic variability displayed by parasite strains .
Trypanosoma cruzi is the protozoan agent of Chagas disease , of major medical and economic significance throughout Latin America , and an emergent threat to global public health [1] . This parasite alternates between blood-sucking triatomine vectors and a wide spectrum of susceptible , mammalian hosts , including humans . Transitions between hosts or from different niches within hosts pose major adaptation challenges and are accordingly accompanied by a complex series of developmental changes [2] . Briefly , epimastigote forms replicate along the digestive tract of insect vectors , and eventually attach to the cuticle of the rectal epithelium where they differentiate into non-dividing and infective metacyclic trypomastigotes . These are in turn deposited on the mammal along with the insect feces during a blood meal and gain access to internal body fluids via a skin lesion or a mucosal surface . In the mammalian host , T . cruzi presents two major morphological stages: trypomastigotes , which are non-dividing and highly motile forms found in the bloodstream and in the extracellular spaces of tissues , and amastigotes , which are intracellular replicative forms . Host cell invasion by T . cruzi trypomastigotes follows a rather complex , active and multi-step process [3 , 4] . The initial recognition and sensitization of the target cell involves various apparently redundant parasite surface receptors such as MASPs ( mucin-associated surface proteins ) and Gp85 molecules [5–7] . Most of these are glycosyl phosphatidylinositol ( GPI ) -anchored glycoproteins encoded by large , polymorphic and developmentally-regulated gene families [6 , 8] . They all bear a similar predicted architecture , in which the outermost and variable N-terminal domain protrudes from the parasite glycocalix , thus ideally suited for engaging with different constituents of the host cell membrane and/or extracellular matrix . Early after invasion , trypomastigotes escape from the parasitophorous vacuole into the host cell cytoplasm where they undergo transformation into amastigote forms . The molecular and cellular mechanisms that regulate amastigogenesis remain poorly understood , although this process can be triggered and recapitulated , at least in vitro , by various factors including temperature or pH shift and starvation [9 , 10] . After several rounds of replication and just before disruption of the parasite-laden cell , amastigotes differentiate back into trypomastigotes , which may infect other cells within the host or may be taken up by the insect vector during a bloodmeal , thus closing the cycle . T . cruzi trypomastigotes are wrapped by a protective coat made up essentially of heavily O-glycosylated mucins and , underneath , glycoinositol phospholipids ( GIPLs ) [11 , 12] . Trypomastigote mucins , also known as tGPI-mucins , are encoded by the TcMUC family of genes , comprising ~800 members [13–15] . TcMUC deduced products share a common structure made up of highly conserved N-terminal signal peptide ( SP ) and cleavable C-terminal GPI attachment signal; and a variable , Thr-rich region that constitutes the ‘mature’ apo-mucin displayed on the parasite surface [13] . The latter region undergoes extensive O-glycosylation in vivo , which confers a strong hydrophilic character to the overall molecule [11 , 16] . Large variability observed for biochemically-purified tGPI-mucins may be attributed to the simultaneous expression of multiple TcMUC genes showing differences in the length , sequence and number of putative glycosylation-acceptor sites as well as in the extent and/or structure of attached oligosaccharides [13–17] . O-glycans in tGPI-mucins start with the addition of a single αN-acetylglucosamine ( αGlcNAc ) unit to the hydroxyl group of Thr residues , which may remain unsubstituted or become elongated with different carbohydrates ( mainly Galactopyranoses , Galp ) in multiple linkages and configurations [16 , 18] . In the presence of suitable donors , tGPI-mucins become rapidly sialylated by means of a parasite-encoded trans-sialidase ( TS ) [11 , 18–20] . This reaction involves the cleavage of a sialic acid ( SA ) residue linked in α ( 2–3 ) configuration to a terminal βGalp ( SAα2–3βGalp ) in the donor macromolecule and the subsequent formation of the same linkage on tGPI-mucins [18 , 21] . Alternatively , certain βGalp terminal units of tGPI-mucins may become modified with αGalp residues within the parasite secretory pathway , which generates highly antigenic structures and precludes their further elaboration with SA on the trypomastigote membrane [22] . Importantly , SA-containing neo-glycotopes on tGPI-mucins are involved in trypomastigote recognition and invasion of mammalian cells [23 , 24] , as well as in providing protection against lytic antibodies [11] and/or complement opsonization [25] . In addition to their surface-associated roles , tGPI-mucins are also shed to the milieu as part of multi-cargo and multi-tasking micro-vesicles ( MVs ) that bud from the trypomastigote plasma membrane [26 , 27] . The lipid anchor of tGPI-mucins triggers the synthesis of proinflammatory cytokines in macrophages and other immunomodulatory phenomena in the infected host [28–30] . TSSA ( Trypomastigote Small Surface Antigen ) is a highly antigenic GPI-anchored molecule expressed by T . cruzi trypomastigotes , likely involved in its interaction with host cells [31–35] . TSSA displays substantial sequence conservation to members of the TcMUC family , particularly along its deduced N- and C-terminal regions , and was thus included within this family [15 , 32] . However , TSSA codes for a short polypeptide with no Thr-rich region that migrates in non-reducing SDS-PAGE as a single ~15 kDa species [32] , thus well below the molecular mass determined for tGPI-mucins ( ~45–220 kDa [16] ) . Also at variance with tGPI-mucins , TSSA adhesion properties seem to rely on the exposition of peptidic rather than glycan moieties [23 , 31] . This point was further assessed by direct binding assays of recombinant TSSA molecules to cultured mammalian cells [31] . Detailed genetic characterization of the TSSA locus disclosed sequence variations among parasite strains that , when considered as a whole , turned out to be diagnostic for each of the 6 phylogenetic groups ( named as TcI to TcVI ) delineated within the T . cruzi taxon [36 , 37] . Interestingly , two TSSA variants expressed by TcI and TcVI extant T . cruzi clades were shown to exhibit significant differences in their antigenicity as well as in their host cell binding and signaling properties [31 , 32 , 38] . Here we show that although encoded by TcMUC genes and being co-expressed on the trypomastigote coat , TSSA and tGPI-mucins are completely different species . While tGPI-mucins undergo extensive O-glycosylation and terminal sialylation to fulfill their main protective roles , TSSA seems to be displayed as a hypo-glycosylated molecule . Both kind of molecules segregate to mutually exclusive membrane domains , further supporting their functional diversification . We next used a variety of biochemical and genetic approaches to demonstrate that exposition of ‘naked’ peptidic motifs is critical for TSSA engagement to host cell surface receptor ( s ) prior to trypomastigote internalization .
CL Brener and Sylvio X-10 clones , belonging to TcVI and TcI T . cruzi phylogenetic groups , respectively , were used in this study . Parasite developmental forms were obtained and purified as described [39 , 40] . Briefly , epimastigote forms were grown at 28°C in brain-heart tryptose ( BHT ) medium ( BD ) supplemented with 10% fetal calf serum ( FCS , GIBCO Laboratories ) whereas cell-derived trypomastigotes ( henceforth trypomastigotes ) and amastigotes were harvested from the supernatant of infected , Mycoplasma-free monkey fibroblast Vero cells ( ATCC ) . Vero and human HeLa cells ( ATCC ) were grown at 37°C and 5% CO2 in DMEM supplemented with 10% FCS , 0 . 292 g/L L-glutamine , 100 IU/mL Penicillin and 100 μg/mL Streptomycin ( all from GIBCO Laboratories ) . Tagged versions of the TSSA variant encoded in the Esmeraldo-like chromosome of the CL Brener clone ( TSSA-CL; GenBank Accession Number ACY54510 ) , and the TSSA variant encoded by the Sylvio X-10 clone ( TSSA-Sy; GenBank Accession Number ACY02865 . 1 ) have been described [31] . The tagged version of a canonical TcMUC gene ( GenBank Accession Number U32448 ) has also been described [41] . All of these constructs bear a single , in-frame FLAG epitope placed immediately upstream of the GPI-anchoring signal , which does not interfere with protein trafficking , processing and/or surface display [31 , 41] . These constructs were subcloned into the episomal pTEX-omni plasmid [42] by using the XbaI and XhoI restriction sites . Cloning was checked by restriction mapping analysis and DNA sequencing . For parasite transfection , exponentially growing epimastigotes ( 3 x 108 ) were harvested , washed with phosphate-buffer saline ( PBS ) , transferred to a 0 . 2 cm gap cuvette ( Bio-Rad ) with 10 μg of purified DNA and electroporated as described [42] . Parasites were not cloned by limited dilution or enriched by any means , and antibiotic selection ( 500 μg/mL G418; GIBCO Laboratories ) was sustained over time once stable transfected populations were obtained . For growth curves , epimastigotes were seeded at a density of 1 x 106 parasites per mL in BHT 10% FCS without G418 and the cell number was quantified at the indicated time-points using a Neubauer chamber . For metacyclogenesis assays exponentially growing epimastigote forms ( 2 x 106 ) were harvested , diluted in BHT without G418 and maintained at 28°C without agitation . Metacyclogenesis was evaluated at different time-points by direct counting on Neubauer chambers . For each sample , at least 150 fixed parasites were counted and epimastigote vs metacyclic forms were discriminated morphologically [2] . Total parasite lysates were run in SDS-PAGE , transferred to PVDF membranes ( GE Healthcare ) , and probed with mouse monoclonal antibody ( mAb ) anti-FLAG ( clone M2 , Sigma ) followed by HRP-conjugated anti-mouse IgG ( Sigma ) ( both at 1:5 , 000 dilution ) and the SuperSignal West Femto Chemiluminescent Substrate ( Pierce ) . Antiserum to TSSA-CL [31 , 32] , to TcMUC [19] , to T . cruzi TcSMUG L [43] and to T . cruzi Glutamate Dehydrogenase ( TcGDh , [44] ) were used at 1:5 , 000 , 1:2 , 000 , 1:5 , 000 and 1:5 , 000 dilution , respectively . Western blots on total parasite extracts were performed to assess the recognition protein profiles for antisera to TcMUC and TSSA-CL ( S1 Fig ) . For flow cytometry analyses , parasites ( 1 . 5 x 106 ) were washed , blocked in PBS 10% FCS , and incubated with rabbit polyclonal antibodies to FLAG ( Sigma ) or antiserum to TSSA-CL ( both at 1:200 dilution ) , in an ice-water bath followed by Alexa Fluor-conjugated secondary antibodies ( 1:500 dilution ) ( Molecular Probes ) . After several washes , parasites were resuspended in 300 μL of PBS containing 4% ( w/v ) paraformaldehyde ( PFA ) , extensively washed with PBS and analyzed using FACS CyFLOW Partec and FloMax software . Parasites were harvested , washed in PBS , adhered to poly-L-lysine ( Sigma ) coated cover-slips , fixed for 30 min in PBS 4% PFA , blocked for 30 min in 4% Bovine Serum Albumin ( BSA , Sigma ) in PBS ( PBS-BSA ) supplemented with 0 . 5% saponin ( Sigma ) for permeabilization , and probed with mAb anti-FLAG diluted 1:500 in PBS-BSA or rabbit antiserum to TSSA-CL ( 1:4 , 000 dilution ) . After extensive washings with PBS , secondary Alexa Fluor-conjugated antibodies were added at 1:500 dilution in PBS-BSA . Samples were extensively washed with PBS and mounted . Images were obtained with a Nikon Eclipse 80i epi-fluorescence microscope coupled to a DS-Qi1 CCD camera or with an IX-81 microscope attached with a FV-1000 confocal module . In the latter case , the objective was a PLAN APO 60X NA 1 . 42 oil immersion ( Olympus , Japan ) and the acquisition software used was FV 10-ASW 3 . 1 . Images were treated using ImageJ 1 . 45s Software ( NIH , USA ) . Vero cells ( 5 x 104 ) grown on 24-well culture plates were added with 1 or 2 x 105 Sylvio X-10 or CL Brener trypomastigote forms ( with up to 5% of contaminant amastigote forms ) , respectively . When indicated , parasites were mixed with an equal volume of a specific peptide at 100 μg/mL or PBS before being added to the cell monolayer . After 3 h of incubation at 4°C ( for adhesion assays ) or at 37°C ( for infection assays ) , cells were washed with PBS to remove non-attached parasites and fixed with PBS-PFA immediately ( for adhesion assays ) or after additional 36 h incubation in DMEM 4% FCS at 37°C ( for infection assays ) . Following extensive washings in PBS , cells were processed for IIF assays as described [31] . Infection rate was determined by manual counting of infected and total cells whereas adhesion rate was determined by counting cells with adhered/recently internalized parasites and total cells using the Image J plug-ins Cell Counter and Nucleus Counter in at least 1 , 000 DAPI-stained cells . Three independent experiments were carried out , each one in duplicate . Vero cells ( 1 x 104 ) were seeded on the bottom of trans-well microplates ( 0 . 4 mm pore size , Corning Fisher , NY ) . After 24 h cells were added with 1 x 105 wild type trypomastigotes whereas 2 x 105 trypomastigotes of the indicated transgenic or parental line were seeded in the upper chamber of the trans-well . Following a 3 h-incubation period at 37°C , the upper trans-well chambers were removed , cells were washed with PBS to remove non-attached parasites and processed for IIF assays after additional 36 h incubation in DMEM 4% FCS . Parasite conditioned medium ( CM ) was prepared as described [19] and diluted 1:2 in fresh MEM 4% FCS for infection assays . When indicated , CM was fractionated onto 25 μL of mAb anti-FLAG-Sepharose ( Sigma ) or control anti-hemmaglutinin-Agarose ( Roche ) as described [43] . Flow-through ( unbound ) fractions were then used for infection assays as above . Glutathione S-transferase ( GST ) fusion proteins bearing the central and mature region of TSSA-Sy ( GST-TSSA-Sy24-61 ) and TSSA-CL ( GST-TSSA-CL24-62 ) , as well as variants spanning partially overlapped sequences from TSSA-CL24-62 have been described [33] . HeLa cells ( 5 x 104 ) placed in 96-well culture plates were grown overnight , fixed with PBS 4% PFA , blocked with PBS 10% FCS for 1 h , and incubated with 200 μg/mL of the indicated GST-fusion protein for 1 h followed by mAb anti-GST ( clone GST-2 , Sigma ) diluted 1:1 , 000 in PBS 2% FCS . Plates were washed with PBS , added with HRP-conjugated secondary antibody diluted 1:5 , 000 in PBS 2% FCS followed by 100 μL of 3 , 3’ , 5 , 5’-Tetramethylbenzidine and 50 μL 2 M sulfuric acid , and the absorbance read at 450 nm . Custom peptides were synthesized by GenScript . Sequences were as follows: pCL22-38 , 22CTTANGGSTSSTPPSGT38; pCL30-44 , 30TSSTPPSGTENKPAT44; pCL42-56 , 42PATGEAPSQPGASSG56; and pSy41-55 , 41TAAGGTPSPSGASSG55 . These were derived from TSSA-CL or TSSA-Sy deduced protein sequences; and residues are numbered according to their position in the corresponding protein . Neither peptide affected parasite and/or cell viability under the assayed conditions , as revealed by propidium iodide uptake and trypan blue exclusion , respectively [45] . Recently harvested trypomastigotes ( 5 x 106 ) were washed twice in PBS supplemented with 2% glucose and incubated for 48 h in MEM without FCS at either pH7 or pH5 , at 37°C . Samples were taken at the indicated time-points , washed twice with PBS and fixed with PBS-PFA 4% . Direct counting of cell-derived trypomastigote and amastigote forms were determined on Neubauer chambers . Alternatively , amastigogenesis was assessed by IIF assays , labeling parasite samples taken at the indicated time-points with an antiserum raised against a fragment ( residues 20 to 67 ) of an amastin protein ( TcCLB . 506437 ) . The use of amastin as a T . cruzi amastigote marker has been validated [46] . Live trypomastigote forms from the CL Brener strain were extensively washed in cold PBS and labeled for 30 min in the presence of 10 mM 2-deoxyglucose ( Sigma ) and 1 mM of the azido-sialyllactose analog N-azidoacetyl neuraminyl α2–3lactose ( Neu5Azα2–3LacβOMe ) [19] . When indicated , recombinant TS was added to the reaction mixture as described [43] . Reaction was heated at 65°C to inactivate TS and non-permeabilized parasites labeled by the Staudinger method with 250 μM Phosphino-FLAG ( Sigma ) for 20 min at room temperature [19] . Following extensive washings , parasites were processed for IIF assay as above or resuspended ( at 500 x 106 per mL ) in ice-cold immunoprecipitation buffer ( 150 mM NaCl , 50 mM Tris/HCl , pH 7 . 6 , 1 mM EDTA , 0 . 1% Nonidet P40 , 1% Triton X-100 , 100 μM Tos-Lys-CH2Cl and 1 mM PMSF ) and incubated on ice for 1 h . After preclearing , samples were fractionated onto 25 μL of mAb anti-FLAG-Sepharose overnight at 4°C . Following several washings , retained molecules were directly cracked in 100 μL for Western Blot . Unbound fractions were precipitated with acetone and processed for Western blot . Trypomastigote forms ( 1–3×109 ) were delipidated by chloroform/methanol/water ( 10:20:8 v/v/v ) treatment as described [16] . Briefly , the soluble fraction was evaporated under N2 stream and then partitioned with butan-1-ol/water ( 2:1 , v/v ) . The butan-1-ol phase ( F1 ) contains mainly lipids , phospholipids and GIPLs , whereas the aqueous phase ( F2 ) is enriched in mucins . Both phases were further extracted as before . Delipidated parasite pellets were also extracted with butan-1-ol/water ( 2:1 , v/v ) at 4°C , and the mucin-rich aqueous phase ( F3 ) was stored . Final parasite pellets ( P ) were resuspended in denaturing loading buffer containing 6 M urea and 100 μg/mL DNAse I ( Sigma ) . Epimastigotes ( 2×108 ) were harvested , washed with PBS , resuspended in 200 μL of lysis buffer ( Tris-HCL 20 mM pH 7 . 6 , EDTA 1mM , Sacarose 0 , 25M , Nonidet-P40 0 , 5% ( v/v ) ) and incubated on ice for 1 h . Upon centrifugation to remove cellular debris supernatants were added with NaOH ( 0 . 1 N final ) and incubated for 4 h at 40°C . Treated and untreated samples were analyzed by Western Blot . CL Brener trypomastigotes ( 2 × 108 ) were subjected to four consecutive centrifugation rounds , two at 2 , 700 × g for 10 min followed by two at 10 , 600 × g for 10 min . The cell-free CM was supplemented with 10% FCS to a final volume of 100 μL . Exosome purification kit ( System Biosciences , CA ) was added to the CM and MVs were purified according to the manufacturer’s protocol . The pellet containing MVs was cracked in a final volume of 100 μL while proteins in the supernatant were precipitated with cold acetone , cracked in a final volume of 100 μL and analysed by Western blot [19] .
One key and defining functional aspect of tGPI-mucins is their role as major SA acceptors on the trypomastigote coat [11 , 19 , 23] . Hence , to evaluate if TSSA may be functionally considered part of tGPI-mucins , we firstly analyzed the distribution of SA-containing glycoconjugates and TSSA on the parasite surface by fluorescence microscopy . To that end , CL Brener trypomastigotes were sialylated in the presence of exogenously added Neu5Azα2-3LacβOMe . This sialyllactose analog is recognized as an appropriate SA residue donor by T . cruzi TS and readily incorporated into the trypomastigote coat [19] . Once incorporated , the azido group of Neu5Az was covalently coupled to a FLAG epitope through a Cu2+-free click chemistry , thus allowing us to assess the distribution of SA-acceptors by anti-FLAG IIF assays . As shown in Fig 1A , the surface of trypomastigotes became strongly labeled upon addition of Neu5Azα2-3LacβOMe . Control IIF assays carried out in the absence of Neu5Azα-3LacβOMe or over heat-killed parasites rendered negative results [19] , indicating the requirement of active TS for effective incorporation of the derivative sialyl residue . As described [19] , SA-acceptors were displayed in discrete domains following a dotted pattern along the entire surface of the parasite body and the flagellum ( Fig 1A ) . TSSA also presented a discontinuous distribution , with multiple anti-TSSA-reactive spots of apparent larger size than those of SA-acceptors scattered along the trypomastigote surface ( Fig 1A ) . Additional TSSA-reactive spots were observed in the vicinity of trypomastigotes , which may correspond to secreted TSSA molecules ( Fig 1A ) . Importantly , minimal co-localization was observed between TSSA and SA-acceptors signals ( Fig 1A ) . We next fractionated Neu5Az-labeled glycoconjugates on to mAb anti-FLAG-Sepharose and evaluated different fractions by Western blot . As shown , SA-acceptors migrated in reducing SDS-PAGE as a broad smear ranging from 115 to 45 kDa ( Fig 1B , left panel ) , a pattern compatible with tGPI-mucins [16 , 19] . Indeed , part of the Neu5Az-labeled material also reacted with an antiserum raised against a canonical TcMUC product ( Fig 1B , middle panel ) . In contrast , the ~15 kDa TSSA-reactive band was observed exclusively in the flow-through ( i . e . non-sialylated ) fraction ( Fig 1B , right panel ) . Further labeling experiments carried out in the presence of recombinant TS ( to disregard possible steric hindrance and/or enzyme shortage that would interfere with TS-TSSA interaction in our previous experimental setup ) yielded similar results ( Fig 1C ) . In such conditions , and in addition to the 45–115 kDa smear of tGPI-mucins we were also able to detect a faint ~10 kDa FLAG-reactive band in the bound fraction ( Fig 1C , left panel ) , which was not further characterized . Again , TSSA partitioned completely to the unbound fraction ( Fig 1C , right panel ) . Together , these results strongly suggest that TSSA molecules displayed on the trypomastigote surface do not constitute significant SA-acceptors for parasite TS in vivo . This may in turn reflect the absence of terminal βGalp residues in appropriate configuration on TSSA or , alternatively the presence of specific modifications ( i . e . α-galactosylation ) on such terminal residues that preclude their further conjugation to SA by means of TS [16] . To distinguish between these possibilities , we next followed a standard butan-1-ol extraction protocol in unlabeled CL Brener trypomastigotes and analyzed different fractions by Western blot . As shown in Fig 1D , TSSA is detected exclusively in the parasite-associated fraction , strongly suggesting that it does not undergo extensive glycosylation in vivo . In contrast , The TcMUC antiserum revealed a broad smear in the aqueous soluble fractions F3 and , to a lesser extent , F2 , which coincides with the extraction pattern of sialylated or α-galactosylated tGPI-mucins ( Fig 1D ) [16 , 19] . We finally evaluated the pattern of spontaneous secretion of TSSA and tGPI-mucins by CL Brener trypomastigotes . As shown in Fig 1E , secreted TSSA is entirely associated to plasma membrane-derived MVs whereas at least part of secreted tGPI-mucins partitioned to the ‘soluble’ fraction of the conditioned medium ( CM ) of trypomastigotes ( Fig 1E ) . Overall , these findings indicate that TSSA and tGPI-mucins display substantial biochemical divergences and thus constitute structurally ( and also likely functionally ) different species on the trypomastigote coat . In a previous work , we had already developed transgenic epimastigote lines over-expressing different TSSA variants with apparently contrasting features , termed TSSA-CL and TSSA-Sy [31] . Unfortunately , these transgenic lines were generated on the Adriana strain background , which did not undergo significant metacyclogenesis in vitro and hence precluded us to assess the impact of TSSA over-expression on trypomastigotes . To circumvent this limitation we attempted to develop TSSA over-expressing ( TSSA ox ) parasite lines on additional parental backgrounds . Constructs bearing FLAG-tagged TSSA variants were therefore sub-cloned into the pTEX omni vector and independently transfected into CL Brener ( TcVI ) epimastigotes . To expand our analysis , we also transfected these constructs into the Sylvio X-10 clone , comprised within the TcI T . cruzi phylogenetic group . As previously observed in the Adriana strain [31] , TSSA ox epimastigotes showed no significant morphological or growth differences in comparison with parental , wild type parasites ( S2 Fig ) . Expression level , surface localization and distribution of TSSA molecules was assessed by flow cytometry ( Fig 2A ) and confocal microscopy-based assays ( Fig 2B ) . As observed in trypomastigotes ( see Fig 1A ) , TSSA ox epimastigotes from both genetic backgrounds bore a punctate pattern over their entire periphery , including the cell body and the flagellum ( Fig 2B ) . Western blot analysis revealed a FLAG-reactive smear ( ~25–45 kDa ) for either TSSA variant in transgenic epimastigotes ( Fig 2C ) . This pattern of migration on SDS-PAGE clearly differed to that displayed by trypomastigote TSSA ( see Fig 1B–1D ) , suggesting differences in the post-translational processing of this molecule along the parasite life cycle . Indeed , and at variance with trypomastigote-expressed TSSA ( see Fig 1D ) , butan-1-ol extraction experiments supported extensive glycosylation of epimastigote-expressed TSSAs ( Fig 2D ) . Part of FLAG-reactive species could be detected in the mucin-rich fraction F3; thus very similar to the extraction pattern obtained for the epimastigote-restricted TcSMUG L mucin-type products assayed in parallel ( Fig 2D ) [43] . In vitro assays revealed that TSSA ox epimastigotes display quite similar metacyclogenesis rates as compared to the corresponding parental lines ( S3 Fig ) . Epimastigote cultures enriched in metacyclic trypomastigotes forms from every transgenic line were therefore used to infect Vero cell monolayers in vitro and , after several rounds of infection , mammal-dwelling forms of the parasite were obtained from culture medium . TSSA ox trypomastigotes were sorted by standard procedures and subjected to different biochemical and phenotypic analyses . Microscopy-based and flow-cytometry assays indicated that FLAG-tagged TSSAs accumulated at roughly similar levels on the surface of trypomastigotes from both genetic backgrounds ( Fig 2E and 2F ) . As shown for native molecules ( Fig 1A ) , FLAG-tagged TSSAs were not evenly spread along the trypomastigote membrane but following a rather patchy distribution ( Fig 2E ) . To estimate the extent of TSSA over-expression in transgenic trypomastigotes , comparative flow cytometry experiments were carried out on CL Brener wild type and TSSA-CL-ox lines . In this case , parasites were labeled with a TSSA-CL antiserum , revealing that the TSSA-CL-ox line expressed ~30% more TSSA-CL than the parental line ( Fig 2G ) . This experiment could not be performed in Sylvio X-10 parasites due to lack of an appropriate TSSA-Sy antiserum . Interestingly , Western blot experiments revealed that FLAG-tagged TSSAs were expressed as a less diffused band of ~15–20 kDa band on TSSA ox trypomastigotes of both genetic backgrounds ( Fig 2H ) , hence quite similar to the native molecules ( Fig 1B–1D ) and distinct of the TSSAs of over-expressing epimastigotes ( Fig 2C ) . The limited number of trypomastigotes yielded by TSSA ox lines ( see below ) precluded us to carry out detailed biochemical characterizations of FLAG-tagged TSSAs . However , anti-FLAG immunoprecipitation assays followed by Western blot using the TSSA-CL antiserum revealed a contrasting recognition profile for epimastigote- vs trypomastigote-expressed products ( Fig 2I ) , further supporting a differential processing for TSSA on distinct developmental forms of the parasite . Moreover , release of O-glycans by ß-elimination led to a shift in the migration pattern of the TSSA-CL expressed by epimastigotes ( from a ~34–45 kDa smear to a ~17 kDa species ) , and to its recognition by the TSSA-CL antiserum ( Fig 2J ) . A much higher ratio of extracellular amastigotes to trypomastigotes was consistently observed in the supernatant of cells infected with TSSA ox parasites as compared with parental , wild type lines . This bias suggested an imbalance in parasite infectivity , intracellular growth and/or trypomastigote-to-amastigote differentiation . To evaluate the latter issue , we carried out extracellular amastigogenesis assays on CL Brener lines . As shown in Fig 3A , the transformation process was completed after 24 h of incubation at pH7 for both TSSA-CL and TSSA-Sy ox trypomastigotes whereas the parental line took > 48 h to assess . To rule out a possible non-specific effect of transfection and/or protein over-expression on this phenotype , we evaluated in parallel the kinetics of transformation of CL Brener TcMUC ox trypomastigotes . These parasites were transfected with a pTEX omni vector bearing a FLAG-tagged , canonical TcMUC product [41] , selected and differentiated as TSSA ox lines . In contrast to TSSA ox lines , however , TcMUC ox trypomastigotes displayed similar kinetics of transformation than the parental , wild type line ( Fig 3A ) . Similar results were obtained for TSSA ox trypomastigotes in the Sylvio X-10 background ( Fig 3B ) , or when trypomastigote-to-amastigote transformation was evaluated at pH5 morphologically or by means of an amastigote-specific antibody ( S4 Fig ) . No significant differences in the viability of distinct parasite lines along the amastigogenesis experiments were observed ( S4 Fig ) . Together , these findings suggest that over-expression of TSSA leads to an exacerbated amastigogenesis . We next performed in vitro infection experiments with transgenic or parental trypomastigotes . The infection rate was determined 36 h afterwards by direct counting of infected and non-infected cells . In the CL Brener background , TSSA-CL ox parasites displayed an enhanced infectivity ( ~17% ) as compared to those over-expressing TSSA-Sy or parental controls ( Fig 4A ) . No significant differences between transgenic and parental lines were however observed when the number of parasites/infected cell was evaluated ( S5 Fig ) , suggesting that over-expression of TSSA does not affect parasite cell cycle and/or intracellular growth . To further address the specificity of these results , we aimed at carrying out similar infection assays in the presence of exogenously added TSSA-derived peptides ( indicated in Fig 4B ) . To identify the host cell binding sequence ( s ) in TSSA-CL we followed a deletion mapping analysis using a panel of GST-fusion molecules . Host cell binding of each purified recombinant protein was assessed as described [31] . As shown in Fig 4C , most of TSSA-CL-derived sequences had negligible binding activity , similar to that of GST or a GST-fusion spanning residues 24 to 61 of TSSA-Sy ( GST-TSSA-Sy24-61 ) used as negative controls [31] . Conversely , GST-TSSA-CL42-56 and to a lower extent its included sequence GST-TSSA-CL48-56 and GST-TSSA-CL30-44 showed significant binding to cultured cells ( Fig 4B and 4C ) , similar to what has been originally described for GST-TSSA-CL24-62 [31] . Moreover , GST-TSSA-CL42-56 presented improved binding than GST-TSSA-CL24-62 ( Fig 4C ) , suggesting structural constraints imposed by other residues in the latter protein that partially impair its interaction with cell surface ligand ( s ) . Importantly , the improved infectivity of TSSA-CL ox trypomastigotes could be specifically counteracted by preincubation with peptides pCL30-44 or pCL42-56 ( Fig 4A ) . Peptide pCL22-38 , which did not show host cell binding properties ( see GST-TSSA-CL24-38 in Fig 4C ) , or peptide pSy41-55 , which encompasses the corresponding sequence of pCL42-56 in the non-adhesive TSSA-Sy protein ( Fig 4B ) did not affect infectivity of TSSA-CL ox parasites when assayed at the same concentration ( Fig 4A ) . Neither peptide affected parasite viability under the assayed conditions , as revealed by propidium iodide uptake ( S5 Fig ) . Similar experiments were carried out in order to determine parasite adhesion rates . In line with above results , TSSA-CL ox parasites showed increased adhesion to cultured cells as compared to wild type or TSSA-Sy ox ones , and this effect could also be specifically counteracted by peptides derived from the adhesive motifs of TSSA-CL but not with control peptides or PBS ( Fig 4D ) . Overall , our data show a strict correlation between peptide ability to bind to host cells and to inhibit TSSA-driven adhesion ( and therefore infectivity ) of CL Brener trypomastigotes . These findings strongly suggest that recombinant and native TSSA-CL display highly similar specificities . In the Sylvio X-10 background , and in sharp contrast to above results , over-expression of neither TSSA variant had a significant impact on trypomastigote infectivity and/or adhesion ( Fig 4A and 4D ) . Considering that TSSA-CL is actively secreted to the medium ( Fig 1E ) , we next evaluated whether it may also modulate infectivity of trypomastigotes in trans . To that end , the infectivity of trypomastigotes was assessed as before , though in the presence of additional parasites placed on the upper chamber of trans-well microplates . As shown in Fig 4E , a significant decrease in the infection rate was observed when wild type CL Brener trypomastigotes were exposed to the CM of CL Brener TSSA-CL ox parasites as compared to those exposed to the CM of wild type or TSSA-Sy ox counterparts . A similar trend was observed when CL Brener trypomastigotes were exposed to the CM of different Sylvio X-10 parasite lines . In this case , however the decrease in the infection rate of wild type CL Brener trypomastigotes exposed to the CM of Silvio X-10 TSSA-CL ox parasites was not of statistical significance ( Fig 4E ) . The infectivity rate of Sylvio X-10 trypomastigotes , on the other hand , was not affected by the CM of every tested trypomastigote line ( Fig 4E ) . A complete profile of the CM from Sylvio X-10 and CL Brener trypomastigotes , as well as from different transgenic lines is shown in S6 Fig . We next assessed the infectivity of wild type CL Brener trypomastigotes directly added with purified CM of different CL Brener lines . As shown in Fig 4F , the CM from TSSA-CL ox trypomastigotes had a significant inhibitory effect as compared to the CM of TSSA-Sy ox or wild type counterparts . The effect exerted by the CM of TSSA-CL ox parasites could be depleted by FLAG-affinity chromatography ( Fig 4F ) , strongly suggesting that FLAG-tagged TSSA-CL molecules , likely associated to actively secreted MVs ( see Fig 1F ) , constitute the ‘soluble factor’ underpinning this inhibition . Altogether , these results strongly support a direct involvement of TSSA-CL , via adhesive motifs of peptidic nature in the interaction between CL Brener trypomastigotes with the target cell prior to parasite internalization .
Antigenicity mapping experiments [33 , 47] and biochemical data [31] converged in suggesting the exposition of peptidic moieties on surface-associated TSSA molecules , which was at odds with TSSA being part of the TcMUC gene family . In the first part of this work , we undertook a series of biochemical approaches to directly address this issue . Microscopy-based experiments showed that TSSA-CL , as previously reported for other trypomastigote surface markers [19] , does not co-localize with sialylated tGPI-mucins . The punctate pattern observed for TSSA-CL builds upon our hypothesis of the trypomastigote membrane as a highly organized structure made up of multiple and discrete nanoscale domains bearing different protein composition [48] . Further affinity fractionation experiments and whole-parasite butan-1-ol partition assays definitely established that TSSA and tGPI-mucins are completely different species . At variance with tGPI-mucins , which undergo extensive O-glycosylation and terminal sialylation to fulfill their main protective and immunomodulatory roles [11] , TSSA seems to be displayed as a hypo-glycosylated ( or non-glycosylated at all ) molecule on the trypomastigote surface . Although this proposal is based on experimental data and seems to have a functional correlate ( see below ) , a detailed structural analysis of TSSA would be required to definitely address this issue . Host cell binding assays and the use of trypomastigote transgenic lines support that exposition of TSSA ‘naked’ peptide sequences is critical for its engagement of yet unidentified host cell surface receptor ( s ) prior to trypomastigote effective internalization . As shown , at least 2 TSSA-CL-derived synthetic peptides ( from residue 42 to 56 and from residue 30 to 44 ) displaying adhesion properties to cultured cells in vitro , were able to interfere with or partially block CL Brener trypomastigote-host cell interactions , indicating that these peptidic motifs are involved in the receptor pairing of the native , surface-associated TSSA-CL molecule in vivo . Sequence polymorphisms between TSSA-CL and TSSA-Sy variants were shown to be focused on the central region of this molecule ( from residue 36 to 51 ) , and likely explain the significant differences in their host cell binding properties [34 , 36 , 38] . Taking into account that TSSA-CL likely contributes to trypomastigote-host cell binding , it may be also speculated that its aggregation in discrete domains along the surface coat increases the avidity of its mediated interactions , hence providing the trypomastigote with a solid grasp to the target cell . Interestingly , anti-TSSA-CL antibodies elicited by Chagasic patients are preferentially directed towards linear epitopes contained within the herein identified adhesive motifs of TSSA-CL and are thus expected to have a direct detrimental effect on parasite infectivity [33 , 47] . In the second part of this work , we generated transgenic epimastigote lines over-expressing TSSA variants to get further insights into its in vivo functional role ( s ) . Independently of the recipient strain and the transfected construct , TSSA over-expression led to the surface accumulation and ‘patchy’ distribution of a FLAG-tagged product , which does not affect epimastigote growth and/or differentiation into metacyclic forms . Curiously , epimastigote-expressed TSSAs undergo mucin-type glycosylation . This kind of ‘aberrant’ processing has been previously observed in TSSA ox epimastigotes from the Adriana strain [31] , suggesting it is a common feature of this parasite stage , likely attributed to its particular profiling of glycosyl transferases [18] . Upon transformation into trypomastigotes , and in strict accordance to previous recombinant protein binding data [31] , CL Brener trypomastigote forms over-expressing TSSA-CL , but not TSSA-Sy , show significantly improved adhesion and infectivity towards non-macrophagic cells . Even though we cannot formally rule out the possible contribution of additional surface adhesin ( s ) whose expression/processing may become deregulated in transgenic parasites , the counteracting effect of peptides spanning TSSA-CL adhesion motifs allowed us to nail down this phenotype to TSSA-CL . Over-expression of TSSA-CL does not have a significant impact on the infectivity and/or adhesion of Sylvio X-10 trypomastigotes , which is in principle difficult to reconcile with our main hypothesis . However , the fact that Sylvio X-10 parasites ( not encoding for an adhesive TSSA variant ) are nonetheless adhesive/infectious in our in vitro system indicates that target cell recognition/invasion capabilities of these parasites rely on a different subset of surface receptor ( s ) . As mentioned , T . cruzi trypomastigotes bear a huge repertoire of apparently redundant adhesins [5–7] . As part of the compensatory mechanisms , we hypothesize that some variants of these adhesion molecules showing improved expression/function may have been selected for in Sylvio X-10 parasites . Indeed , several quantitative and/or qualitative differences of surface molecules between T . cruzi strains , some of which are likely associated to parasite phenotypic variations , have been described [49–52] . The differential ‘surface coat environment’ of Sylvio X-10 trypomastigotes may be responsible for buffering the otherwise net increase in trypomastigote infectiveness caused by over-expression of TSSA-CL . Alternatively , it may be speculated that TSSA-CL undergoes a particular processing in Sylvio X-10 trypomastigotes . However , the facts that this molecule is recognized by TSSA-CL antiserum ( S7 Fig ) and that it is able to partially block the infectivity of CL Brener parasites in trans ( Fig 4E ) argue against this possibility . In addition to its role in trypomastigote-host cell interactions , we show that TSSA ox trypomastigotes independently of the variant and parasite genetic background , display exacerbated amastigogenesis . The use of control trypomastigotes over-expressing a distinct GPI-anchored molecule undergoing similar intracellular processing and trafficking pathways such as TcMUC ox [41] indicate that this phenotype cannot be related to a global secretory system depression and/or saturation of transport mechanisms . The molecular and cellular basis underlying the putative link between TSSA and amastigogenesis remain to be addressed . In summary , we have shown that in spite of being encoded by a TcMUC-like gene and being co-expressed on the trypomastigote coat , TSSA is neither structurally not functionally related to tGPI-mucins . Our data indicates that this molecule plays an important role not only on the infectivity of trypomastigotes but also on the phenotypic variability displayed by T . cruzi strains [37] , and strongly support this molecule as an excellent candidate for molecular intervention and/or vaccine development in Chagas disease . | Infection with Trypanosoma cruzi produces a chronic and debilitating infectious disease known as Chagas disease , of major significance in Latin America and an emergent threat to global public health . In the absence of vaccines and/or appropriate chemotherapies , the search for parasite effectors that support infection of mammalian cells is a focus of significant interest . One such candidate is the Trypomastigote Small Surface Antigen ( TSSA ) , a polymorphic molecule expressed on the surface coat of infective trypomastigote forms . Previous data indicated that recombinant versions of two different TSSA variants ( termed TSSA-CL and TSSA-Sy ) encoded by parasite strains belonging to extant phylogenetic groups exhibited contrasting host cell binding and signaling abilities . Here , we generated genetically modified strains of T . cruzi over-expressing different TSSAs to address this issue . Trypomastigotes over-expressing TSSA-CL , the ‘adhesive variant’ , displayed improved adhesion and infectivity towards non-macrophagic cell lines as compared to those over-expressing TSSA-Sy or parental lines . In addition , and irrespective of the protein variant , TSSA over-expression enhanced trypomastigote-to-amastigote conversion . Overall , our data strongly suggest that TSSA plays an important role not only on the infectivity and differentiation of T . cruzi trypomastigotes but also on the phenotypic variability displayed by different strains of this parasite . These data , together with the fact that TSSA recalls a strong and likely protective humoral response during human infections , support this molecule as an excellent candidate for molecular intervention and/or vaccine development in Chagas disease . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] | [
"cell",
"binding",
"blood",
"serum",
"cell",
"physiology",
"medicine",
"and",
"health",
"sciences",
"body",
"fluids",
"viral",
"transmission",
"and",
"infection",
"microbiology",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"protozoan",
"life",
"cycles",
"developmental",
"biology",
"trypomastigotes",
"protozoans",
"epimastigotes",
"life",
"cycles",
"amastigotes",
"trypanosoma",
"cruzi",
"trypanosoma",
"host",
"cells",
"blood",
"anatomy",
"cell",
"biology",
"virology",
"physiology",
"biology",
"and",
"life",
"sciences",
"protozoology",
"immune",
"serum",
"organisms"
] | 2017 | The Trypomastigote Small Surface Antigen (TSSA) regulates Trypanosoma cruzi infectivity and differentiation |
Visceral leishmaniasis ( VL ) is a parasitic disease , transmitted by the sand fly species Phlebotomus argentipes in the Indian sub-continent . Effective vector control is highly desirable to reduce vector density and human and vector contact in the endemic communities with the aim to curtail disease transmission . We evaluated the effect of long lasting insecticide treated bed nets ( LLIN ) and bed nets impregnated with slow-release insecticide tablet K-O TAB 1-2-3 ( jointly insecticide-treated nets or ITN ) on VL incidence in a highly endemic sub-district ( upazila ) in Bangladesh . Several distributions of LLIN or K-O TAB 1-2-3 for self-impregnation of bed nets at home took place in Fulbaria upazila , Mymensigh district from 2004 to 2008 under three research projects , respectively funded by CDC , Atlanta , USA ( 2004 ) and WHO-TDR , Geneva , Switzerland ( 2006 & 2008 ) . We included all households ( n = 8142 ) in the 20 villages that had benefited in the past from one of these interventions ( 1295 donated LLIN and 11 , 918 local bed nets impregnated with K-O TAB 1-2-3 ) in the “exposed cohort” . We recruited a “non-exposed cohort” in villages with contemporaneously similar incidence rates who had not received such vector control interventions ( 7729 HHs from nine villages ) . In both cohorts , we visited all families house to house and ascertained any VL cases for the 3 year period before and after the intervention . We evaluated the incidence rate ( IR ) of VL in both cohorts as primary endpoint , applying the difference-in-differences method . The study identified 1011 VL cases ( IR 140 . 47/10 , 000 per year [py] ) before the intervention , of which 534 and 477 cases in the intervention and control areas respectively . The IR was 144 . 13/10 , 000 py ( 534/37050 ) and 136 . 59/10 , 000 py ( 477/34923 ) in the intervention and control areas respectively , with no significant difference ( p = 0 . 3901 ) before the intervention . After the intervention , a total of 555 cases ( IR 77 . 11/10 , 000 py ) were identified of which 178 ( IR 48 . 04/10 , 000 py ) in the intervention and 377 ( 107 . 95/10 , 000 py ) in the control area . The intervention area had a significant lower IR than the control area during follow up , rate difference = –59 . 91 , p<0 . 0001 . The IR during follow up was significantly reduced by 96 . 09/10 , 000 py in the intervention area ( p<0 . 0001 ) and 28 . 63/10 , 000 py in control area ( p<0 . 0001 ) compared to baseline . There was a strong and significant overall effect of the ITN intervention , δ = –67 . 45 , p <0 . 0001 . Sex ( OR = 1 . 36 , p<0 . 0001 ) and age ( OR = 0 . 99 , p<0 . 0001 ) also had a significant effect on VL incidence . Male had a higher risk of VL than female and one year increase in age decreased the likelihood of VL by about 0 . 92% . Two third of the VL incidence occurred in the age range 2 to 30 years ( median age of VL patients was 17 years ) . VL incidence rate was significantly lower in the ITN intervention cohort compared to control in Bangladesh . Some bias due to more intense screen-and-treat activities or other interventions in the intervention area cannot be ruled out . Nonetheless , given their feasibility and sustainability , ITNs should be considered for integrated vector control during the maintenance phase of the VL elimination programme .
Visceral leishmaniasis ( VL ) —also known as kala-azar ( KA ) in the Indian sub-continent—is a deadly parasitic disease transmitted by the female Phlebotomus argentipes sand fly . In the South-East Asia Region , humans are the only proven reservoir of the parasite , Leishmania donovani . Kala-azar has been present in the Bengal territory ( presently West Bengal , India , and Bangladesh ) since the early 1800s [1] and gradually spread along the course of the Ganges and the Brahmaputra rivers , the major transport routes of Bengal . In what is today Bangladesh , KA was first described in 1824 in Jessore district [2] , where an epidemic killed an estimated 75 , 000 people between 1824 and 1827 [1] . The historical records describe the classical picture of KA , with prolonged irregular fever , progressive emaciation despite good appetite , enlargement of liver and spleen and black coloration of skin [3] . In the late 1950s and early 1960s , WHO launched a malaria eradication programme throughout the South Asian sub-continent based primarily on Indoor Residual Spraying ( IRS ) of Dichlorodiphenyltrichloroethane ( DDT ) . During this programme , KA almost disappeared as a collateral benefit [4] . However , within a few years after the end of the malaria eradication efforts , KA returned to Bihar and Bengal on both sides of the India-Bangladesh borders [5] . In Bangladesh , sporadic KA cases were reported again from the late 1960s onwards [6] . Between 1968 and 1980 , 59 KA patients were reported , mostly from 5 districts ( Sirajgang , Pabna , Mymensingh , Rajshahi , and Tangail ) [7] . The numbers of KA cases soared from 1980 onwards , and a major outbreak occurred in Pabna district [1] . Between 1994 and 2013 , the National Programme of Disease Control , Directorate General Health Services ( DGHS ) , Government of Bangladesh reported 109 , 266 KA cases and 329 deaths [8] . Fifty percent of those cases were reported from just five sub-districts ( Upazila ) of Mymensingh district [8] . In 2005 , three countries ( Bangladesh , India , and Nepal ) , supported by WHO , launched a regional initiative to eliminate KA as a public health problem from the region and signed a Memorandum of Understanding ( MoU ) to this effect . The initiative aimed to reduce the KA incidence to one case per 10 , 000 population in each endemic sub-district by 2015 [9] . This deadline was later extended to 2017 [10] , and 2020 in the London Declaration on Neglected Tropical Diseases [11] . Despite an impressive decrease in the number of cases in each country , WHO could not yet validate the KA elimination status in any of them and advocates for more intense and sustained control efforts and disease surveillance . The intervention strategies in the elimination programme are based on case detection and treatment and integrated vector management ( IVM ) [9] . In Bangladesh , however , no specific sand fly control operations were carried out by the programme between 1999 and early 2012 [8 , 12] . It took a long time to register the required insecticides for the indication of sand fly control . The first indoor residual spraying ( IRS ) activity was conducted using deltamethrin 5WP in April/May 2012 in eight highly endemic Upazilas ( sub-districts ) [13] . Till today none of the countries was able to fully implement the IVM strategy in the region , as they tend to implement IRS for sand fly control only , and this in an independent way of any other vector control operation . Although well-performed IRS can reduce vector density dramatically , it remains operationally challenging and expensive , and its acceptance by the community is not always optimal . Several authors have highlighted its limitations , in terms of insecticide resistance , quality of implementation , occupational hazard , cost , sand fly adaptation , etc . [14–17] . Therefore , there is a need for alternative tools which are operationally easy to implement and cost-effective in terms of per household protected . The question of whether there are alternatives to IRS will only become more relevant in the post-elimination era . We briefly summarize here the evidence on P . argentipes control methods from Bangladesh and the region so far . Between 2002 and 2009 several epidemiological and entomological studies were conducted in the highly endemic area of Fulbaria , one of the Upazilas of Mymensingh district , either to assess the KA disease burden and its risk factors [18] , or to evaluate the effectiveness of insecticide-treated nets as an alternative for IRS for controlling the P . argentipes sand fly [19–21] . Consistent use of non-treated local bed nets in summer was associated with reduced risk for VL in an observational study [18] . This study also showed that use of bed nets is acceptable in the rural community of Bangladesh and found a high percentage of households owning at least one bed net [18] , similar to evaluations in India and Nepal [22] . Inspired by the effectiveness of insecticide treated bed nets for malaria control , several intervention studies evaluated either donated long-lasting insecticide-impregnated bed nets ( LLIN ) or local bed nets impregnated with slow release insecticide tablet K-O TAB 1-2-3 , on entomological endpoints [19–21] . For ease of understanding , we regroup both interventions as “Insecticide-Treated Nets” ( ITN ) in the remainder of the text . The two multi-country intervention studies found significant reductions in sand fly density , ranging from 60% to 80% . A less pronounced 25% reduction of sand fly density was found in a cluster randomized trial ( CRT ) conducted in India and Nepal comparing households covered by LLIN with households where no LLINs were used , which were allowed to continue to use their own commercially available non-treated nets [23] . However , the CRT study in India and Nepal did not find any effect of the LLIN distribution on Leishmania donovani infection nor KA incidence , notwithstanding a high coverage of all household members and effective use of the LLINs [24] . Authors suggested this negative finding might be related to exposure outside the peridomestic environment due to changing sand fly behaviour , which was partly confirmed later [25] . Long-standing insecticide pressure because of the repeated IRS campaigns in India and Nepal might have forced sand flies to adapt again to the outdoor environment . It is worthwhile to study the same question in Bangladesh though , as the sociocultural and environmental parameters are somewhat different . In Bangladesh , in contrast to India , no IRS was in place for a very long time in the KA endemic areas , so there was no insecticide pressure on the peridomestic sand fly habitat . Therefore , we set out to investigate the impact of ITNs on KA incidence in Bangladesh through a retrospective cohort analysis , as staging another CRT would raise ethical questions and would not be feasible in the present context of very low incidence rates near elimination .
The present study protocol was approved by the Ethical Review Committees of the Bangladesh Medical Research Council ( BMRC ) and the Special Program for Research and Training in Tropical Diseases/Regional Office for South-East Asia , World Health Organization ( WHO/SEARO ) , India . Informed consent in the household survey was signed by the head of household before their voluntary participation in the study . This study is a retrospective cohort analysis set in Fulbaria sub-district , Bangladesh . Fulbaria is located 111 km from the capital city Dhaka , and 23 kilometers away from the district headquarters respectively . Fulbaria has 13 unions ( lowest administrative unit ) and 116 villages . The Upazila occupies an area of 398 . 70 km2 including 14 . 76 km2 forest area . In the Fulbaria population , we retrospectively defined two distinct cohorts; the exposed cohort was the one who benefited from an ITN intervention in the recent past , and the unexposed were those who did not . The first , “exposed cohort” was therefore composed of all the communities who had benefited previously from a LLIN or K-O TAB 1-2-3 distribution in 1 of three distinct studies ( 18–21 ) . The non-exposed cohort ( control ) ( i . e . , families that did not receive any donated long-lasting nets or whose local nets were not impregnated ) , was composed of villages of similar population size with a comparable KA incidence rate in the corresponding study period of each of the three published studies ( 18–21 ) . Based on data from the epidemiological records of the Ministry of Health ( passive surveillance data ) , we then ranked all the KA endemic villages of Fulbaria ( excluding those already included in the exposed cohort ) , according to the number of reported cases , for the corresponding time period when the respective ITN interventions were carried out ( 2004 , 2006 , 2008 ) . We then randomly selected nine endemic villages from the 15 top-ranked villages ( Baddiyan bari , Balashawr , Palashtali , Deoli , Dhamar , Shibpur , Kathgara , Harirumbari , and Palashihata ) ( Fig 1 ) , and included these nine communities ( n = 7729 ) in the control cohort . We describe here the different ITN interventions that took place in the exposed cohort . A first epidemiological study was conducted in a total of 506 households from three paras ( sub-villages ) namely Nadirpar , Lakxmipur , and Bamonbaid of Chouder village ( Fig 1 ) of Fulbaria union , Fulbaria Upazila , Mymensingh district between 2002 and 2004 [18] . The study was funded by CDC , Atlanta , USA and implemented by icddr , b . After completion of the study , each HH was donated one unit of LLIN ( manufactured by Vestergaard Frandsen Private Limited ) . Between 2006 and 2007 , a cluster randomized trial was conducted with four arms ( 3-intervention [arm-1: IRS using deltamethrin 5 WP , arm-2: LLIN and arm-3: environmental management] and 1-control arm where LLIN were donated after completion of the study period ) in Fulbaria Upazila ( Fig 1 ) , Mymensingh district [19 , 20] . This study involved a total of 596 households . The study was funded by the Special Programme for Research and Training in Tropical Diseases ( TDR ) , WHO , Geneva , Switzerland and conducted by the National Institute of Preventive and Social Medicine ( NIPSOM ) , Bangladesh . LLIN ( manufactured by Vestergaard Frandsen Private Limited ) were donated in two arms . Last , a community-based study was conducted with 6967 households in Putijana union ( Fig 1 ) of Fulbaria Upazila , Mymensingh district between 2007 and 2008 [21] . In this study , all existing bed nets at HH level were impregnated with slow release insecticide tablet K-O TAB 1-2-3 ( 0 . 4g deltamethrin in a 1 . 6 g tablet and a chemical binder ) manufactured by Bayer Crop Science , Isando , South Africa . The study was supported by the Special Programme for Research and Training in Tropical Diseases ( TDR ) , WHO , Geneva , Switzerland and conducted by National Institute of Preventive and Social Medicine ( NIPSOM ) , Bangladesh . We assessed the outcome “KA” independently from the original research projects , and in the same way for the intervention and control area , as follows . We exploited the full database for Fulbaria sub-district for the period of 2001 to 2011 to identify reported KA cases in the intervention and control areas and visited all affected communities in an exhaustive house to house survey . All households ( HH ) of both cohorts were visited between 2011 and 2012 , and the head of the HH/responsible adult was interviewed in order to ascertain the number of reported KA cases in the period of three years before and after the intervention for the three distinct study sites described above , and in a matching time frame for the control cohort . The period of observation was seven years for each intervention , and the respective time windows were as follows: for the 2004 CDC funded study: 2001–2007; for the 2006 TDR study: 2003–2009; and for the 2008 TDR study: 2005–2011 . Additionally , information about current bed net use and washing practices was also collected in the intervention area . Trained Research Assistants conducted the interview using a structured questionnaire . A standard data entry interface was designed using Microsoft Office Access for entering study data . Data were checked and cleaned before analysis . Percentages were used to summarise the demographic and study variables . VL incidence rate was calculated for control and intervention areas ( per 10 , 000 persons per year ) for baseline and follow-up period separately . Z-test was used to compare the VL incidence rates between the intervention and control area , and p-values at the 0 . 05 significance level were used . Difference In Difference ( DID ) estimates ( δ ) were calculated to estimate the effect of the intervention at the community level . Binary logistic regression was used to calculate odds ratios for the effect of gender and age on VL incidence rate . STATA/MP 13 . 0 for Windows ( StataCorp LP , College station , TX77845 , USA ) was used for data analysis .
Of a total of 15 , 871 HHs ( 71 , 973 population ) , 8142 HHs ( 37 , 050 population ) and 7729 HHs ( 34 , 923 population ) were included in the study as exposed and control cohort . Table 1 shows that their baseline characteristics are very comparable , including for the frequency of KA at baseline in household level . In the household survey , we investigated the persistent use of bed nets in the intervention area . Of 8142 HHs that benefited at one point in the past from an ITN distribution in the intervention area , more than 92 . 2% HHs had at least one bed net in their house at the time of our household survey . Among those , 80 . 1% were ITNs , either self-impregnated with K-O TAB 1-2-3 or LLIN , the others were non-impregnated commercial nets ( Table 2 ) . About 33 . 9% HHs even had two nets , of which 88 . 9% were ITNs . However , 7 . 8% of all HHs did not have any bed net at the time of survey . More than 84 . 3% HHs ( 6864/8142 ) informed that they were always sleeping under a bed net . Sixty-five percent of all HHs reported that they felt impregnated nets were effective against mosquitoes along with other insects while about 32% HHs informed nets only effective against mosquitoes . About 82% HHs stated that there were less kala-azar ( VL ) cases in their community after the introduction of impregnated nets while about 17% respondents had no opinion . Only 12% of our respondents knew that kala-azar is transmitted by a sand fly bite , while the majority ( 74 . 4% ) said it is transmitted by mosquitoes . Almost all HHs ( 98% ) expressed a demand for ITNs , and the majority ( 72 . 7% respondents ) asked for a free-of-cost distribution as a government donation ( Table 2 ) . ITNs had been washed upto 5 times in 57 . 6% , 23 . 0% and 67 . 5% of HHs and 6–10 times in 29 . 6% , 54 . 1% and 32 . 4% in Putijana union; Chouder village; and Bhalukjan , Panch Kushmail , Neogi Kushmail , Baruka villages respectively ( Table 3 ) . Regarding washing practice of nets , 88 . 5% , 88 . 2% and 94 . 9% HHs in the Putijana union; Chouder village; and Bhalukjan , Panch Kushmail , Neogi Kushmail , Baruka villages reported that they washed their nets in the pond ( Table 3 ) which is not recommended . In Putijana union , the majority of the respondents ( 98 . 8% ) said that they dried their nets in direct sunlight ( also not recommended ) , while this was 76 . 5% and 53 . 9% in Chouder village and Bhalukjan , Panch Kushmail , Neogi Kushmail , Baruka villages respectively . In the house-to-house survey , we recorded a total of 1011 VL cases ( 140 . 47/10 , 000/year ) in the three years preceding the respective research projects of which 534 ( 144 . 13/10 , 000/year ) and 477 cases ( 136 . 59 per 10 , 000/year ) in the intervention and control areas respectively ( Table 4; Fig 2 ) . The difference in incidence rate ( IR ) was not statistically significant ( p = 0 . 3901 ) . In the three years after the research projects , we identified a total of 555 KA cases ( incidence rate 77 . 11/10 , 000/year ) of which 178 ( 48 . 04/10 , 000 per year ) in the intervention area and 377 ( 107 . 95/10 , 000 per year ) in control area ( Table 4; Fig 2 ) . The area that benefited from ITN had a significantly lower incidence rate than the control area in the 3-years follow up period , the rate difference was –59 . 91 , p<0 . 0001 . The VL incidence rate during follow-up was significantly reduced both in the intervention and control areas , by 96 . 09/10 , 000/year in intervention area ( p<0 . 0001 ) and 28 . 63/10 , 000/year population in control area ( p<0 . 0001 ) compared to baseline . The effect of the intervention was strongly significant , δ = –67 . 45 , p <0 . 0001 . The estimated reduction of VL incidence rate by the intervention was 46 . 80% ( p<0 . 0001 ) . Moreover , sex ( OR = 1 . 36 , p<0 . 0001 ) and age ( OR = 0 . 99 , p<0 . 0001 ) also had a significant effect on VL incidence . Male were more affected by VL than females . A one year increase in age decreased the likelihood of VL by about 0 . 92% . Seventy five percent of the VL incidence occurred in the age range of 2 to 30 years ( median age of VL patients was 17 years ) .
The main finding of our analysis is that the introduction of ITN in rural highly endemic communities in Bangladesh was associated with a significantly greater reduction of the VL incidence , compared to unexposed communities that also experienced a reduction over time but of lesser size . The present study confirmed that the use of bed nets is a common practice in the rural community of Bangladesh as observed by others [18] . We found that many households in the intervention cohort were still using the nets which had been distributed during the previous studies . A certain proportion of HHs ( about 8% ) were not having bed nets , and those were most likely the poorest families , are mostly related to poverty as it is well established that VL affects the poorest communities in the Indian sub-continent ( ISC ) [26–28] . It has been observed that high coverage of bed net use has community effect on vector sand fly in India and Nepal [23] , similar impact found for malaria vector in Tanzania [29] , so unavailability of bed net in the small number of HHs might not have negative impact . However , the study findings suggest that the washing practices of the ITNs require some change to preserve their effectiveness . Impregnated nets should not dry in the direct sun light as no UV protection is in place in the net . Large number of people dried their nets in the direct sun in the intervention areas which may have reduced their efficacy . It is hard to explain why HHs dried their nets in the direct sun though they were informed to dry nets in the shady place . The possible reason could be HHs want to make sure bed net get dried before sunset in the same day of wash as they may not have extra net . Moreover , it is also recommended that impregnated net should not be washed in ponds or rivers as deltamethrin ( synthetic pyrethroid ) is poisonous for aquatic animals especially for fishes [30] . Unfortunately very few people washed their nets using tube well water . It is well established that VL endemic communities are poorest of the poor , due to this reason many of the study families may not have own tube well which forces them to wash their bed nets in the pond as it is convenient . The strength of our design is that we were able to control for a declining temporal trend by comparing the effect in the intervention area with that of a contemporaneous control area . We also acknowledge two important limitations of our study design . As it is non-experimental in nature , there could be other factors that explain the trend in IRs in the cohorts , such as e . g . a more intense screen-and-treat as the baseline IR were of the highest in the region , and communities might have been targeted preferentially by the programme . We believe the influence of such factor to be minor , as prior to 2009 the elimination programme in Bangladesh was not yet in full swing [12] . At the time , except for some training , little governmental control activities took place . VL patients were in theory entitled to receive all medication free of cost in the government health facilities , but in practice there were severe drug shortages of Sodium Stibogluconate [18] until the introduction of Miltefosine as first-line treatment option in 2009 [8] . It later appeared that one of the batches of Miltefosine supplied by the national programme was a fake drug with no active substance [31] , so we may consider that the effect of case management was minimal during that period . Similarly , no sand fly control activities were conducted by the government up to early 2012 [8 , 12] since banned of DDT in 1997 [32] , as the registration process of deltamethrin for sand fly control took a long time [13] . However , in 2013 the national programme distributed two pieces of LLIN to each patient who had completed VL treatment between 2009 and 2011 [8] . Secondly , our comparison is a one-to-one comparison of one cohort compared to another , and given the erratic behaviour of VL in small areas , the lack of replicates limits the robustness of our findings . Randomization of a sufficient number of study units to either intervention or control cohorts would undoubtedly lead to less biased results , but in the given context of very low case incidence , the organization of such trial is deemed not feasible . Unfortunately , very few studies evaluated the impact of ITN on VL incidence in the ISC . The only study evaluating the impact of local nets impregnated with slow release insecticide on VL in Bangladesh found a 66 . 5% incidence reduction after one year of use in a comparison of one intervention to one control area [33] . Our study showed a significant reduction of VL incidence after three year of use . In Sudan , another observational study found a 59% reduction of VL after using impregnated bed net [34] which is in line with our findings as well . However , our findings contrast with those of pair-matched cluster randomized trial of LLIN distribution in India and Nepal where no VL incidence reduction was found [24] . However , the same study showed a significant reduction on malaria incidence , and the LLIN reduced about 25% P . argentipes sand fly density at household level [23] . The difference between Bangladesh and India/Nepal could be that long-term DDT spraying in India and synthetic pyrethroid spraying in Nepal induced some adaptation of sand fly behaviour towards more outdoor resting or feeding behaviour which is partially supported by a study from India [25] . To eliminate or control a vector-borne disease it is highly important to reduce human-vector contact and vector density . Till today except for IRS no other interventions are included in the vector control strategy of the VL elimination initiative . In the MoU , it was noted that IVM should be adopted as regional strategy for vector control , and this requires more than one tool [9] . Operationally IRS is a more challenging and also more expensive method than ITN distribution . Studies in Bangladesh , India and Nepal identified that the impact of IRS is sub-optimal when it was carried out by the national programme [13 , 16] . Furthermore , VL cases are sharply reducing in the countries so that it will not be sustainable to continue blanket IRS operations in all endemic sub-districts in the country . Health authorities in the region may no longer allocate enough funding for IRS because they have many other health priorities to respond to . It is worth to mention that Bangladesh and Nepal so far did not receive any external funding to control the VL vector in contrast to India ( personal observation , RC ) . In this regard , the present study provides observational evidence of the effect of ITNs in the absence of other governmental control interventions . Given the affordability of ITNs [15] , their ease of implementation and their acceptability , they should be given consideration for inclusion in integrated vector management , definitely in the era of post-VL-elimination [35 , 36] . | Visceral leishmaniasis ( VL ) is a deadly parasitic disease , transmitted by the sand fly species Phlebotomus argentipes in the Indian sub-continent . Humans are the only proven reservoir of the parasite , Leishmania donovani . Effective vector control is highly desirable to reduce vector density and human and vector contact in the endemic communities to stop the disease transmission . We evaluated the effect of long lasting insecticide treated bed nets ( LLIN ) and bed nets impregnated with slow-release insecticide tablet K-O TAB 1-2-3 ( jointly insecticide-treated nets or ITN ) on VL incidence in a highly endemic sub-district ( upazila ) in Bangladesh . The nets were either donated or impregnated between 2004 to 2008 under three studies and defined as “exposed cohort” comparing their effect on VL incidence with “non-exposed cohort” ( no donation of impregnated nets ) for a 3 year period before and after the intervention . The study identified 1011 VL cases ( IR 140 . 47/10 , 000 per year [py] ) before the intervention , of which 534 and 477 cases in the intervention and control areas respectively . There was a strong and significant overall effect of the ITN intervention , δ = –67 . 45 , p <0 . 0001 . The VL incidence rate was significantly lower in the ITN intervention cohort compared to control in Bangladesh , though some bias cannot be totally ruled out . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"kala-azar",
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"tropical",
"diseases",
"geographical",
"locations",
"sand",
"flies",
"parasitic",
"diseases",
"animals",
"neglected",
"tropical",
"diseases",
"infectious",
"disease",
"control",
"insect",
"vectors",
"bangladesh",
"agrochemicals",
"infectious",
"diseases",
"zoonoses",
"protozoan",
"infections",
"disease",
"vectors",
"insects",
"agriculture",
"arthropoda",
"insecticides",
"people",
"and",
"places",
"mosquitoes",
"eukaryota",
"asia",
"leishmaniasis",
"biology",
"and",
"life",
"sciences",
"species",
"interactions",
"organisms"
] | 2019 | Effect of insecticide-treated bed nets on visceral leishmaniasis incidence in Bangladesh. A retrospective cohort analysis |
Human African trypanosomiasis progresses from an early ( hemolymphatic ) stage , through CNS invasion to the late ( meningoencephalitic ) stage . In experimental infections disease progression is associated with neuroinflammatory responses and neurological symptoms , but this concept requires evaluation in African trypanosomiasis patients , where correct diagnosis of the disease stage is of critical therapeutic importance . This was a retrospective study on a cohort of 115 T . b . rhodesiense HAT patients recruited in Eastern Uganda . Paired plasma and CSF samples allowed the measurement of peripheral and CNS immunoglobulin and of CSF cytokine synthesis . Cytokine and immunoglobulin expression were evaluated in relation to disease duration , stage progression and neurological symptoms . Neurological symptoms were not related to stage progression ( with the exception of moderate coma ) . Increases in CNS immunoglobulin , IL-10 and TNF-α synthesis were associated with stage progression and were mirrored by a reduction in TGF-β levels in the CSF . There were no significant associations between CNS immunoglobulin and cytokine production and neurological signs of disease with the exception of moderate coma cases . Within the study group we identified diagnostically early stage cases with no CSF pleocytosis but intrathecal immunoglobulin synthesis and diagnostically late stage cases with marginal CSF pleocytosis and no detectable trypanosomes in the CSF . Our results demonstrate that there is not a direct linkage between stage progression , neurological signs of infection and neuroinflammatory responses in rhodesiense HAT . Neurological signs are observed in both early and late stages , and while intrathecal immunoglobulin synthesis is associated with neurological signs , these are also observed in cases lacking a CNS inflammatory response . While there is an increase in inflammatory cytokine production with stage progression , this is paralleled by increases in CSF IL-10 . As stage diagnostics , the CSF immunoglobulins and cytokines studied do not have sufficient sensitivity to be of clinical value .
Human African trypanosomiasis ( HAT ) , also known as Sleeping Sickness , is caused by the protozoan hemoflagellate Trypanosoma brucei ssp . After inoculation of the parasite by the tsetse fly vector , the disease progresses through two stages . In the hemolymphatic , or early stage of disease , parasites proliferate in the blood and lymphatic system . In the meningoencephalitic , or late stage , parasites penetrate the blood brain barrier ( BBB ) and persist and proliferate in the CNS , causing an encephalitic reaction that leads to death if untreated or inadequately treated [1] . Two sub-species of African trypanosome give rise to HAT . T . b . gambiense is endemic to West and Central Africa , with a chronic course of infection in which late stage may not commence for months or years after infection , and for which there is recent evidence for asymptomatic infection [2] , [3] , [4] . T . b . rhodesiense is endemic in East and Southern Africa , is distinguished by the SRA ( serum resistance associated gene ) and exhibits a more acute pattern of progression than T . b . gambiense , although there is considerable diversity in progression rate that may be related to parasite virulence variation and host immunogenetics [5] . Animal model studies of T . brucei infection demonstrate that dysregulated inflammatory responses are a major contributor to the pathophysiology of infection , both systemically [6] and in the brain [7] , [8] , where it was hypothesised that the development of neuropathology is associated with an astrocytosis regulated by the CNS inflammatory/counter-inflammatory cytokine balance [9] . In humans , direct measurements of immune cell activation in the brain are not possible for obvious ethical reasons . While gross inflammatory pathology analogous to that observed in rodent models has been described in post-mortem material [10] , our limited understanding of the pathophysiology of CNS infection in HAT derives from the observation of neurological symptoms and analysis of patients' cerebrospinal fluid ( CSF ) [11] , [12] , [13] . A spectrum of neurological symptoms is observed in HAT infection . This includes sleep , sensory , motor and psychiatric disorders as well as the characteristic sleep disturbances that have given this disease its common name of Sleeping Sickness [1] . Staging is critical to therapeutic decision making as late stage infections of T . b . rhodesiense are currently treated with arsenical drugs that induce a severe and sometimes fatal reaction known as the post-treatment reactive encephalopathy ( PTRE ) in about 10% of treated patients , half of whom die as a result giving an overall drug mortality of 5% [12] . Currently , disease staging primarily relies on the detection of trypanosomes in the CSF and/or an elevation in the CSF white blood cell ( WBC ) count . The most widely applied diagnostic cut off for CSF WBC counts to indicate a late stage infection ( WHO criteria ) is >5 cells/µl [14] , although in the case of T . b . gambiense infection there is some evidence that this value is too low and that effective early stage treatment may still be administered in patients with up to 20 WBC/µl in the CSF . It has also been proposed that HAT cases with between 5 and 20 CSF WBC/µl fall into an intermediate stage category , regardless of whether trypanosomes are detected in the CSF [11] . In addition to CSF cell counts , other biochemical and immunological markers have been investigated to improve the sensitivity and specificity of diagnostic staging . In late stage HAT there is an increase in CSF protein level that is largely accounted for by immunoglobulin ( Ig ) expression . High levels of intrathecal IgM synthesis are typical and have been shown to be a sensitive marker for intrathecal inflammatory responses and therefore of stage diagnostic value in T . b . gambiense infections [15] . A number of stage-specific alterations in cytokine and chemokine response have also been described in CSF from HAT patients [8] , [16] , [17] , [18] , and IL10 has been proposed as a potential diagnostic marker for infection and cure owing to the speed with which levels return to normal after treatment [19] . As part of a study of the clinical evolution of T . b . rhodesiense HAT [20] , we have analysed in detail the parasitological and clinical progression of HAT in a cohort of patients recruited in Serere in 2003 , in Eastern Uganda . Because the parasites circulating in this epidemic were genetically homogeneous [21] , and the host population comprised a single ethnolinguistic group , this set of cases offers an opportunity to describe the evolution of HAT independent of variations in parasite virulence , and to use the clinical data to explore the relationship of inflammatory responses in the CNS to clinical disease . In particular , we tested the hypothesis that disease progression and neurological dysfunction would be associated with increasing inflammatory ( agonist ) and decreasing anti-inflammatory ( antagonist ) responses . We also evaluated the stage diagnostic potential of CSF immunoglobulin and cytokine responses in T . b . rhodesiense infection .
This study was conducted according to the principles expressed in the Declaration of Helsinki . All patients recruited received written and verbal information explaining the purpose of this study and gave informed written consent . All protocols were approved by ethics committees in Uganda ( Ministry of Health ) and UK ( Grampian Joint Ethics Committee ) . Ethical consent forms were designed in English and also translated into local languages . Consent was given as a signature or a thumb-print after verbal explanation . For those under 16 consent was given by their legal guardian , and for those whose clinical condition prohibited full understanding of the recruitment process , consent was gained from a spouse or other family member . 115 HAT patients were recruited at Serere Health Centre , Serere District , Eastern Uganda , between August 2002 and July 2003 . This cohort of cases was drawn from of a larger multi-centre study , for which study sites , recruitment protocols , treatment regimens , disease progression characteristics and clinical examination methods have been published elsewhere [20] . All patients belonged to the Ateso ( Eastern Nilosaharan ) ethnolinguistic group . Patients with intercurrent infections of malaria , filariasis or schistosomiasis were excluded from the study . Staging was carried out in accordance with WHO criteria [14] . These define late stage by the presence of parasites in the lumbar CSF and/or a CSF WBC>5/µl in parasitemic individuals . In this cohort , parasite counts in the CSF were determined by Neubauer hemocytometer , and the definitive presence or absence of parasites in the CSF by double centrifugation [22] . Plasma and lumbar CSF were collected from all patients as part of routine diagnostic and stage determination procedures . Paired plasma and CSF samples were frozen in liquid nitrogen within 1 h of collection , and maintained in liquid nitrogen until required for analysis . Control plasma and CSF samples for cytokine analysis were obtained from 17 HAT suspects at Serere Health Centre and 18 HAT suspects at the LIRI Health Centre , Tororo , Uganda respectively who were all later diagnosed as non-infected . In some assays , limitations on the volume of plasma and CSF available meant that a subset of patient samples that was effectively randomly selected was analysed . Cytokines IFN-γ , IL-6 and IL-10 concentrations were measured in CSF using a solid phase analyte capture sandwich ELISA ( OptiEIA set , Becton Dickinson-Pharmingen , Oxford , U . K . ) as previously described in [23] . Free TNF-α was measured using a receptor binding assay as described previously ( BioLISA , Bender Med Systems , Wien , Austria ) [24] . Cytokine assays limits of detection were IFN-γ:1 . 8 pg/ml; IL-10:1 . 6 pg/ml; TGF-β:19 . 2 pg/ml; IL-6:8 . 3 pg/ml; TNF-α:22 pg/ml . For descriptive and inferential statistical analysis , results below the limit of detection were assumed to be ( 0 . 5× limit of detection value ) . As rank statistical methods were used this assumption did not bias significance tests . Total IgM , IgA , IgG and albumin were determined by nephelometry ( ProSpec , Dade-Behring , Marburg , Germany ) as described in [15] . Blood-CSF barrier function was evaluated using the albumin quotient ( QALB ) . The cut off for dysfunction was calculated using the formulaIntrathecal immunoglobulin synthesis ( IgLOC ) was evaluated using the method of Reiber [25] . Briefly , this analytic approach is based on a reference set of 4300 normal CSF samples , from which an upper hyperbolic discrimination curve QLIM defines the upper limit of the immunoglobulin quotient ( Q ( Ig ) ) in the absence of intrathecal Ig synthesis . Intrathecal Ig synthesis results in a Q ( Ig ) lying above QLIM . The level of intrathecal synthesis of each isotype ( IgLOC ) is derived from the formulaThe results of this analysis were presented in quotient diagrams [26] using CSF Statistics Tool software ( CoMed GmbH , Soest , Germany ) . None of the continuous variable parameters examined could be transformed to normality . Therefore differences between groups were tested using the Mann-Whitney U-test , or the Kruskal Wallis test followed by Dunn's post-hoc test . Bivariate correlations were evaluated using Spearmann's rank correlation coefficient ( rs ) . Diagnostic outcomes as dependent variables were tested on cytokine concentrations using logistic regression , and diagnostic power was assessed using receiver operating characteristic analysis . Diagnostic panel candidates were selected using mixed stepwise logistic regression and optimal diagnostic cut offs were determined using the prediction profiler in JMP6 . 0 ( SAS Institute , Cary , NC , USA ) .
The study population comprised 35 early stage and 80 late stage patients . The age , gender and diagnostic stage of these individuals are presented in Table 1 . There were no significant associations between either gender or age and infection stage . Of the 80 late stage cases , 77 were confirmed by detection of trypanosomes in the CSF after double centrifugation . The remaining 3 where trypanosomes could not be detected exhibited WBC counts between 6 and 20/µl . Progression to late stage in this focus was rapid with a median reported duration of illness of 8 weeks for late stage cases . Bloodstream parasitemias were scored on thick blood films . Late stage median parasitemia was lower than in early stage , but this difference was not significant ( Table 2 ) . However , parasitemia was significantly inversely correlated to reported duration of illness . In contrast , both CSF trypanosome and WBC counts increased significantly with increasing duration of illness ( rs = 0 . 53 p<0 . 0001 and 0 . 42 p<0 . 0001 respectively ) . Also , as would be expected given that CSF trypanosome and WBC are diagnostic criteria , both were significantly higher in late stage cases compared to early stage cases . Total CSF protein was significantly higher in late stage cases compared to early stage cases and also increased in relation to duration of disease ( Table 2 ) . Plasma albumin concentration was below the normal reference range for European populations [15] in both early and late stage cases and decreased with disease progression as measured by both disease stage and reported duration of illness ( Table 2 ) . In contrast , CSF albumin concentration was significantly higher in late stage cases compared to early stage cases . Likewise , the albumin quotient was significantly increased in late stage cases compared to early stage cases and also correlated to reported duration of disease ( Table 2 ) . Using the age related cut off for normal albumin quotient ( QALB ) , blood brain barrier ( BBB ) dysfunction was indicated in 6% of early stage cases and 42% of late stage cases . Plasma IgG and IgA levels ( Table 2 ) did not differ significantly between early and late stage cases and fell within the normal European reference range [15] . However plasma IgM levels in both early and late stage cases were increased above the reference range and increased with disease progression as estimated by stage of infection and duration of disease . In early stage CSF samples , all Ig isotype concentrations were within the reference range . However , in late stage CSF samples IgM , IgG and IgA concentrations were significantly increased compared to both early stage and normal reference range values , and also were significantly correlated with duration of disease . Intrathecal synthesis of immunglobulins was determined ( mg/l ) in relation to the upper hyperbolic discrimination line ( QLim ) [26] in quotient diagrams ( Figure S1 ) . Intrathecal Ig synthesis was detected in 12% early and 74% late stage cases ( Table 3 ) . The proportion of intrathecally synthesised immunoglobulin in relation to total CSF concentration also varied according to isotype , reaching a median of 44% in the case of IgM synthesis in late stage cases . Overall , of those late stage cases ( n = 54 ) where intrathecal IgM synthesis was detected , intrathecal synthesis of a second isotype occurred in 61% ( IgA ) and 35% ( IgG ) of cases respectively . Intrathecal synthesis of all 3 isotypes was detected in 30% of these cases . There was also a significant correlation of intrathecal immunoglobulin synthesis with duration of illness , CSF trypanosome concentration and CSF white cell concentration for all isotypes . The CSF IL-10 ( Figure 1a ) concentration was significantly increased over control in both early and late infection stages and also increased with progression from early to late stage . The IL-10 concentration also showed a significant positive correlation with disease duration ( rs = 0 . 41 p<0 . 001 ) . There was no significant difference between early and late stage cases for IFN-γ ( Figure 1b ) . TNF-α levels ( Figure 1c ) in all control and early stage CSF samples were below the assay limit of detection , but were detectable at a significant level in 5/21 late stage cases . TGF-β levels ( Figure 1d ) were significantly higher in early stage samples compared to late stage although there was no significant correlation with duration of infection . The CSF IL-6 ( Figure 1e ) concentration was increased above control levels in both early and late stage cases , there was also significant correlation to duration of infection ( rs = 0 . 46 p<0 . 01 ) . When the relationships between the expression of each of the CSF cytokines were evaluated , significant positive relationships were identified between TNF-α and IL-10 ( rs = 0 . 42 p<0 . 01 ) ; IL-6 and IL-10 ( rs- = 0 . 38 p<0 . 05 ) ; IL-6 and IFN-γ ( rs = 0 . 37 p<0 . 05 ) and a negative relationship between TGF-β and IL10 ( rs = −0 . 4 p<0 . 001 ) . IL-10 and TGF-β concentrations were significantly predictive of diagnosis stage in univariate logistic regression ( likelihood ratio test p<0 . 0001 for both cytokines ) . Each log unit increase in CSF IL-10 and TGF-β concentrations were associated with odds ratios ( OR [95% CI] ) for late stage diagnosis of 4 . 0 ( 2 . 5–7 . 4 ) and 0 . 04 ( 0 . 01–0 . 0 . 17 ) respectively . To determine if either would be of utility as a late stage diagnostic , receiver operating characteristic ( ROC ) curves were analysed . For IL-10 the area under the ROC curve ( AUROCC ) was 0 . 85 but in order to achieve 100% specificity a cut off value of 275 pg/ml only offered 14% sensitivity . For TGF-β , the AUROCC was 0 . 86 , and similarly the cut off for 100% specificity ( 50 pg/ml ) only offered 5% sensitivity . An optimal combined panel of CSF IL-10 , TGF-β , and IgM concentration was identified using stepwise logistic regression analysis and with discriminatory cut off levels for late stage determined on the patient data of IL-10>66 . 4 pg/ml; TGF-β<159 . 5 pg/ml and CSF IgM>89 . 2 mg/l . This panel offered an AUROCC of 0 . 97 , with 70% sensitivity for 100% late stage specificity . Neurological signs including altered gait , tremors , incontinence , cranial nerve neuropathy ( facial nerve palsies ) , somnolence and reduced Glasgow coma score ( GCS ) were observed in both early and late stage patients , indicating early onset of neurological involvement , with only moderate coma ( GCS<12 ) being unrepresented in any early stage cases ( Table 4 ) . Of the neurological signs assessed , only somnolence was observed to show a significant ( p<0 . 01 ) increase in frequency in relation to duration of illness ( OR for each log unit increase in duration of illness = 2 . 1 ( 95% CI = 1 . 3–3 . 8 ) ) . The relationships of neurological signs to intrathecal immunoglobulin and cytokine synthesis were examined . There were no significant differences in either intrathecal immunoglobulin levels ( Igloc ) or cytokine concentrations ( IL-10 , IFN-γ , TNF-α , TGF-β or IL-6 ) in relation to the presence or absence of gait ataxia , tremors , urinary incontinence , although cases with facial nerve palsies exhibited a slight increase in intrathecal IgA concentration ( median ( IQR ) 0 . 0 ( 0 . 0–4 . 6 ) mg/l versus 0 . 0 ( 0 . 0–0 . 8 ) mg/l p<0 . 05 ) . However , cases with moderate coma ( GCS<13 ) exhibited substantially and significantly higher levels of intrathecal synthesis of all Ig isotypes as well as IL-10 and IL-6 ( Table 5 ) . This effect was not observed in cases with mild coma ( GCS 13–14 ) . In T . b . gambiense infection , it has been proposed that CSF white cell counts of 5–20/µl , regardless of whether trypanosomes are detected in the CSF , should be regarded as an intermediate stage [11] that may be treated with pentamidine . In the cohort of T . b . rhodesiense patients described in this paper , only 3 individuals fell into this category . These patients were treated successfully with melarsoprol and followed up for two years . The characteristics of these subjects are presented in Table 6 . All presented with normal GCS , gait , and an absence of somnolence . All exhibited normal BBB function . Unlike the late stage cases classified with white cell counts >20/µl that all were positive for CSF trypanosomes after double centrifugation , all three of these possible “intermediate” stage cases were negative for CSF trypanosomes . However 1/3 exhibited incontinence and cranial neuropathy . Furthermore , 2/3 exhibited intense intrathecal IgM synthesis , and 1/3 intrathecal IgA and IgG synthesis ( Table 6 and Figure 1b closed triangles ) .
Our understanding of pathogenesis in HAT is limited by the logistic difficulties of clinical studies in endemic areas . Studies in animal models suggest that the development of systemic and CNS pathology follows a dysregulation of host-inflammatory responses , and that host-immune response variation may control the severity of pathology [27] . African trypanosome CNS infection model studies indicate that the development of neuropathology results from inflammatory responses in the brain and it has been hypothesised that pathogenesis may be controlled or limited by counter-inflammatory responses [8] . In this study we selected a cohort of T . b . rhodesiense HAT patients from Eastern Uganda who were exposed to a genetically homogeneous parasite population [21] and were drawn from a homogeneous ethnolinguistic ( Ateso ) host population . This approach enables us to minimise any confounding effects of natural parasite variability [5] on the parameters and also to a limited extent the effects of host genetic diversity on disease progression assuming the close relationship between language and genetic variation observed in other nilo-saharan populations [28] . We analysed the development of CNS humoral and cellular immune responses and their relationship to disease progression and neurological signs of HAT . Disease progression was assessed both in relation to diagnostic staging and the reported duration of disease . Development of disease in the study group reported here was rapid , with a median disease period of 8 weeks for late stage infection , consistent with results previously reported from a larger cohort of patients from the Soroti focus [20] . Overall parasitemia levels were highest in early stage cases and fell in relation to duration of infection . In the plasma , IgM concentration increased with both progression from early to late stage and reported disease duration . No similar increases were evident for plasma IgA and plasma IgG , which also remained within or close to the reference range concentrations described for healthy European subjects . This result is consistent with observations in both mouse and bovine models of a predominant polyclonal B-cell activation in trypanosomiasis [29] . Although IgM responses have been previously described in T . b . gambiense HAT , there has only been one previous report of elevated IgM in the serum and CSF of T . b . rhodesiense HAT patients [30] . Plasma albumin levels fell with disease progression . While this phenomenon may be expected as part of a negative acute phase response in early infection [31] , in more chronic infection it is possibly also a consequence of liver pathology , albuminuria resulting from kidney damage , and cachexia [32] . In the CSF , total protein increased with stage progression and reported disease duration , consistent with previous observations for both T . b . gambiense and T . b . rhodesiense . The CSF protein level has been recognised as a useful stage diagnostic tool in HAT [14] but it is interesting to note that in this study of T . b . rhodesiense patients , the median late stage CSF protein concentration is lower than the recommended diagnostic cut off for staging employed in T . b . gambiense cases ( 370 mg/l ) . The increase in CSF protein is accounted for at least in part by the large and significant increases in CSF IgM , IgG and IgA concentration . Similar increases have been described previously for T . b . gambiense infection [15] . The increase in CSF immunoglobulin level is a product of both intrathecal immunoglobulin synthesis and accumulation of serum-derived immunoglobulins in the CSF as a result of a reduction in CSF flow and turnover rate [25] , [33] . This is evident in this study from the significant increase in QALB levels with both stage progression and disease duration , and indicates a reduction of either CSF production in the choroid plexus or outflow of CSF into venous circulation . In order to measure intrathecal synthesis of immunoglobulin we used the hyperbolic discrimination curve QLIM as cut off for non-CNS derived Ig [33] . Intrathecal Ig synthesis ( IgLOC ) was detected in few ( 12% ) early stage cases but commonly ( 70% ) in late stage cases . In contrast to the findings from similar studies in T . b . gambiense infection [15] , intrathecal Ig synthesis in T . b . rhodesiense HAT was predominantly a single class ( IgM ) response and 2 or 3 class responses were considerably less common . Overall , while intrathecal Ig synthesis in late stage infection provides clear evidence of the activation of humoral immune responses by trypanosomes in the CNS , it does not offer the sensitivity or specificity required to be effective as a stage diagnostic . None of the early stage cases in this study where intrathecal Ig synthesis was detected relapsed ( over a 1 year follow up period ) after suramin treatment , indicating that the diagnostic decision to classify these cases as early stage was correct . In T . b . gambiense infection a similar frequency of early stage cases exhibiting Ig synthesis has been described [15] . CSF cytokine concentrations were measured as indicators of cellular immune activation in the CNS . Increases in IL-10 and IL-6 concentration with stage progression are consistent with previous studies in both T . b . rhodesiense [19] and T . b . gambiense infection [16] . In this study we further observed an increase in TNF-α concentration and a decrease in TGF-β concentration with progression to late stage . The increases in CSF cytokine concentrations were not restricted to late stage cases only . CSF concentrations of IL-6 and IL-10 were also elevated over control levels in early cases , and provide evidence of early activation of CNS cellular responses and suggest , as has been shown in rodent models [10] , that there may be very early CNS involvement in HAT at a stage when patients are diagnostically classified as early stage and effectively treated with suramin . While further research is required into this phenomenon , one possibility is raised by observation that trypanosomes may readily penetrate the vascular endothelial basement membrane in some regions of the brain while still being unable to traverse the parenchymal basement membrane [34] . Such a process would bring trypanosomes in contact with astrocytes , and thus initiate neuroinflammatory responses . An increase in TNF-α concentration was only observed in late stage cases , and therefore is consistent with the mouse model of CNS infection [8] . With respect to the reduction in CSF TGF-β concentration with stage progression , while we were unable to measure CSF TGF-β in a sympatric control population , published data on CSF TGF-β levels in normal subjects ( 266 pg/ml [35] ) suggest that in both early and late stage of infection TGF-β CSF concentrations are reduced below control levels . The reciprocal relationship of TGF-β and TNF-α concentrations in the CSF observed in disease progression is consistent with the mutually antagonistic anti- and pro-inflammatory properties of these two cytokines . TNF-α is a mediator of inflammatory neuropathology and its expression has been observed in association with astrocyte activation in mouse models of late stage HAT [36] as well as having been associated with disease severity in HAT [37] . TGF-β functions as a regulatory cytokine that can modulate inflammatory reactions in the CNS [38] , for example through suppression of pro-inflammatory TNF-α expression in astrocytes [39] , and the host TGF-β response has been implicated in the mild presentation of HAT observed in Malawi [23] . However , this does not account for the increase of CSF IL-10 with disease progression , as IL-10 is also an anti-inflammatory mediator [40] . The strong correlation between TNF-α and IL-10 levels in the CSF indicates the activation of distinct cellular compartments in the CNS during infection , and identification of the CNS cellular sources of IL-10 , TNF-α and TGF-β in HAT will require further work in model systems , and to determine if distinct subsets of brain macrophages are involved in inflammatory/counterinflammatory regulation as has been observed in in vitro models of murine brain macrophage activation [41] . We observed no difference in CSF IFN-γ concentration between early and late stage . This result is not consistent with the close relationship of brain IFN-γ synthesis with disease progression described in the mouse model of African trypanosomiasis [8] . Cytokines in the CSF have previously been proposed as potential stage diagnostic markers [16] , [19] . In this study , while IL-10 and TGF-β were predictive of diagnostic stage , they were insufficiently sensitive to be developed as effective staging markers , either individually or in combination with CSF IgM concentration . Through analysis of immunoglobulin and cytokine responses in the CNS in relation to stage progression in T . b . rhodesiense HAT , it is possible to test the hypothesis that neurological dysfunctions observed in HAT are manifestations of inflammatory neuropathology . Neurological signs ( gait abnormalities , tremors , incontinence , cranial nerve neuropathy , somnolence and mild coma ) were equally probable to be observed in both early and late stage cases , and the only significant difference in incidence of neurological symptoms was for moderate coma which was never observed in early stage cases . This result is consistent with observations in a larger multi-centre study in Uganda [20] . While somnolence was equally probable in early and late stage cases , its incidence did increase with increasing reported duration of infection . We then analysed whether any relationship existed between CNS humoral and cytokine responses and neurological signs . Intrathecal Ig and CSF cytokine synthesis did not vary according to the presentation of gait ataxias , tremors , incontinence , facial nerve palsies , somnolence and mild coma . Therefore these neurological symptoms of HAT may have a non-immunological basis . However , individuals presenting with moderate coma presented elevated intrathecal IgG , IgA and IgM synthesis as well as significantly increased CSF IL-10 and IL-6 concentrations . In one respect this finding is to be expected , as of all the neurological signs that were investigated , moderate coma was only observed in diagnostic late stage cases . However , the moderate coma cases showed no significant increase in CSF IFN-γ , TNF-α or TGF-β in relation to mild or no coma cases . This suggests that inflammatory cytokine responses including TNF-α and IFN-γ do not increase with disease severity and this is not consistent with findings in experimental models and other clinical studies [10] . Finally , we observed a small number ( 3 ) of cases of HAT with CSF WBC concentration 6 and 20/µl . These cases were classified as late stage cases , however it was noted that these were the only 3 cases in the study where trypanosomes were not detected in the CSF even after double centrifugation . It is intriguing to speculate that these cases might present the first indication that , as is the case in T . b . gambiense infection , there may be an “intermediate” or “early-second stage” of infection [13] , [15] in T . b . rhodesiense HAT . However such an interpretation would require the study of considerably more cases of this type , and while all three cases showed normal BBB function , 2 of the cases presented an intense intrathecal IgM synthesis ( Fig . 1b ) despite neither of these individuals showing any neurological signs . In conclusion , in T . b . rhodesiense HAT , increasing levels of intrathecal immunoglobulin synthesis and CNS pro-inflammatory cytokine expression were associated with disease progression from early to late stage , although these were of limited diagnostic value . While intrathecal immunoglobulin synthesis was associated with the development of coma , it was not associated with any of the other typical neurological sequelae of HAT , which were also not related to CSF inflammatory or counter-inflammatory cytokine levels . Neuroinflammatory responses in correctly diagnosed early stage cases and cases with a similarity to the intermediate stage of T . b . gambiense HAT suggest there may be an early CNS involvement prior to the detectable invasion of the brain by the parasite and that effects on the CNS may be mediated indirectly by the parasite while it is still localised in the haemolymphatic system . | Human African trypanosomiasis , caused by the parasites Trypanosoma brucei rhodesiense and T . b . gambiense , is clinically defined by two diagnostic stages , an early stage where the parasites appear to be localised to the blood and lymphatic systems , and a late stage where the parasites are also localised in the central nervous system ( CNS ) and cause a meningoencephalitis , which is fatal if untreated . We studied the progression between these stages of infection in T . b . rhodesiense infections in Uganda . Progression from early to late stage was associated with an increase in inflammatory responses in the CNS as measured by analysis of the cerebrospinal fluid . However , contrary to predictions based on experimental model studies , neither disease stage progression nor CNS inflammatory responses were directly associated with development of neurological symptoms . Our results suggest that biological basis of the boundary between the two diagnostic stages in this infection may not be clear cut , with implications for therapeutic decision making . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"infectious",
"diseases",
"clinical",
"immunology",
"immunology",
"parasitic",
"diseases",
"immune",
"response"
] | 2012 | Stage Progression and Neurological Symptoms in Trypanosoma brucei rhodesiense Sleeping Sickness: Role of the CNS Inflammatory Response |
Pentameric ligand-gated ion channels ( pLGICs ) are neurotransmitter-activated receptors that mediate fast synaptic transmission . In pLGICs , binding of agonist to the extracellular domain triggers a structural rearrangement that leads to the opening of an ion-conducting pore in the transmembrane domain and , in the continued presence of neurotransmitter , the channels desensitize ( close ) . The flexible loops in each subunit that connect the extracellular binding domain ( loops 2 , 7 , and 9 ) to the transmembrane channel domain ( M2–M3 loop ) are essential for coupling ligand binding to channel gating . Comparing the crystal structures of two bacterial pLGIC homologues , ELIC and the proton-activated GLIC , suggests channel gating is associated with rearrangements in these loops , but whether these motions accurately predict the motions in functional lipid-embedded pLGICs is unknown . Here , using site-directed spin labeling ( SDSL ) electron paramagnetic resonance ( EPR ) spectroscopy and functional GLIC channels reconstituted into liposomes , we examined if , and how far , the loops at the ECD/TMD gating interface move during proton-dependent gating transitions from the resting to desensitized state . Loop 9 moves ∼9 Å inward toward the channel lumen in response to proton-induced desensitization . Loop 9 motions were not observed when GLIC was in detergent micelles , suggesting detergent solubilization traps the protein in a nonactivatable state and lipids are required for functional gating transitions . Proton-induced desensitization immobilizes loop 2 with little change in position . Proton-induced motion of the M2–M3 loop was not observed , suggesting its conformation is nearly identical in closed and desensitized states . Our experimentally derived distance measurements of spin-labeled GLIC suggest ELIC is not a good model for the functional resting state of GLIC , and that the crystal structure of GLIC does not correspond to a desensitized state . These findings advance our understanding of the molecular mechanisms underlying pLGIC gating .
Chemical signaling in the brain and periphery relies on the rapid opening and closing of pentameric ligand-gated ion channels ( pLGICs ) , which include nicotinic acetylcholine ( nAChRs ) , serotonin-type-3 ( 5-HT3Rs ) , γ-aminobutyric acid-A ( GABAARs ) , and glycine ( GlyRs ) receptors [1] . These receptors exist in at least three distinct , interconvertible states: resting ( unliganded , closed channel ) , activated ( liganded , open channel ) , and desensitized ( liganded , closed channel ) , and the binding of agonists , antagonists , and allosteric drugs alters the equilibria between these states . Neurotransmitter binding in the extracellular ligand-binding domain triggers rapid opening of an intrinsic ion channel more than 60 Å away in the transmembrane domain of the receptor , and with prolonged neurotransmitter exposure , the channel moves into a nonconducting desensitized state . Although we know a fair amount about the structure of these receptors , the mechanisms by which the binding of neurotransmitter triggers channel opening and desensitization are still unfolding , and our understanding of the protein motions underlying these processes is limited . pLGICs are composed of five identical or homologous subunits arranged pseudosymmetrically around a central ion-conducting channel . Our current structural knowledge of these proteins comes from cryo-EM structures of the Torpedo nAChR in a presumed unliganded closed state ( 4 Å resolution ) and liganded open state ( 6 . 2 Å resolution ) [2] , [3] , high-resolution crystal structures of the extracellular binding domains of the nAChR α1 and α7 subunits [4] , [5] , crystal structures of full-length prokaryotic pLGIC homologs from Erwinia chrysanthemi ( ELIC ) and Gloeobacter violaceus ( GLIC ) solved in presumed closed and open channel conformations [6]–[8] , respectively , and a recent crystal structure of a glutamate-activated chloride channel ( GluCl ) in an open channel conformation from C . elegans [9] . In general , each subunit can be divided into two parts: an extracellular binding domain ( ECD ) folded into a β-sandwich core and a transmembrane channel domain ( TMD ) consisting of four α-helical membrane-spanning segments ( M1 to M4 ) . Neurotransmitter binding occurs at sites located at interfaces between subunits in the ECD ( reviewed in [1] ) , and the M2 helices of each of the subunits form the ion-conducting channel . In each subunit , flexible loops ( loop 2 , loop 7 , loop 9 , and the M2–M3 loop ) connect the binding domain to the channel domain ( Figure 1 ) and play a critical role in coupling binding site movements to the channel [1] . Comparison of ELIC and GLIC structures suggests that channel activation is associated with an anticlockwise concerted twist of each ECD β-sandwich and a radial tilting of the pore lining M2 α-helices away from the channel axis [7] , [10] . Rearrangements in the flexible loops that form the interface between the ECD ( loop 2 , loop 7 , and loop 9 ) and the TMD ( M2–M3 loop ) ( Figure 1 ) are also observed . Some studies , however , suggest that the GLIC structure may correspond to a desensitized state [11] , [12] and the ELIC structure to an “uncoupled” nonfunctional conformation [11] , [13] . Thus , whether the motions inferred by comparing the static structures of two related ( but with only 18% sequence identity ) proteins solved in detergent micelles in uncertain functional states accurately predict the dynamic gating motions of a pLGIC in its native environment is unknown . In this study , we used site-directed spin labeling ( SDSL ) electron paramagnetic resonance ( EPR ) spectroscopy and functional GLIC channels reconstituted into liposomes to examine if , and how far , the loops at the ECD/TMD gating interface move during proton-dependent gating transitions from the resting to desensitized state . SDSL EPR spectroscopy is a powerful method for monitoring the structure and dynamics of membrane proteins in conditions closely resembling the proteins' native environment [14] , [15] . In SDSL EPR , a cysteine residue is introduced at a site of interest , and a sulfhydryl-specific nitroxide reagent ( typically 1-oxyl-2 , 2 , 5 , 5-tetramethyl-3-pyrroline-3-methyl methanethiosulfonate spin label , MTSL ) is covalently attached to the free sulfhydryl as a paramagnetic probe to create the R1 side chain ( Figure S1A ) . Backbone and side chain mobility can be detected with the continuous wave ( CW ) method , and distances and distance changes between pairs of probes can be measured by double electron electron resonance ( DEER ) spectroscopy ( up to ∼60 Å ) [16] . SDSL EPR spectroscopy is the ideal complement to high-resolution static snapshots of crystal structures and has been used successfully to study the dynamic motions of the voltage-gated K+ channel [17]–[19] , other membrane proteins ( e . g . , the mechanosensitive channel of small conductance , MscS [20] , and rhodopsin [21] ) and recently , GLIC [22] , [23] . Here , we show that proton-dependent GLIC gating from resting to desensitized conformation induces a large inward movement of loop 9 towards the channel lumen and an immobilization of loop 2 , which is accompanied by substantial rearrangements of the intra- and intersubunit interface between the ECD and TMD . No appreciable proton-induced motions in the M2–M3 loop in the TMD were detected , demonstrating the conformation of this critical loop is similar in resting ( closed , unliganded ) and desensitized ( closed , liganded ) states . Proton-induced motions in GLIC were absent when the protein was in detergent micelles , indicating that lipids are required for functional gating transitions and suggesting that the detergents used for protein solubilization and crystallization may influence the conformations captured in the crystal structures . In general , residue positions and the proton-induced motions in functional GLIC protein embedded in lipid differ from those predicted based on the crystal structures of GLIC and ELIC obtained in detergent micelles , suggesting the GLIC crystal structure does not correspond to a desensitized conformation and that ELIC is not a suitable model for the structure of the M2–M3 loop of GLIC in the resting , closed state .
To study proton-induced motions in loops forming the ECD/TMD interface of GLIC by EPR spectroscopy , we generated a cys-free GLIC mutant by replacing the lone native cysteine , C26 , with alanine ( Figure 1A , B ) . We then individually mutated K32 ( loop 2 ) , T157 ( loop 9 ) and K247 and P249 ( M2–M3 loop ) to cysteine in the mutant C26A background ( Figure 1A , B ) . We expressed wild-type and mutant proteins in Xenopus laevis oocytes and measured proton-induced currents using two-electrode voltage clamp ( Figure S1B ) . All of the mutants formed functional channels with wild-type GLIC properties ( pH50 = 5 . 2±0 . 1 , Hill coefficient nH = 1 . 6±0 . 1 ) . We also measured currents elicited by pH50 concentrations before and after reaction with the sulfhydryl-specific MTSL to determine if the wild-type cysteine ( C26 ) and the introduced cysteines could be labeled by MTSL . For C26 , K32C , T157C , and P249C , treatment with 1 µM MTSL for 2 min inhibited pH50 currents ( 30%–70% ) , demonstrating that the cysteines were accessible to modification with MTSL ( Figure S1 and Table S1 ) . The MTSL modification shifted the pH50 to more acidic values but did not alter maximal proton-activated currents ( data not shown ) . For the mutants C26A and K247C , MTSL treatment had no effect on subsequent proton-activated currents . For K247C , treatment with the bulkier sulfhydryl-modifying reagent , MTSEA-biotin , inhibited proton-induced currents . To test whether MTSL modified K247C , we applied MTSL prior to MTSEA-biotin . MTSL blocked the ability of MTSEA-biotin to inhibit proton-induced currents , indicating that MTSL labels K247C but has no functional effect on channel activation . We then expressed wild-type and mutant GLIC proteins in E . coli , purified the proteins in n-Dodecyl-β-D-maltoside ( DDM ) , and labeled them with MTSL . To test whether the purified GLIC proteins were functional , we reconstituted mutant C26A into liposomes formed with 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoethanolamine ( PE ) and 1-palmitoyl-2-oleoyl-sn-glycero-3-phospho- ( 1′-rac-glycerol ) ( PG ) , and recorded single-channel currents in planar lipid bilayers also formed with PE∶PG ( Figure 2A ) . PE and PG are major components of the inner cell membrane of most bacteria , and PE∶PG bilayers are good models for the bacterial membrane [24] . The purified C26A mutant reconstituted into PE∶PG liposomes produced proton-elicited single-channel currents with unitary conductance of 13 . 6±0 . 6 pS , which is comparable to the 8 pS value reported for wild-type GLIC in HEK293 cells [25] . We next tested whether the addition of cholesterol or cardiolipin along with PE and PG would affect GLIC single-channel properties . Cholesterol is essential for eukaryotic pLGIC function [26]–[30] , and was recently shown to increase GLIC current desensitization rates [22] . Cardiolipin is an anionic lipid typically found in the bacterial cell membrane , and previous studies have reported that anionic lipids can modulate eukaryotic pLGIC function [29] , [31]–[33] . We reconstituted mutant C26A into liposomes formed with PE∶PG∶cholesterol at a 3 . 4∶1 . 3∶1 molar ratio or PE∶PG∶cardiolipin at a 5 . 8∶2 . 3∶1 molar ratio , and recorded single-channel currents in planar lipid bilayers formed with the same lipids ( Figure 2A ) . The single-channel conductance of C26A mutant GLIC was not altered by cholesterol ( 12 . 8±0 . 5 pS ) or cardiolipin ( 10 . 0±1 . 6 pS ) . The open dwell time in the presence of cholesterol ( 10 . 15±0 . 08 ms , at −100 mV ) was similar to that in PE∶PG ( 11 . 83±0 . 06 ms , at −100 mV ) , whereas in the presence of cardiolipin , it was slightly increased ( 19 . 64±0 . 06 ms , at −100 mV ) . Overall , the data demonstrate that purified GLIC C26A mutant protein reconstituted in PE∶PG liposomes is functional and that cholesterol and cardiolipin have little effect on the GLIC single channel properties measured . We also confirmed that the purified reconstituted cysteine mutant GLIC proteins were functional . Single-channel currents recorded in PE∶PG bilayers for K32C , T157C , and P249C mutant protein ( Figure 2B ) had single channel conductances from 14 . 8±0 . 4 pS ( T157C ) to 5 . 8±0 . 8 pS ( P249C ) , comparable to mutant C26A and wild-type GLIC [25] . In addition , we injected purified , reconstituted , MTSL-labeled protein directly into Xenopus oocytes to verify their functionality [34] , [35] . We recorded significantly larger proton-dependent induced currents ( approximately 300 nA–1 µA ) from oocytes injected with mutant C26A , K32R1 , T157R1 , K247R1 , and P249R1 as compared to noninjected oocytes ( 70 nA ) ( Figure 2C ) . Overall , the data demonstrate that the purified mutant GLIC proteins reconstituted into liposomes were functional and proton-sensitive . We initially recorded the CW spectra of MTSL-labeled wild-type GLIC ( C26R1 ) , at pH 7 . 6 and pH 4 . 6 ( Figure 3 ) . The shape of the CW spectrum reflects the mobility of the R1 side chain , which ultimately depends on packing of its surroundings . Therefore , proton-induced changes in the CW spectrum reflect structural rearrangements that alter the local environment around R1 ( i . e . , neighboring side chain motions and backbone flexibility ) . The CW EPR measurements were collected at room temperature over an hour ( steady-state conditions ) . Based on our proton-concentration response curves , at pH 7 . 6 , the channels will predominantly be in an unliganded , resting conformation . At pH 4 . 6 , the channels will predominantly be in a desensitized conformation . At both pH values , the CW spectra of C26R1 showed the spin labels were largely immobile ( Figure 3 ) , indicating a tightly packed environment near the C26R1 side-chain , consistent with its buried location on β-strand 1 . Switching to pH 4 . 6 had no effect on the shape of the C26R1 CW spectrum , indicating that probe mobility did not change , which suggests the local environment near the probe is the same or it rearranged into an equally packed conformation . The CW spectrum of an MTSL-treated C26A mutant showed virtually no signal ( Figure S2 ) , demonstrating the absence of spin-labeled protein contaminants . To detect proton-induced conformational rearrangements in loop 2 , loop 9 , and the M2–M3 loop , we recorded the CW spectra of the MTSL-labeled GLIC mutants K32R1 , T157R1 , K247R1 , and P249R1 at pH 7 . 6 and pH 4 . 6 ( Figure 3 ) . In the ECD , the CW spectra of K32R1 , located in loop 2 , showed two distinct EPR spectral components , mobile and immobile ( indicated by arrows in Figure 3 ) , likely associated with two alternative rotameric spin-label conformations [36] . At pH 7 . 6 ( closed , resting state ) , a greater proportion of the spin probes were in a mobile conformation , whereas at pH 4 . 6 ( desensitized ) a greater proportion were immobile , indicating a proton-induced structural rearrangement occurred that resulted in a more densely packed environment near the spin probe . Proton-induced changes were also detected in the CW spectra of T157R1 ( Figure 3 ) , located in loop 9 . The low field regions of the spectra were entirely different at the two pH values , indicating a completely new motional environment , with the spin probes predominantly immobile at pH 7 . 6 ( resting ) and mostly mobile at pH 4 . 6 ( desensitized ) . For both K32R1 and T157R1 , switching to pH 4 . 6 did not result in significant spectral broadening or changes in center resonance line amplitude , indicating that the observed differences reflect changes in probe mobility and not intersubunit dipolar spin–spin interactions . Changes in spin probe mobility are plotted in Figure 4 and were calculated by measuring the inverse width of the central line , ΔH0−1: an increase in ΔH0−1 indicates an increase in motion , whereas a decrease in ΔH0−1 indicates a decrease in motion [16] , [17] , [37] . We also collected CW spectra of T157R1 reconstituted in PE∶PG∶cholesterol and in PE∶PG∶cardiolipin at pH 7 . 6 and pH 4 . 6 ( Figure 5 ) to test the effects of lipids on proton-induced motions in GLIC . In the presence of cholesterol , the CW spectra were essentially indistinguishable from the spectra of T157R1 reconstituted in PE∶PG , indicating that cholesterol had no effect on the proton-induced structural rearrangements near loop 9 . In the presence of cardiolipin , there was a marked decrease in the population of spin probes that switched to the more mobile conformation at pH 4 . 6 ( Figure 5 ) , suggesting that cardiolipin hinders proton-induced motions near loop 9 . In the TMD , the CW spectra of K247R1 and P249R1 , located in the M2–M3 loop , also revealed the spin labels were motionally restricted , indicating a sterically packed environment ( Figure 3 ) . For K247R1 , we observed an additional slight decrease in mobility upon switching to pH 4 . 6 and no changes in mobility for P249R1 . The lack of significant proton-induced changes in probe mobility was unexpected , since motions in the M2–M3 loop have been suggested to play an important role in coupling agonist binding to channel gating [7] , [38]–[40] . To ensure that we were maximally activating the MTSL-labeled GLIC protein , we also collected CW spectra at pH 3 . 0 ( Figure S3 ) . Upon changing the pH to 3 . 0 , probe mobility for K247R1 decreased slightly more , whereas no changes were seen for P249R1 , as judged by the inverse central line width ΔH0−1 ( Figure 4 ) . In general , no significant changes in the overall line shape of the CW spectra at pH 3 . 0 compared to pH 4 . 6 or pH 7 . 6 were observed ( Figure S3 ) . Using DEER spectroscopy , distances in the range of 18 to 60 Å between paramagnetic centers in a membrane protein can be measured [41]–[43] . Because GLIC is a homopentamer , two distances are expected at each labeled position: one between spin probes on adjacent subunits , another between probes on nonadjacent subunits ( Figure 6 , Figure S4A ) with a theoretical nonadjacent and adjacent distance ratio of 1 . 6 expected . We measured the distances between probes in GLIC at pH 7 . 6 , which stabilizes the closed state , and at pH 4 . 6 , which favors desensitized states , to test if , and how far , the loops at the ECD/TMD gating interface ( e . g . , K32R1 , T157R1 , K247R1 , and P249R1 ) move during proton-dependent gating transitions . Currently , a high-resolution structure of GLIC is only available in an apparently open channel conformation , and little is known about the process of desensitization at the structural level . While there are uncertainties in assigning functional states to the ELIC and GLIC crystal structures , comparing ELIC ( PDB entry 2VL0 ) and GLIC ( PDB entry 3EHZ ) solved in apparently closed and open channel conformations , respectively [6] , [7] , [10] , suggests that loops 2 and 9 move inward toward the channel lumen ( ∼1 . 7 Å for K32 relative to ELIC's L29 , and ∼5 Å for T157 relative to ELIC's D158 ) , whereas the M2–M3 loop moves ∼6 Å outward away from the channel lumen ( K247 relative to ELIC's R254 and P249 relative to ELIC's P256 ) with channel activation ( Figure 1C , D; Table 1; see Materials and Methods and Figure S4B for displacement calculations ) . By comparing our experimental DEER distances obtained from functional protein in lipids to those predicted from the crystal structures , we can begin to assess the conformational states to which these structures correspond . We initially examined C26R1 , which is located on β-strand 1 in the ECD , for intersubunit distances by DEER spectroscopy . Figure 6 shows the background subtracted dipolar evolution fit using Tikhonov regularization , a model-free approach . The interspin DEER-derived distances were 22 Å and 35 Å at pH 7 . 6 ( adjacent and nonadjacent subunits , respectively ) , and 22 Å and 34 Å at pH 4 . 6 ( Figure 6 , Table 1 ) . Similar distance distributions were obtained when we fit the data using 2-Gaussian or 2 Rice3D model-based approaches . The nonadjacent∶adjacent distance ratios were 1 . 6 , in excellent agreement with the theoretical value for a homopentameric labeled protein . The DEER-derived interspin distances were slightly shorter than the Cβ–Cβ distances ( Table 1 ) measured in the crystal structures of ELIC and GLIC . The absence of detectable pH-induced distance changes indicates either a lack of motion or a concerted rigid-body motion for the ECD β-cores . For K32R1 , in loop 2 , the experimental distances were 20 Å and 28 Å at pH 7 . 6 ( adjacent and nonadjacent subunits , respectively ) , and 19 Å and 27 Å at pH 4 . 6 ( Figure 6 , Table 1 ) . The nonadjacent∶adjacent distance ratios were 1 . 4 , which are smaller than the 1 . 6 theoretical value , suggesting that the probe locations were not perfectly symmetrical . The small ( less than 1 Å ) proton-dependent change in the interprobe center distances suggests that there is little to no proton-induced displacement of loop 2 . For T157R1 , in loop 9 , the interspin DEER-derived distances were 30 Å and 49 Å at pH 7 . 6 ( adjacent and nonadjacent subunits , respectively ) , and 19 Å and 31 Å at pH 4 . 6 ( Figure 6 ) . The nonadjacent∶adjacent distance ratios were 1 . 6 , in excellent agreement with the theoretical value for a homopentameric labeled protein . At pH 4 . 6 , we collected data out to a shorter dipolar evolution time ( solid blue line ) to increase the quality of the data . When we collected data out to the same evolution time as the pH 7 . 6 sample ( dotted blue line ) , intersubunit distances longer than 31 Å were not observed . Upon switching to pH 4 . 6 , the interprobe distance changed more than 17 Å for the nonadjacent distance , indicating that the probe attached to loop 9 undergoes a large proton-induced inward movement toward the channel lumen , with a displacement of 9 . 2 Å ( Figure S4 ) . Since T157R1 is a good reporter of proton-induced conformational motions , we examined the ability of GLIC to undergo these rearrangements in detergent micelles . For T157R1 in DDM micelles , the interspin DEER-derived distances were 31 Å and 49 Å at pH 7 . 6 ( adjacent and nonadjacent subunits , respectively ) , and 30 Å and 49 Å at pH 4 . 6 ( Figure 7 ) . The distances were nearly identical at both pH values and matched the distances obtained from GLIC reconstituted into PE∶PG liposomes at pH 7 . 6—that is , in the resting , closed channel state ( Figure 6 and Table 1 ) . The data suggest that DDM inhibits proton-induced motions in GLIC and locks GLIC in a conformation resembling the resting state . For K247R1 , in the M2–M3 loop , at pH 7 . 6 only one interspin distance of 30 Å was obtained using Tikhonov regularization ( Figure 6 ) . These data were also fit using 2-Gaussians and 2 Rice3D model-based approaches , which also resulted in only one distance ( i . e . , two distances of the same value resulted ) . The interspin distance does not correlate with either of the Cβ–Cβ distances in the ELIC crystal structure ( 13 . 8 Å adjacent , 22 . 3 Å nonadjacent ) but is comparable to the nonadjacent Cβ–Cβ distance in GLIC ( 33 . 7 Å ) . At pH 4 . 6 , we detected two distances of 18 Å and 30 Å ( adjacent and nonadjacent subunits , respectively ) similar to Cβ–Cβ distances measured in the GLIC crystal structure ( Table 1 ) . The apparent absence of any proton-induced change in the nonadjacent 30 Å distance suggests that K247R1 occupies the same location at both pH values , which is consistent with the very modest changes we observed in the CW EPR spectra upon switching to pH 4 . 6 ( Figure 3 ) . We cannot , however , rule out the possibility that the interspin 30 Å distance measured at pH 7 . 6 is between adjacent subunits . Given the short phase memory time for this position ( Tm = 0 . 6 µs ) , a nonadjacent distance of approximately 49 Å ( expected for a 30 Å adjacent distance ) is beyond the range that can be reliably measured [41]–[43] . For P249R1 , also in the M2–M3 loop , the interspin distances were 20 and 30 Å ( adjacent and nonadjacent subunits , respectively ) at pH 7 . 6 , and 20 Å and 28 Å at pH 4 . 6 ( Figure 6 ) , indicating little to no proton-induced changes in distances , consistent with the lack of proton-elicited changes seen in the CW EPR spectra ( Figure 3 ) . Overall , the data suggest that the M2–M3 loop is in a similar position in both the resting and desensitized GLIC conformational states . This result is in contrast to its approximately 6 Å outward displacement away from the channel lumen predicted by comparing the crystal structures of ELIC and GLIC [7] . We also used computer modeling to evaluate how well the ELIC and GLIC crystal structures predict our experimental DEER data . We built a homology model of GLIC based on the ELIC crystal structure and used the PRONOX program to estimate the distances between spin labels in the GLIC model ( Table S2 ) . The computed distances were then compared to our experimental DEER distances . Using standard conditions in the program , no distances were computed for C26R1 , T157R1 , and P249R1 due to clashes ( i . e . , MTSL did not fit using favored rotamer conformations ) , and the average interspin distances computed for K32R1 and K247R1 were shorter than our experimental data . When we relaxed conditions and allowed additional MTSL rotamer conformations , the program still could not compute distances for C26R1 and P249R1 and the distance for T157R1 was shorter than our experimental data . In general , the modeling suggests the ELIC structure obtained in detergent micelles , at least for these positions , is not a good model for the resting state of GLIC embedded in lipids . We also used the PRONOX program to estimate the distances between spin labels using the crystal structure of GLIC ( PDB entry 3EAM ) as the input ( Table S2 ) . Using standard or relaxed conditions , no distances were computed for C26R1 . For T157R1 and K247R1 , the estimated distances were similar to our experimental DEER distances obtained at pH 4 . 6 . The computed distances for K32R1 and P249R1 were much shorter than the experimental DEER distances at pH 4 . 6 , suggesting the GLIC crystal structure , at least at these positions , does not correspond to a desensitized conformation .
The structural rearrangements underlying how pLGICs transition between closed , open , and desensitized states are still unclear . While high-resolution crystal structures of pLGICs in apparently closed and open channel conformations [6] , [7] , [9] , [10] have provided insights into possible activation mechanisms , whether these static protein structures , solved in detergent micelles , accurately capture the conformational gating transitions that a functional pLGIC undergoes when embedded in a lipid bilayer is unknown . Using SDSL EPR spectroscopy and functional GLIC channels reconstituted in liposomes , we measured protein motions associated with GLIC gating under conditions that promote conformational transitions from closed to desensitized states . We focused on the loops forming the interface between the ECD and the TMD , specifically loops 2 and 9 in the ECD and the M2–M3 loop in the TMD , which previous studies have shown to be important for coupling agonist binding to channel gating [39] , [40] , [44] , [45] . Comparing the crystal structures of ELIC and GLIC suggests that these loops undergo structural rearrangements during activation , with loops 2 and 9 moving inward toward the channel lumen , whereas the M2–M3 loop moves outward ( Figure 1C , D ) . Whether and if these loops move during desensitization is unknown . Here , we show that proton-dependent GLIC channel gating transitions into a desensitized state induces substantial rearrangements of the intra- and intersubunit interface between the ECD and TMD . The biggest change occurred in loop 9 , where T157R1 underwent a large ( ∼9 . 2 Å ) proton-induced inward movement toward the channel lumen ( Figure 6 ) . The displacement was accompanied by concurrent rearrangements in its surrounding tertiary contacts , with the CW spectra ( Figure 3 ) revealing a densely packed environment in the resting state ( pH 7 . 6 ) that becomes less packed in the desensitized state ( pH 4 . 6 ) . The DEER-derived interspin distances for T157R1 at pH 7 . 6 are longer than the Cβ–Cβ distances in the ELIC crystal structure and the intersubunit distances estimated computationally , whereas the DEER distances at pH 4 . 6 are shorter than the Cβ–Cβ distances in the GLIC crystal structure and the distances estimated computationally ( Table 1 , Table S2 ) . Thus , the resulting proton-mediated 9 . 2 Å inward displacement of loop 9 measured by DEER spectroscopy is larger than the predicted motion based on comparing the ELIC and GLIC crystal structures . Since the residues in loop 9 in ELIC and GLIC differ substantially [7] , [10] , it is not surprising that the magnitude of the DEER-derived displacement measured in a single protein , GLIC , does not match the displacement calculated by comparing the structures of two different proteins . Also , the loop 9 displacement measured by DEER spectroscopy in a functional protein reconstituted in liposomes may differ from the motion observed when the protein is constrained in a crystal lattice in detergent . In support of this latter possibility , our DEER data demonstrate that the proton-mediated motion of T157R1 is lost in DDM micelles compared to T157R1 reconstituted in PE∶PG liposomes ( Figures 6 and 7 ) . This finding is consistent with the loss of channel function observed for eukaryotic nicotinic acetylcholine receptors solubilized in DDM [46] and supports the idea that detergent-solubilization of membrane proteins can affect structural dynamics and result in conformational ambiguity of the crystal structures solved in the presence of detergents ( reviewed in [13] ) . Our DEER spectroscopy experiments measured the intersubunit distances of T157R1 in GLIC in a resting , closed state ( pH 7 . 6 ) and in a desensitized , closed state ( pH 4 . 6 ) . Currently , our experiments cannot distinguish whether loop 9 moves in the open state and remains displaced during desensitization or whether its movement occurs specifically in the desensitized state . Nonetheless , the data directly demonstrate that a large inward motion of loop 9 occurs during GLIC gating transitions , which results in substantial rearrangements of the intersubunit interface . We predict that a similar inward motion of loop 9 in eukaryotic pLGICs occurs during agonist-mediated channel gating transitions . By measuring changes in cysteine accessibility , disulfide bond formation , and attached fluorophore emissions , agonist-induced local rearrangements near loop 9 have been detected in nicotinic acetylcholine receptors [47] , [48] , GABAA receptors [49]–[52] , glycine receptors [53] , and serotonin-type 3 receptors [54] . Proton-mediated structural rearrangements in the local protein environment near K32R1 in loop 2 were also observed . CW EPR spectroscopy ( Figure 3 ) revealed K32R1 is in a more densely packed environment in the desensitized state ( pH 4 . 6 ) compared to the resting state ( pH 7 . 6 ) , suggesting that during channel activation to desensitization loop 2 becomes less mobile . Since our DEER measurements at pH 7 . 6 and 4 . 6 demonstrate loop 2 undergoes minimal proton-induced displacement ( Figure 6 ) , the decrease in K32R1 mobility likely arises primarily from an increase in its surrounding tertiary interactions . Residues in loop 7 , loop 9 ( adjacent subunit ) , M2–M3 loop , and the pre-M1 are in close proximity to loop 2 , and functional studies in eukaryotic pLGICs have shown that a network of electrostatic and hydrophobic interactions between these regions and loop 2 play a role in coupling binding to gating [45] , [55] . Our data suggest that increases in these interactions and a resulting immobilization of loop 2 accompany GLIC channel gating transitions into the desensitized state . Proton-induced conformational rearrangements near the M2–M3 loop in the TMD were minimal , at least for the two positions we examined , K247R1 and P249R1 . In the resting closed channel state ( pH 7 . 6 ) , the EPR spectra ( Figure 3 ) showed the spin-probes at both positions were motionally restricted , reflecting a sterically packed environment near the probes . Upon switching to pH 4 . 6 or 3 . 0 ( desensitized state ) , only a modest decrease in K247R1 mobility was observed , whereas no change in P249R1 mobility was seen ( Figures 3 and 4 ) . Moreover , our DEER data indicate that there were essentially no proton-mediated changes in intersubunit distances between probes at these positions ( Figure 6 ) . Overall , our data suggest the M2–M3 loop does not undergo significant movement . Based on the crystal structures of ELIC and GLIC , the M2–M3 loop is predicted to move ∼6 Å ( Cβ–Cβ ) outward away from the channel lumen during activation ( Figure 1C and Table 1 ) . One possible explanation for the difference between our data and the structure-based predictions is that in the desensitized state , which we are preferentially monitoring at pH 4 . 6 , the M2–M3 loop adopts a conformation that resembles its conformation in the resting state . Consistent with this idea , photolabeling of a residue in the M2–M3 loop of the nicotinic acetylcholine receptor delta subunit is state-dependent with robust labeling seen only in open and fast desensitized states and little to no labeling in the resting and slow desensitized states [56] . In a cysteine accessibility study , modification of cysteines introduced into the GLIC M2 helix and the M2–M3 loop were faster at pH 5 . 0 than pH 7 . 5 , indicating a proton-mediated increase in accessibility [12] . Again , a possible explanation for the difference between their data and ours is that we are monitoring desensitized states at pH 4 . 6 , whereas in their study the channels are submaximally activated at pH 5 . 0 and in a mixture of open and possibly resting and desensitized conformations . Overall , our SDSL EPR data suggest that the M2–M3 loop is in a relatively packed environment in the resting state that is relatively unchanged in the desensitized state . A recent work [23] suggests residues in the middle of M2 move inward to occlude the channel during desensitization , whereas residues in the extracellular end of M2 remain displaced outward . One might expect then that the M2–M3 loop , which is attached to M2 , would remain displaced outward . Our data suggest that this is not the case . Whether the GLIC M2–M3 loop moves substantially during proton-induced channel opening and then moves back to a position similar to that adopted in the resting state during desensitization is unclear . The motions inferred from static crystal structures of two different proteins in uncertain functional conformations might not accurately reflect gating motions of a functional protein embedded in a lipid membrane . The intersubunit distances we measured at pH 7 . 6 in the resting state for both K247R1 and P249R1 are substantially different from the Cβ–Cβ distances in the crystal structure of ELIC for the structurally aligned residues R254 and P256 and from the intersubunit distances that we estimated computationally , suggesting that the ELIC structure is not a good model for the conformation of GLIC's M2–M3 loop in the resting state . In the recently crystallized locally closed GLIC structures , the three conformations of the M2–M3 loop are different than the M2–M3 loop conformation in ELIC [57] . In addition , comparing one of the locally closed channel GLIC structures ( LC2 conformation , PDB entry 3TLS ) with GLIC in an apparently open channel state ( PDB entry 3EAM ) suggests channel closing/opening can occur without significant rearrangements of the M2–M3 loop . MTSL labeling and/or detergent solubilization and membrane-reconstitution could lock the receptor in a nonactivatable “uncoupled” state , which abolishes its ability to undergo conformational transitions in response to pH changes . Previous work on nicotinic acetylcholine receptors have shown that lipid composition and choice of detergent are critically important for maintaining optimal receptor functionality [32] , [33] , [46] , [58] . Since we record robust proton-elicited currents in bilayers with our purified PE∶PG lipid reconstituted GLIC protein , and when we directly inject purified MTSL-labeled lipid reconstituted protein into oocytes ( Figure 2 ) , these possibilities seem unlikely . Moreover , in a recent study , the ability of GLIC to undergo proton-dependent gating transitions was maintained following its reconstitution in a variety of different lipids [59] , consistent with our findings . Both cholesterol and anionic lipids are well-known modulators of pLGIC function [60] , and it was recently shown that cholesterol modulates GLIC gating kinetics , speeding its desensitization [22] . Our data using CW EPR spectroscopy showed that cholesterol , when added to PE∶PG liposomes , had no effect on the proton-induced structural rearrangements near loop 9 ( Figure 5 ) . In addition , cholesterol had little effect on GLIC single channel conductance or open-dwell time . On the other hand , adding cardiolipin to PE∶PG liposomes reduced probe mobility at pH 4 . 6 compared to PE∶PG alone or PE∶PG∶cholesterol , suggesting that cardiolipin inhibits proton-induced motions near loop 9 ( Figure 5 ) . In a recent report , GLIC protein appeared slightly more rigid when reconstituted into membranes formed by E . coli lipids ( i . e . , a PE∶PG∶cardiolipin mixture ) as compared to reconstitution in asolectin or phosphocholine [59] , consistent with our finding that cardiolipin decreases GLIC mobility . In summary , using SDSL EPR spectroscopy , we present new information about the structural changes associated with ligand-induced gating motions in the prokaryotic pLGIC GLIC using a functional protein reconstituted into a native-like lipid environment . We provide direct experimental evidence that structural rearrangements of the intra- and intersubunit interface between the ECD and TMD accompany pLGIC gating transitions from closed to desensitized states . Specifically , in the ECD , proton-induced gating transitions from closed to desensitized states decrease local side-chain interactions with loop 9 , which increases loop 9 mobility and results in a large inward movement of loop 9 , whereas loop 2 becomes more immobilized . These data suggest that desensitization not only involves structural changes in the M2 channel helix to block ion conduction [23] but also entails motions in the ECD that likely change the network of interactions between residues in loop 2 , loop 7 , loop 9 , preM1 , and the M2–M3 linker . In the resting state , the M2–M3 loop in the TMD domain is relatively immobile and in a packed environment , and remains in nearly the same position in the desensitized state . The position of P249R1 in the M2–M3 loop in the desensitized state is substantially different than that observed in the GLIC apparently open channel structure , suggesting the crystal structure is not in a desensitized conformation . Currently , a resting , closed channel state structure of GLIC is not available . Our DEER data provide a first glimpse of the positions of GLIC residues in the resting state and suggest that the ELIC structure is not a good model for the resting state . These findings advance our understanding of the molecular mechanisms underlying pLGIC gating .
The DNA sequence encoding GLIC ( residues 44–359 ) was extracted by PCR amplification from G . violaceus cells ( ATCC ) , and subcloned in vectors pUNIV [61] for two-electrode voltage clamp experiments , and pET-26b ( Novagen ) for expression in E . coli . GLIC DNA sequence was preceded in pUNIV by the DNA sequence encoding the signal peptide of the GABAA receptor β2 subunit to promote cell surface expression . pET-26b incorporates an N-terminal pelB signal sequence for potential periplasmic localization . In addition , DNA sequence for maltose-binding protein ( MBP ) followed by a ∼20 amino acid linker containing a consensus sequence for thrombin cleavage was cloned following the pelB signal and N-terminal to GLIC . GLIC mutants were created using the QuikChange site-directed mutagenesis kit ( Stratagene ) . Mutations were confirmed by DNA sequencing . Capped cRNAs encoding WT and mutant GLIC were transcribed in vitro using the mMessage mMachine T7 kit ( Ambion ) . Single X . laevis oocytes were injected with 27 nL of cRNA ( 50–100 ng/µL/subunit ) . Injected oocytes were incubated at 16°C in ND96 ( 5 mM HEPES pH 7 . 4 , 96 mM NaCl , 2 mM KCl , 1 mM MgCl2 , 1 . 8 mM CaCl2 ) supplemented with 100 µg/ml of gentamycin and 100 µg/mL of bovine serum albumin for 2–5 d before use for electrophysiological recordings . Oocytes were perfused continuously with ND96 at pH 7 . 4 at a flow rate of 5 mL/min , while being held under two-electrode voltage clamp at −60 mV in a bath volume of 200 µL . Borosilicate glass electrodes ( Warner Instruments ) used for recordings were filled with 3 M KCl and had resistances of 0 . 4 to 1 . 0 MΩ . Electrophysiological data were collected using Oocyte Clamp OC-725C ( Warner Instruments ) interfaced to a computer with an ITC-16 A/D device ( Instrutech ) and were recorded using the Whole Cell Program , version 4 . 0 . 9 ( kindly provided by J . Dempster , University of Strathclyde , Glasgow , UK ) . Proton-induced currents were measured by perfusing ND96 buffered at pH 6 . 5–3 . 8 . For pH 5 . 0–3 . 8 HEPES was replaced with 5 mM Na Citrate as the buffering agent . For pH 6 . 5–6 . 0 5 mM MES was used as the buffering agent . GraphPad Prism 4 was used for data analysis and fitting . pH response data were fit to the equation I = Imax/ ( 1+10 ( pH-pH50 ) *n ) , where I is the peak response at a given pH , Imax is the maximum amplitude of current , pH50 is the pH inducing half maximal response , and n is the Hill coefficient . The functional effect of modifying substituted cysteines with 1-oxyl-2 , 2 , 5 , 5-tetramethyl-3-pyrroline-3-methyl methanethiosulfonate spin label ( MTSL ) was evaluated in oocytes using two-electrode voltage clamp . Proton-induced currents were measured at pH 5 . 0 until peak current amplitudes varied by <5% . Oocytes were then treated with 1 µM MTSL at pH 7 . 4 for 2 min , washed for 5 min , and proton-induced currents were measured again at pH 5 . 0 . Extent of modification was quantified as ( 1−Iafter MTSL/Ibefore MTSL ) *100% . WT and C26A were incubated with 100 µM MTSL . E . coli BL21 ( DE3 ) strain cells ( Invitrogen ) were transformed with the pET-26b vector encoding the GLIC constructs . Cells were cultured in LB medium at 37°C to OD600∼1 . 0–1 . 4 , and then expression was induced overnight at 20°C with 0 . 2 mM isopropyl β-D-1-thiogalactopyranoside ( IPTG ) . Cells were harvested and lysed with an EmulsiFlex C-5 homogenizer ( Avestin ) in 20 mM Tris-HCl pH 7 . 6 , 150 mM NaCl ( buffer B1 ) supplemented with 1 mM PMSF , 2 µM pepstatin-A , and 2 µg/mL leupeptin as protease inhibitors . The lysate was cleared by centrifugation at 18 , 000 rpm for 30 min at 4°C , and then the pellet was resuspended in B1 with 2% n-dodecyl-β-D-maltoside ( DDM , Anatrace ) and gently agitated overnight at 4°C for protein extraction from cell membranes . Solubilized pellet was cleared by ultracentrifugation at 45 , 000 rpm with a 50 . 2 Ti rotor ( Beckman ) for 45 min at 4°C and purified by affinity chromatography with amylose resin ( New England Biolabs ) . Amylose resin with bound MBP-GLIC was washed with 10 volumes of B1 with 0 . 1% DDM followed by 10 volumes of B1 with 0 . 02% DDM ( buffer B2 ) , and then the fusion protein was eluted in B2 supplemented with 20 mM maltose . MBP-GLIC was concentrated in Amicon Ultra-4 ( 100 KDa molecular weight cutoff ) concentrator tubes ( Millipore ) and subjected to size exclusion gel filtration in a Superose6 GL10/300 column ( GE Healthcare ) previously equilibrated in B2 . Fractions of the peak corresponding to pentameric MBP-GLIC ( ∼400 kDa ) were combined and treated with MTSL and thrombin under gentle agitation at 4°C overnight . In detail , protein was first treated with 5-fold molar excess of DTT for 5 min at room temperature , and then 2- to 60-fold molar excess of MTSL ( Toronto Research ) was added to specifically label the unique cysteines , followed by 1 U of thrombin ( bovine , plasminogen-free , Calbiochem ) per 100 µg of pentameric MBP-GLIC . The digested product was applied to amylose resin for a second round of affinity chromatography to purify the cleaved , MTSL-labeled GLIC from the excess spin label and MBP . GLIC was subjected to a final gel filtration , and peak fractions corresponding to the pentameric form of the protein ( ∼180 kDa ) were combined and concentrated to 3–6 mg/mL , flash frozen in liquid nitrogen , and stored at −80°C . Purified GLIC protein was reconstituted into liposomes formed with 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoethanolamine ( PE ) and 1-palmitoyl-2-oleoyl-sn-glycero-3-phospho- ( 1′-rac-glycerol ) ( PG ) at a PE∶PG = 2 . 7∶1 molar ratio , or with PE∶PG∶cholesterol at a 3 . 4∶1 . 3∶1 molar ratio . Lipid mixtures were prepared at a concentration of 20 mg/mL in buffer B1 , sonicated , and mixed with GLIC purified in DDM , typically at 6 , 000-fold molar excess . After a 3 h incubation at 4°C , the protein∶lipid mixture was diluted 2-fold in buffer B1 containing 10% glycerol and incubated overnight at 4°C . To remove DDM , Biobeads ( BioRad ) were added for 8–10 h and then removed . Finally , the Biobead-free solution was ultracentrifuged at 100 , 000 rpm , and the pellets of GLIC reconstituted into liposomes were stored at −80°C . Pellets of purified GLIC mutants reconstituted into liposomes were thawed on ice . The amount of protein in a pellet was estimated by assuming 70% reconstitution efficiency . The pellets were resuspended in buffer B1 to a protein concentration of approximately 1–2 mg/mL and were subjected to two rounds of freeze-thaw . The proteoliposome solution was somewhat viscous and to facilitate protein injection into the oocytes , the diameter of the glass injection pipet was adjusted to about 5–10 µm . Protein-injected oocytes were incubated for 5–8 h at 16°C before recording . Two-electrode voltage clamp of oocytes injected with lipid reconstuted GLIC protein was performed in the same manner as oocytes injected with GLIC cRNA . pH-induced currents from uninjected oocytes were used as controls . For preparation of planar lipid bilayers , lipid mixtures of PE∶PG ( 2 . 7∶1 ) , PE∶PG∶cholesterol ( 3 . 4∶1 . 3∶1 ) , or PE∶PG∶cardiolipin ( 5 . 8∶2 . 3∶1 ) were prepared in n-decane at a concentration of 20 mg/mL . Planar lipid bilayers were painted with a glass rod across a 150 µm aperture in a Delrin cup . In order to create both an ionic and a pH gradient , the trans chamber was filled with 150 mM NaCl at pH 7 . 6 , whereas the cis chamber ( where the protein was added to ) was filled with 450 mM NaCl at pH 5 . 2 . Prior to adding the GLIC K32R1 , T157R1 , and P249R1 protein to the chamber , the protein was treated with 10 mM dithiothreitol ( DTT ) to remove the majority of the spin label . Once the protein was incorporated into the planar lipid bilayer , the pH of the cis chamber was dropped to 4 . 6 by adding 10% ( v∶v ) of 1 M Na Citrate . Single-channel currents were recorded using an Axopatch 200B amplifier ( Axon Instruments ) , filtered with an 8-pole low-pass Bessel filter ( Frequency Devices ) set at 100 Hz , and digitized at a rate of 4 kHz with a Digidata 1440A interface ( Axon Instruments ) . Data acquisition and analysis were performed with pClamp10 . 2 . Continuous wave ( CW ) EPR spectroscopy was carried out at room temperature on a Bruker ELEXSYS 500 X-band spectrometer equipped with a superhigh Q ( SHQ ) cavity ( Bruker Biospin ) . Upon change in pH to 4 . 6 or 3 . 0 , the proteolipid samples were freeze-thawed to ensure even distribution of the protons inside and outside the vesicles . Spectra were then recorded over 100 G under nonsaturating conditions with a 100 kHz field modulation of 1 G . Samples were typically 20 µL in volume and contained in a glass capillary . Protein concentrations were typically 30 µM . The DEER spectroscopy experiments were carried out at the Ohio Advanced EPR Laboratory at Miami University using a Bruker ELEXSYS 580 Q-band spectrometer equipped with a Bruker EN5107D2 dielectric resonator or at the National Biomedical EPR Center using a Bruker ELEXSYS 580 X-band spectrometer equipped with a Bruker 3 mm split-ring cavity . Samples were typically 10 µL for Q-band and 25 µL for X-band at a concentration of 35–50 µM , contained 20% deuterated glycerol as a cryoprotectant , were flash frozen using a dry ice-acetone slurry , and run at 80 K . A four-pulse DEER sequence [42] was used with two-step phase cycling . The dipolar evolution data were analyzed for distance distributions using DeerAnalysis2011 software [62] and model-free Tikhonov regularization as it gave the best fit to the background-corrected data . Distribution curves obtained from model-free Tikhonov regularization were then fit to Gaussian shapes using Peak Fitter ( T . O'Haver , MATLAB File Exchange ) to obtain the mean peak center distance values . Rice3D and Gaussian analyses of the dipolar evolution data yielded similar results as the model-free Tikhonov regularization analysis . All DEER data distributions shown are the result of model-free Tikhonov regularization . Pairs of data were recorded on the same spectrometers and under identical conditions . The minimum displacement δ for a spin probe is calculated using the formula: ( 1a ) or ( 1b ) where AC and AO are the DEER-determined distances between probes in adjacent subunits ( indicated as “short” in Table 1 ) at pH 7 . 6 and 4 . 6 , respectively , and NC and NO are the distances between probes in nonadjacent subunits ( indicated as “long” in Table 1 ) at pH 7 . 6 and 4 . 6 , respectively ( see Figure S4B ) . The formulas are also used to calculate the displacement of a residue in GLIC relative to the structurally aligned amino acid in ELIC using the crystal structures . In this case , AC and AO are the Cβ–Cβ distances separating pairs of equivalent residues in adjacent subunits in ELIC and GLIC , respectively , whereas NC and NO are the distances separating pairs of equivalent residues in nonadjacent subunits in ELIC and GLIC , respectively . In this method the displacement δ depends on distances calculated within each crystal structure ( i . e . , the intrinsic coordinates ) , which is a more reliable method than positioning the two structures on top of each other and measuring distances between aligned residues . A closed state homology model of GLIC based on the crystal structure of ELIC ( PDB entry 2VL0 ) was built using Modeller as described by Ghosh and co-workers [63] . The PRONOX program ( http://rockscluster . hsc . usc . edu/research/software/pronox/pronox . html ) was used as described by Hatmal and colleagues [64] to estimate the distances between spin labels using our GLIC homology model and the crystal structure of GLIC ( PDB entry 3EAM ) as inputs . In general , the PRONOX distances were estimated using the standard approach . For some positions , we used the fine search option to help remove clashes . All distances are from N to N atoms . The computed PRONOX distances calculated for the GLIC homology model and the GLIC crystal structure were compared to our experimental DEER distances measured at pH 7 . 6 and pH 4 . 6 , respectively . Note , the DEER distances are obtained from lipid-embedded functional GLIC protein , while the PRONOX distances are based on static X-ray crystal structures of two different proteins in detergent micelles . | Ligand-gated ion channels reside in the membranes of nerve and muscle cells . These proteins form channels that span the membrane , where they transduce chemical signals into changes in electrical excitability . Neurotransmitters bind to the extracellular surface of these proteins to trigger global structural rearrangements that open the channel , allowing ions to flow across the cell membrane . In the continued presence of neurotransmitters , the channels desensitize and close . Channel opening and closing regulate muscle contraction and signaling in the brain , and defects in these channels lead to a variety of diseases . While crystal structures have provided frozen snapshots of these proteins in presumed closed and open channel states , little is known about how the channels desensitize and move during actual signaling events . Here , we applied a technique to investigate the structure and local dynamics of proteins known as site-directed spin labeling to a prototypical ligand-gated channel , GLIC . We directly quantified ligand-induced motions in regions at the boundary between the binding domain ( loops 2 and 9 ) and the channel domain ( M2–M3 loop ) . We show that a large movement of loop 9 and an immobilization of loop 2 , which rearranges the interface between the binding and channel domains , accompanies GLIC channel gating transitions into a desensitized state . These data provide new insights into the protein movements that underlie electrochemical transmission of signals between cells . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2013 | Site-Directed Spin Labeling Reveals Pentameric Ligand-Gated Ion Channel Gating Motions |
To cause infections microbes need to evade host defense systems , one of these being the evolutionarily old and important arm of innate immunity , the alternative pathway of complement . It can attack all kinds of targets and is tightly controlled in plasma and on host cells by plasma complement regulator factor H ( FH ) . FH binds simultaneously to host cell surface structures such as heparin or glycosaminoglycans via domain 20 and to the main complement opsonin C3b via domain 19 . Many pathogenic microbes protect themselves from complement by recruiting host FH . We analyzed how and why different microbes bind FH via domains 19–20 ( FH19-20 ) . We used a selection of FH19-20 point mutants to reveal the binding sites of several microbial proteins and whole microbes ( Haemophilus influenzae , Bordetella pertussis , Pseudomonas aeruginosa , Streptococcus pneumonia , Candida albicans , Borrelia burgdorferi , and Borrelia hermsii ) . We show that all studied microbes use the same binding region located on one side of domain 20 . Binding of FH to the microbial proteins was inhibited with heparin showing that the common microbial binding site overlaps with the heparin site needed for efficient binding of FH to host cells . Surprisingly , the microbial proteins enhanced binding of FH19-20 to C3b and down-regulation of complement activation . We show that this is caused by formation of a tripartite complex between the microbial protein , FH , and C3b . In this study we reveal that seven microbes representing different phyla utilize a common binding site on the domain 20 of FH for complement evasion . Binding via this site not only mimics the glycosaminoglycans of the host cells , but also enhances function of FH on the microbial surfaces via the novel mechanism of tripartite complex formation . This is a unique example of convergent evolution resulting in enhanced immune evasion of important pathogens via utilization of a “superevasion site . ”
Complement system ( C ) is an important part of innate immunity in human plasma , and the alternative pathway of complement ( AP ) is the first line of defense against invading microbes . AP is spontaneously activated on all unprotected surfaces leading to covalent binding of the main complement opsonin C3b to hydroxyl or amine groups . Surface-attached C3b forms a base for enzymatic convertases , which cleave intact C3-molecules until the activator surface is covered with C3b-molecules . This opsonization leads to opsonophagocytosis , propagation of the cascade resulting in release of chemotactic and anaphylatoxic peptides , and formation of lytic membrane attack complexes . To prevent attack against host structures and over consumption of the components in plasma , complement needs to be tightly regulated . The main regulator of the AP in plasma is factor H ( FH ) . FH is a 150 kDa glycoprotein and consists of twenty globular complement control protein modules ( CCPs ) , each approximately 60 residues long . The AP control activity of FH is in domains 1–4 ( FH1-4 ) [1] , [2] . The so-called cofactor activity of FH is needed for inactivation of the central complement opsonin C3b by the serine-protease factor I . In addition to this , FH regulates AP activation by competing with factor B in binding to C3b and accelerating the decay of AP convertase C3bBb [3] , [4] . To regulate complement , FH has to discriminate between host and non-host surfaces , as activation is warranted on microbial surfaces , but obviously not on host surfaces . This “target recognition” site is known to be in the carboxyl-terminal domains 19–20 ( FH19-20 ) [5] , [6] . Our structures of domains 19–20 alone [7] and complexed with C3d [8] showed how SCR20 can bind to cellular and glycosaminoglycan containing surfaces while SCR19 binds simultaneously to C3d part of C3b facilitating control of the AP . This dual binding ability facilitates target recognition by the AP . The necessity of FH and its ability to distinguish between host and non-host surfaces is demonstrated by mutations in the carboxyl-terminus of FH . Even heterozygous mutations in this region can lead to uncontrolled AP activation on host cells causing severe damage to endothelial cells , red cells , and platelets , resulting in a serious systemic disease , atypical hemolytic uremic syndrome [9] . Another important target binding region in FH is within domain 7 and polymorphism in this domain is strongly associated with age-related macular degeneration , the most common cause of blindness in elderly people in industrialized countries [10] , [11] . FH is utilized by several pathogenic microbes for protection against complement attack [12] . Binding of FH down regulates opsonization and prevents further amplification of the C cascade followed by formation of cytolytic membrane attack complexes . While prevention of opsonization and subsequent phagocytosis is beneficial for practically all microbes , evasion of membrane attack complex formation is especially important for Gram-negative bacteria and spirochetes . Acquisition of FH is important or even essential for pathogens; increasing numbers of them have been shown to bind FH [12] . There are two main interaction sites on FH for microbial binding ( Table S1 ) ; one is within domains 6–7 , and group A streptococci [13] and Neisseria [14] , for example , utilize this site . Binding via domains 6–7 facilitates also utilization of FHL-1 , an alternatively spliced transcript derived from FH-gene which contains domains 1–7 of FH and has cofactor-activity like FH [15] . Many microbes have been shown to bind both FH and FHL-1 [16] . The other microbial interaction site on FH is in the carboxyl-terminal domains 19–20 . It seems that most microbes utilize both sites: for instance , B . burgdorferi sensu stricto , which causes Lyme disease , binds FH via domain 7 using protein CRASP-1 [17] and via domains 19–20 using outer surface protein E ( OspE ) and its paralogs [18] . This ability for dual binding facilitates efficient protection against the AP attack . Due to the high homology between the C-terminus of FH and C-termini of FH-related proteins ( FHRs ) , some microbes bind also certain FHRs but the significance of this phenomenon for immune evasion is not clear yet . We wanted to analyze in detail how and especially why different microbes utilize FH via the carboxyl-terminus . We selected pathogens representing Gram-negative , Gram-positive , and eukaryote microbes known to bind FH , and three microbial proteins , OspE ( from B . burgdorferi sensu stricto ) [18] , FhbA ( from B . hermsii ) [19] , and Tuf ( from P . aeruginosa ) [20] . We discovered that they all share a common binding site in domain 20 that overlaps but is not identical with the heparin and cellular binding sites . We also showed that FH bound to the microbial binding site forms a tripartite complex with C3b and furthermore , formation of this complex not only facilitates regulation of the AP but also enhances it .
We first characterized at the molecular level how microbes bind FH via domains 19–20 . We generated point mutations to 14 surface exposed residues of a recombinant fragment of FH domains 19–20 and used five different microbes isolated from patients: three Gram-negative bacteria P . aeruginosa ( Pa ) [20] , ( H . influenzae ( Hi ) [21] , B . pertussis ( Bp ) [22] ) , one Gram-positive bacterium ( S . pneumoniae ( Sp ) [23] ) , and one eukaryotic pathogen ( C . albicans ( Ca ) [24] ) . We also measured binding of full FH to strains used and noticed they all bind FH , as expected on the basis of previous reports ( Figure S1 ) . Binding of 125I-labeled wild type ( wt ) FH19-20 was measured in the presence of increasing amounts ( up to 7 µM ) of the mutant FH19-20 constructs . Concentrations of the mutants that inhibited 50% of the wt FH19-20 binding ( IC50 ) were calculated from binding curves of three experiments done in triplicate ( examples are shown in Figure S2 ) and shown in Figure 1 as a reciprocal value ( 1/IC50 ) for clarity ( diminished value indicating diminished binding ) . Three mutations , R1182A , R1203A , and R1206A , caused decreased binding to all five microbes ( p<0 . 05 ) ; K1188A had reduced binding to four microbes ( Hi , Pa , Sp , Ca ) ; R1210A to three ( Hi , Pa , Sp ) ; and the K1186A and R1215Q mutations reduced binding to one microbe ( Hi ) ( Figure 1 ) . Four other mutations ( W1183L , T1184R , L1189R , E1198A ) in domain 20 and three ( D1119G , Q1139A , W1157L ) in domain 19 showed no reduction in binding compared to wt . To further characterize interaction of FH with microbial surfaces , similar binding inhibition assays were used with three non-homologous and structurally unrelated bacterial outer surface proteins: OspE , a 15 kDa protein from a Lyme borreliosis agent B . burgdorferi [18] , FhbA , a 20 kDa protein from a relapsing fever spirochete B . hermsii [25] , and Tuf , a 43 kDa protein from P . aeruginosa [20] . Binding of 125I-FH19-20 to the recombinant proteins was measured in the presence of increasing concentrations of the 14 mutant proteins and the IC50 values were calculated from the binding curves as above . When compared to wt FH19-20 , two mutant proteins , R1182A and R1206A , showed decreased affinity to all the three microbial proteins , five mutants ( W1183L , L1189R , E1198A , R1203A , R1215Q ) to two microbial proteins and one mutant ( R1210A ) to one protein ( p<0 . 05 ) ( Figure 2 , Panels A–C , shown as a reciprocal value ( 1/IC50 ) for clarity ) . The effect of three mutants ( T1184R , K1186A , K1188A ) in domain 20 and three ( D1119G , Q1139A , W1157L ) in domain 19 was comparable to wt FH19-20 ( p>0 . 05 ) . Six of the mutants showed decreased binding to both OspE and FhbA suggesting an overlap of the binding sites . The overlap was confirmed using cross inhibition assays with OspE and FhbA ( Figure 2 , Panels D and E ) . Taken together , the binding inhibition assays revealed that all mutants that affected binding were in the domain 20 . Furthermore , we identified one mutant ( R1182A ) with significantly decreased binding to all the microbes or microbial proteins analyzed and two mutants ( R1203A , R1206A ) with significantly reduced binding to seven out of eight targets ( p<0 . 05 ) ( Table 1 ) . In addition , the three central residues in microbial binding , R1182A , R1203A , and R1206A , are close to each other in the crystal structure of FH19-20 [7] . They are within 14 Å of each other on domain 20 and three residues ( K1188A , R1210A , R1215Q ) involved in binding to several microbes are also nearby ( Figure 3 ) . Folding of all these mutants was comparable to wt FH19-20 according to a circular dichroism analyses ( Figure S3 ) . One binding site for glycosaminoglycans/heparin is located at FH domain 20 [26] . We next analyzed if microbes could utilize this site by analyzing binding of 125I-FH19-20 to OspE , FhbA , and Tuf in the presence of heparin , a model substance for cell surface glycosaminoglycans . We showed that heparin inhibits binding of FH19-20 efficiently to Tuf and slightly also to OspE and FhbA ( Figure 4 ) . The data are consistent with previous data showing that glycosaminoglycans bind to residues R1203 , R1206 , R1210 , and R1215 at the very carboxyl-terminus of FH20 [27] . This suggests that the microbial binding site on FH overlaps to some extent with , but is not identical to , the heparin binding site needed for recruitment of FH to eliminate C3b on host cells . Down-regulation of the AP by FH on host cells occurs because FH20 binds to glycosaminoglycans/heparin while FH19 binds simultaneously to the C3d part of C3b [8] , [28] . Next we tested if microbes could utilize FH similarly , i . e . facilitating a two point binding of FH19-20 to surface-bound C3b , one site binding to the microbial protein and the other to C3b . There are two binding sites on FH19-20 for the central complement opsonin C3b , one in domain 19 and the other in domain 20 [8] . Structural analysis shows that the site on domain 20 overlaps with the microbial site , while the site on domain 19 of FH is clearly distinct from it [8] . In agreement with our model , binding of C3d did not inhibit binding of FH19-20 to the microbial proteins ( Figure 4 , panels A–C ) but , to our surprise , actually enhanced it . As C3d enhanced binding of FH19-20 to microbial proteins , we analyzed further if microbial proteins could enhance binding of FH19 to its main physiological ligand , C3b . We measured the binding of 125I-FH19-20 to C3b in the presence of OspE , FhbA , and Tuf . OspE and FhbA enhanced binding of FH19-20 to C3b statistically significantly while enhancement with Tuf was smaller and not significant ( Figure 5 , Panel A ) . This suggests that a microbial protein , FH19-20 , and C3b together form a tripartite complex . We were able to prove this by measuring binding of 125I-OspE to solid phase C3b in the presence of FH19-20 ( Figure 5 , Panel B ) . This means that the tripartite complex must form , because OspE alone does not bind C3b [18] . Mutation of four central residues in the C3d/C3b binding site on domain 19 of FH ( FH19del-20 ) [8] significantly reduced the formation of the tripartite complex , indicating that the C3d/C3b binding site on domain 19 is essential for the interaction ( Figure 5 , Panel B ) . These experiments show that FH19-20 can bind simultaneously to a microbial protein and C3b , and that binding of microbial proteins to FH19-20 enhances the FH-C3b interaction . To further test formation of the tripartite complexes on microbial surfaces we measured effect of C3d ( 100 µg/ml ) on binding of FH19-20 to the surface of whole microbes ( B . burgdorferi , S . pneumoniae , P . aeruginosa , H . influenzae and C . albicans ) . A small increase in FH19-20 binding was observed with all the used microbes , most clearly with S . pneumoniae and C . albicans ( Figure S4 ) . No binding of 125I-C3d to any microbes was seen ( data not shown ) . By modeling the tripartite complex on a surface using the structure of FH19-20 in complex with C3d [8] , C3b ( containing the C3d part ) [29] , and our recent crystal structure of FH19-20 in complex with borrelial OspE protein ( Bhattacharjee et al . , submitted ) , a model of a microbial surface protein , we could also show that formation of a tripartite complex is possible without any steric clashes . Furthermore , in this model the thioester site of C3b faces towards the membrane indicating that a surface-bound microbial protein can enhance binding of FH to C3b on the same surface ( Figure 5 , Panel C ) . The results above suggested that , by enhancing the interaction between FH and C3b , microbes might be able to down-regulate complement activation more efficiently . The main regulatory function of FH is to act as a cofactor for serine protease factor I in inactivation of C3b . We therefore measured the cofactor activity of full length FH in factor I mediated cleavage of C3b in the presence of the three microbial proteins , OspE , FhbA , or Tuf ( Figure 6 , Panels A and B ) . All tested microbial proteins enhanced significantly the cofactor activity of FH ( p<0 . 05 at ≥20 µg/ml for all of the proteins ) . The enhancement was due to the carboxyl-terminal part of FH , since it did not clearly occur when FH1-4 was used instead of full length FH ( Figure 6 , Panel C ) , i . e . enhancement obviously requires domains 19 and 20 that mediate formation of the tripartite complex .
Escape of the complement system , and especially its alternative pathway amplification cascade , is a prerequisite for microbial virulence since this first line immune mechanism is spontaneously activated on all non-protected surfaces . Microbes are known to protect themselves by binding host complement regulators from plasma or other body fluids: FH for protection against the alternative pathway activation and C4b-binding protein for inhibition of the classical and lectin pathways . Binding of FH has been thought to be simple recruitment of host FH onto the microbial surface since FH acts as a cofactor for factor I in the degradation of the central complement component C3b [30] . This inactivation is essential for microbial survival in nonimmune plasma or blood , since it prevents opsonophagocytosis and microbial lysis by the membrane attack complexes [31] . Microbes recruit host FH by binding it via two separate sites , one within the domains 6–7 and the other in the C-terminal FH19-20 ( Table S1 ) , but the reason for using these sites has remained unexplained . Our new data show , first , that the microbes we studied not only use FH19-20 , but in particular the same area on FH domain 20 , which we have named the “common microbial binding site” ( Figure 3 , panel B ) . Second , our data show that binding via this particular site allows the formation of a tripartite microbial protein∶FH∶C3b complex ( Figure 5 , panel C ) . Third , and most importantly , our data show that formation of the tripartite complex enhances FH-mediated inactivation of C3b . This explains why many kinds of microbes have evolved to utilize this common microbial binding site on FH . We analyzed the interaction site between the carboxyl-terminus of FH and microbes by measuring the effect of mutant FH19-20 proteins on binding of wt FH19-20 to five important human pathogens ( Gram-negative and Gram-positive bacteria and a yeast ) . Next we analyzed FH19-20 binding by three structurally non-related , FH binding proteins , two from spirochetes , OspE from B . burgdorferi sensu stricto [18] and FhbA from B . hermsii [32] , and Tuf from P . aeruginosa [21] . To our great surprise all the microbes and microbial proteins studied bound FH via heavily overlapping binding sites on domain 20 ( Table 1 , Figure 3 , panel A ) . We found three key amino acids ( R1182 , R1203 , R1206 ) that affected binding to all the studied microbes and three more ( K1188A , R1210A , R1215A ) that affected binding to at least three out of seven microbes analyzed . We believe that this site , the common microbial binding site , will be found to be used by many other pathogenic microbes too . We did not use full length FH with point mutations in these experiments since microbes have often two binding sites for FH ( Table 1 ) and expression and purification of full-length FH with mutations in both the microbial binding sites might not result in easily interpretable results . Since the different microbial proteins are non-homologous it is expected that they use slightly different residues within or next to the common microbial binding site on FH20 to form , for example , hydrogen bonds and hydrophobic contacts . An example of this is seen with OspE since mutations of two residues of FH19-20 ( W1183 and E1198 ) that are not used by several other microbes had the most striking effect on OspE binding to FH19-20 . Use of variable residues within the same area does not compromise the key finding that the used microbes share a common binding area on FH domain 20 but indicates variability in the structure of the microbial molecules binding to the common shared site on FH . It is obvious that only detailed structural analysis of different microbial FH-binding proteins in complex with FH19-20 will show how important each residue within or next to the common site is for the interaction . At least three non-homologous microbial proteins and , in addition , four microbial species without known homologues of these proteins utilize the same site on FH20 . For some of these microbes , it is not known which surface molecule recruits host FH and it is possible that , at least in some cases , the surface molecules are not proteins but carbohydrates . FH is known to bind to several negatively charged carbohydrates [33] and the common microbial binding site on FH20 overlaps with the site responsible for binding to at least one host carbohydrate , heparin ( Figure 4 , panels A–C ) [27] . It remains to be studied if any microbe binds to the common microbial binding site on FH20 via a carbohydrate , and if carbohydrate binding to FH domain 20 could promote the FH∶C3b interaction through formation of a tripartite complex , similarly to the studied microbial proteins . Why have different microbes evolved to utilize domain 20 , and practically the same particular site on this domain , in recruitment of FH ? Our results provide three reasons for this . First , our work shows that FH bound to microbial surface via domain 20 can also bind the C3d part of C3b by domain 19 ( Figure 5 , panels A and B ) . This brings FH near to its main target , C3b , and allows complement inhibition . On the basis of the superimposition of three structures , our recently solved structure of a microbial FH-binding protein ( OspE ) in complex with FH19-20 ( Bhattacharjee et al , submitted ) and the previously solved structures of FH19-20 in complex with C3d , [8] and of the C3b ( containing the C3d [29] ) , it became clear that FH19-20 can bind simultaneously to a microbial protein and C3b ( Figure 5 , panel C ) . Furthermore , in this superimposition the microbial binding site is also directed towards the surface to which C3b is bound to via the thioester site and is therefore readily available for the microbial molecules in general . Second , the site on domain 20 is available under physiological conditions: the previously described physiologically important heparin binding site [27] , [34] , [35] and the common microbial site overlap to some extent ( Figure 4 , panels A–C ) . Ferreira and coworkers [36] have also emphasized that the cofactor site on domains 1–4 must not be disturbed upon binding of FH to a microbe and our common microbial binding site fulfills also this criterion . Third , utilization of this particular common microbial binding site provides more efficient down regulation of complement activation on the microbe . The intact α′-chain of C3b disappears more efficiently when FH and microbial proteins are present ( Figure 6 , Panels A and B ) . This is due to the carboxyl-terminus of FH , as practically no enhancement was seen by using FH1-4 , which has cofactor activity but does not have the common microbial binding site on domain 20 ( Figure 6 , Panel C ) . The tripartite microbial protein∶FH∶C3b/C3d complexes are clearly formed in fluid phase in vitro ( Figure 5 ) . It is , however , uneasy to demonstrate the complex formation on microbial surface since C3b is bound covalently to the molecules on the target surface and the formed tripartite complex is easily broken upon purification of C3b from the cells . To indicate the complex formation on cell surfaces we therefore used an experimental setup where binding of FH19-20 to whole intact microbes was analyzed in the presence of soluble C3d and saw that addition of C3d could , indeed , enhance the binding ( Figure S4 ) . Since the extrinsic C3d can enhance binding of FH onto the microbial surface it is highly likely that the microbial proteins could also enhance formation of the tripartite complex on the microbial surface , at least if the density of C3b depositions was high enough . The highest C3b concentration occurs on the target surface areas where alternative pathway activation is vigorously amplified via the feedback loop . It would be most beneficial for the microbe if the tripartite complexes were formed within those areas leading to maximal complement down regulation exactly at the spots where it is needed most . For the tripartite complex formation the microbial FH-binding molecule needs to be – or bend – next to the C3b molecule but most , if not all , of the FH-binding microbial proteins which have been structurally characterize are either long molecules ( e . g . streptococcal M protein ) or have a flexible tail that allows twisting and tilting ( e . g . OspE , Bhattacharjee et al , submitted ) . Therefore at least some of the microbial FH-binding molecules seem to be able to operate on a broader area of the surface than just the exact spot they are attached to . The area where the tripartite complexes could be formed might in addition be expanded by lateral movement of the lipid tail or membrane anchor of the microbial FH-binding molecules on at least surface of Gram-negative bacilli . An FH-related protein found in plasma , FHR-1 , has a C-terminal domain that differs from domain 20 of FH only by two residues . The differences are located close to the microbial binding site and it remains to be studied if FHR-1 binds similarly to the used microbes , and if possible recruitment of FHR-1 is functionally beneficial or unfavourable for the microbes as FHR-1 does not have any cofactor-activity . Clearly , formation of the tripartite complex is the reason for the increase in the regulatory function of FH caused by the microbial proteins . As far as we know , this kind of enhancement has neither been suggested , nor studied before . Instead , it has been suggested that microbes mimic host structures and thereby bind FH and other complement regulators [37] . Although microbes , heparin or endothelial cells do bind to overlapping sites on FH , this is not exactly molecular mimicry as the binding sites are not identical . The structures involved are completely different and they appear to differ from organism to organism . We and others have recently shown that host cells recruit FH via domain 20 [27] , [35] and it remains to be studied if this leads to elevated FH function due to tripartite complex as in the microbial proteins [8] , [28] . If this were the case , microbes utilizing the common microbial binding site on FH domain 20 would have functional , not molecular , mimicry of host cells . So far there is , however , no evidence of this . The identified common microbial binding site on FH domain 20 represents a surprising type of host-pathogen contact – a single site on a host molecule utilized by several kinds of microbes in immune evasion . Such a common immune evasion site for both bacterial and eukaryotic pathogens has not been reported earlier . We call this kind of conserved site for microbial immune evasion a “superevasion site” and suggest that superevasion sites may occur on other powerful down regulators of host immunity , too . The concept of a microbial superevasion site is valid not only for down regulators of immunity , such as FH , but also for host immune activator molecules such as immunoglobulins . It is probable that , for example , staphylococcal protein A [38] , streptococcal protein G [39] , and E . coli protein EibD [40] are not the only microbial proteins that bind to a conserved site on IgG leading to prevention of the effector functions of immunoglobulins . This site on the Fc part of IgG is probably an example of a superevasion site on immune activator molecules . In this study we have identified a conserved microbial binding site on domain 20 of the important complement regulator FH . We have shown that , by binding to the common binding site on FH , microbial proteins enhance the FH∶C3b interaction by enhancing their interaction , thereby increasing down regulation of C3b and leading to efficient evasion of complement attack and presumably to increased survival of the microbes in the host . The identified common microbial binding site on FH is the first example of a “superevasion site” pointing to new avenues not only in research on immune evasion by microbes but also in research aimed at novel vaccines and antimicrobial agents .
The outer surface proteins OspE and OspA from B . burgdorferi sensu stricto strain N40 were cloned , expressed and purified as described [18] . FhbA was cloned and purified from B . hermsii strain MAN [32] , and Tuf from a P . aeruginosa blood isolate strain similarly as described earlier [20] . Cloning and purification of wt FH19-20 and the FH19-20 mutants have been described earlier [7] , [27] , [41] . Circular dichroism spectras of six mutants ( R1182A , W1183L , K1188A , E1198A , R1203A , R1206A ) were compared to wt to confirm proper folding of the mutants ( Figure S3 , panel A ) . The capacity of these mutants to form oligomers was compared to wt FH19-20 using gel filtration on a Superdex 75 10/300 GL column ( Figure S3 , panel B ) . FH1-4 was produced as described [42] . C3 and FH were purified from human plasma and C3b generated with trypsin as described [43] . C3d was a kind gift from Prof . D . Isenman , Univ . of Toronto , Canada . Factor I was purchased from Calbiochem/MerckMillipore ( Merck , Darmstadt , Germany ) and BSA , gelatin and heparin from Sigma-Aldrich ( St . Louis , MO , US ) . The wt FH19-20 , FH , OspE , and C3b were labeled with 125I using the IodoGen method ( Thermo Scientific Pierce , Rockford , IL , US ) . The strains of Pseudomonas aeruginosa , Haemophilus influenzae , Streptococcus pneumoniae , Staphylococcus aureus and Candida albicans we used were isolated from blood cultures of septic patients and were kind gifts of Dr . K . Haapasalo-Tuomainen , HUSLAB , Helsinki Univ . Central Hospital , and Univ . of Helsinki , Finland . Bordetella pertussis was a kind gift of Dr . Quishui He , Pertussis Reference Laboratory , Turku , Finland . The used serum sensitive Haemophilus influenzae strain is isolated from a throat swab of a healthy individual . To detect binding of FH or FH19-20 to the microbes , the bacteria and yeast were first washed three times with PBS . Approximately 1×108 cells/reaction were incubated with radiolabeled FH or FH19-20 ( 40 , 000 cpm/reaction ) in the absence or presence of C3d ( 0–100 µg/ml ) in 50% PBS containing 0 . 1% gelatin ( GPBS ) at 37°C for 20 min with agitation ( 1 , 200 rpm ) . Cell-associated and free radioactive proteins were separated by centrifugation ( 10 , 000× g , 3 min ) of the samples through 20% sucrose in GPBS . Radioactivities in the supernatant and pellet fractions were measured with a gammacounter ( Wallac , Turku , Finland ) . The amounts of bound proteins were calculated as percentages of the total radioactivities in the corresponding pellets and supernatants . The experiments were performed three times in triplicate . Nunc Polysorp BreakApart plates ( Thermo Scientific , Rockford , IL , US ) were coated with either bacteria ( 1×106/well in phosphate-buffered saline , PBS , at 37°C for 12 hours ) or proteins ( 5–25 µg/ml in PBS at 4°C for 12 hours ) . The wells were blocked ( 0 . 5% BSA/PBS , 60 min at 22°C , or 0 . 5% BSA/50% PBS for the experiment shown in the Figure 4 , Panel C ) and washed with PBS . Serial dilutions of proteins were mixed with 125I-FH19-20 or 125I-OspE ( 50 , 000 cpm/well ) in a separate 96-well microtitre plate ( Greiner Bio One , Frickenhausen , Germany ) before transferring into the coated wells . After incubation ( 37°C , 60 min ) and washing with PBS ( or 50% PBS for the experiment shown in Figure 4 , Panel C ) , the radioactivity in each well was measured with a gamma-counter ( Wallac , Turku , Finland ) . The inhibition curves were fitted using non-linear regression of a “log ( inhibitor ) vs . response” model using GraphPad Prism software ( version 5 . 0b , GraphPad Software , CA , US ) . The mean inhibitory concentrations ( IC50-values ) were calculated from the fitted curves . All the assays were performed three times using triplicate wells . To measure cofactor activity 125I-C3b ( 100 , 000 cpm/assay ) was mixed with factor I ( 16 µg/ml ) in the absence or presence of FH or FH1-4 ( 8–85 µg/ml ) and OspE , FhbA , and Tuf ( 50 µg/ml ) . Mixtures were incubated at 37°C for 5 min and , after adding β-mercaptoethanol , the samples were heated ( 3 min at 93°C ) and run on 10% SDS-PAGE gels . The gels were subjected to autoradiography and cofactor activity was evaluated as the intensity of the C3b α′-chain measured with GelEval-programme ( FrogDance Software , Dundee , UK ) . Values are expressed as means ± SD . All statistical analyses were performed using GraphPad Prism software and statistical differences were calculated with unpaired t-tests . | Complement is an important arm of innate immunity . Activation of this plasma protein cascade leads to opsonization of targets for phagocytosis , direct lysis of Gram-negative bacteria , and enhancement of the inflammatory and acquired immune responses . No specific signal is needed for activation of the alternative pathway of complement , leading to its activation on all unprotected surfaces . Pathogenic microbes need to evade this pathway , and several species are known to recruit host complement inhibitor factor H ( FH ) to prevent the activation . FH is important for protection of host cells , too , as defects in FH lead to a severe autoreactive disease , atypical hemolytic uremic syndrome . We have now identified at the molecular level a common mechanism by which seven different microbes , Haemophilus influenzae , Bordetella pertussis , Pseudomonas aeruginosa , Streptococcus pneumoniae , Candida albicans , Borrelia burgdorferi and B . hermsii , recruit FH . All microbes bind FH via a common site on domain 20 , which facilitates formation of a tripartite complex between the microbial protein , the main complement opsonin C3b , and FH . We show that , by utilizing the common microbial binding site on FH20 , microbes can inhibit complement more efficiently . This detailed knowledge on mechanism of complement evasion can be used in developing novel antimicrobial chemotherapy . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"complement",
"system",
"immunology",
"microbiology",
"host-pathogen",
"interaction",
"immune",
"defense",
"bacterial",
"pathogens",
"mycology",
"medical",
"microbiology",
"microbial",
"pathogens",
"biology",
"yeast",
"immune",
"system",
"immunity",
"innate",
"immunity"
] | 2013 | Microbes Bind Complement Inhibitor Factor H via a Common Site |
Assess the feasibility of engaging youth to disseminate accurate information about gene by environmental ( GxE ) influences on podoconiosis , a neglected tropical lymphedema endemic in southern Ethiopia . A cross sectional survey was conducted with 377 youth randomly selected from 2 districts of Southern Ethiopia . Measures included GxE knowledge ( 4 true/false statements ) , preventive action knowledge ( endorse wearing shoes and foot hygiene ) , causal misconceptions ( 11 items related to contagion ) and confidence to explain GxE ( 9 disagree/agree statements ) . Over half ( 59% ) accurately endorsed joint contributions of gene and environment to podoconiosis and preventive mechanisms ( e . g . , wearing protective shoes and keeping foot hygiene ) . Multivariable logistic regression showed that youth with accurate understanding about GxE contributors reported having: some education , friends or kin who were affected by the condition , and prior interactions with health extension workers . Surprisingly , higher accurate GxE knowledge was positively associated with endorsing contagion as a causal factor . Accuracy of GxE and preventive action knowledge were positively associated with youth’s confidence to explain podoconiosis-related information . Youth have the potential to be competent disseminators of GxE information about podoconiosis . Interventions to foster confidence among youth in social or kin relationships with affected individuals may be most promising . Efforts to challenge youth’s co-existing inaccurate beliefs about contagion could strengthen the link of GxE explanations to preventive actions .
Advances in genomics are increasing scientific understanding that most health conditions worldwide are caused by the joint influence of genetic and environmental ( GxE ) factors [1] . However , the mechanisms underlying GxE interactions are complex and not well understood by the public [2] . Accordingly , misunderstandings that health conditions with genetic underpinnings cannot be prevented have been well documented in the developing world [3–6] . Global leaders have called for stepping up efforts to increase genomic literacy in low and middle-income countries ( LMICs ) [7] [8 , 9] . Among the challenges to achieving this imperative is that LMICs have limited health service infrastructure , low levels of general literacy and a majority of the populace lives in isolated rural settings . Funding opportunities such as the Human Heredity and Health in Africa ( H3Africa ) have been initiated to address these challenges [10] . Concurrent with advances in genomics , around the world , cadres of lay health workers ( LHWs ) have been engaged to expand the reach of preventive health services for a broad array of communicable and noncommunicable health conditions [11 , 12] . LHWs carry out a number of functions to promote health including the provision of culturally relevant health information that can be conveyed in everyday community settings [13 , 14] . Though it has been suggested that such LHWs might play an important role in promoting community-wide GxE literacy in the developed world to date there has been little consideration of this possibility in LMICs [15 , 16] [17] . Efforts directed to disease prevention often target youth as a means to encourage the establishment of healthy behaviors . Considering youth as potential disseminators of information about GxE health influences offers several other advantages . As members of the community , youth are aware of existing interpretations of the causes of health conditions held by their peers and adults in the community [18] . In the context of HIV/AIDs and other STIs , reproductive health , and malaria , studies suggest that youth may more easily process and understand new information [3] . Youth also have been found to have greater accuracy in considering the influence of genes on health conditions than older adults [19] . These information-processing assets have been attributed to youth’s neurodevelopment stage , exposure to schools , youth networks and the media [20] . The feasibility of engaging LHWs to increase GxE literacy in LMICs rests on whether individuals can be identified with the capability and confidence to serve as GxE information disseminators . Podoconiosis , a non-filarial lymphedema endemic in highland Ethiopia , offers an excellent context to consider the feasibility of engaging youth LHWs for promoting GxE literacy . Globally , it is estimated that four million people in tropical Africa , Central and South America , and Southeast Asia have the condition; approximately 1 . 5 million people are affected in Ethiopia [4] . The condition develops when genetically susceptible individuals are exposed to irritant particles in volcanic soil via walking and farming barefoot [21–23] . Podoconiosis can be prevented if susceptible individuals begin wearing shoes at an early age and do so consistently [21–23] . Prior health education efforts to promote shoe wearing among at-risk individuals have been based on extensive qualitative and rigorous quantitative evaluation of these interventions show that adults in these communities continue to harbor misconceptions that the condition is not preventable because of its heritability [5 , 24 , 25] [5 , 6] [26] . A number of other misconceptions are common including beliefs that the condition is contagious or influenced by other environmental exposures ( e . g . , snake bites ) [6 , 25 , 27 , 28] . In turn , these inaccurate beliefs are associated with risk behaviors such as consistently walking barefoot [5 , 6 , 25] and holding stigmatizing attitudes towards patients [5 , 6 , 29 , 30] . Whether engaging youth as disseminators of accurate GxE information could help to eliminate these misconceptions , reduce social stigma towards affected individuals and encourage preventive actions ( i . e wearing shoes ) is largely unexplored . In order to consider the feasibility of engaging rural Ethiopian youth to serve as information disseminators in the spirit of LHWs , we posed four research questions: ( 1 ) what is the prevalence of youth misconceptions about the causes of podoconiosis ? ; ( 2 ) what factors are associated with accurate understanding of GxE contributions to podoconiosis among youth ? ; ( 3 ) is GxE accuracy associated with correct endorsement of preventive actions ? ; and ( 4 ) is GxE accuracy associated with confidence to explain the causes of podoconiosis ?
The questionnaire was initially developed in English , translated into Amharic and Wolayitigna ( common dialect in the targeted kebeles ) , then checked for accuracy and back-translated to English . Interviewers who had previous experience in similar studies were recruited using informal networks . These individuals ( N = 10 ) received three-days of training on the objectives of the study , items included in the survey questionnaire , and how to carry out the survey . The survey was pilot tested in two nearby rural kebels ( Sodo Zuria and Damot Woydie woredas ) with 30 youth to assess the adequacy of the instruments . These youth were not included in the main survey . Results of the pilot indicated a few limitations such as spelling or grammatical errors that resulted in respondents showing minor hesitation and request for clarification revisions were made . Interviewers administered the survey to the selected youth at their homes . The expectation was that the interviews would take approximately 45 minutes . Written informed consent was obtained from all participants including thumbprints for those unable to sign . Consent from parents or guardians was obtained for those ages 15 to 18 . The respondents were given exercise books and pens as compensation for participation . All aspects of this study were approved by the Addis Ababa University College of Health Science Institutional Review Board . Selection of survey measures assessed domains of knowledge , and self-efficacy related to conveying GxE information . These constructs of Social Cognitive Theory [34 , 35] , were based on prior literature on requisite competencies for LHWs to be effective [36] and on our extensive prior qualitative data collected with adults in these communities [28] . Demographic characteristics assessed included: gender , education level , and age . Additional measures of civic engagement and interactions with health extension workers also were assessed . Three domains of knowledge were assessed: GxE knowledge was based on four questions that were adapted to the causal factors contributing to podoconiosis [23 , 25 , 37] . Youth were asked to rate each statements as “true , ” “false” or “don’t know” ( e . g . , “A person can inherit proneness to have their feet be irritated by the soil . ” ) . These measures were dichotomized based on the median score ( 2 . 1 ) . Youth who scored three and four were labeled as “mostly accurate” ( 1 ) and those who scored less than three were labeled as “mostly inaccurate” ( 0 ) . Knowledge about preventive actions was measured by rated agreement with a list of 10 potential preventive actions that included both accurate and inaccurate options ( e . g . , wearing protective shoes every day–accurate; taking vaccination–inaccurate ) . Each accurate response was assigned one point . The preventive action score was dichotomized due to skewedness , with zero representing inaccurate understanding ( 0–3 ) and one indicating mostly accurate understanding ( 4–10 ) . Causal misconceptions were based on 14 questions derived from prior research conducted with adults living in podoconiosis-endemic communities [26] . Misconceptions were assessed with two subscales , a “contagion score” based on responses to 11 questions assessing perception of podoconiosis as a contagious disease . A second subscale was labeled “other misconceptions” and included three questions related to beliefs about bacteria , poor nutrition and evil eye as causes of the disease . Each endorsed misperception was assigned ine point and then summed for a total score . Confidence to explain ( i . e . , self-efficacy ) was based on nine questions ( e . g . “I am confident that I could explain to other people why some individuals develop podoconiosis and others do not” ) with three response categories ( 1 = Disagree , 2 = Undecided , 3 = Agree ) [26] . Scores were dichotomized as being above or below the median , where zero represented “less confidence” and one indicated “more confidence” . Extracurricular civic engagement was based on self-reported involvement in extracurricular activities at school , youth association , Sunday schools and other leadership-oriented experiences . In Ethiopia more than 30 , 000 health extension workers provide outreach services and disseminate health information to the general public to encourage health promoting habits ( Federal Ministry of Health , 2007 ) . Reported contact with health extension workers ( HEWs ) was assessed as they may have provided opportunities for youth to increase knowledge of health issues and podoconiosis , specifically . Youth who were visited by HEWs at their residence one year prior to the study were compared to those reporting no contact or visit . Data were analyzed using SPSS version 20 software . Frequencies and distributions were examined to check for out-of-range values and other errors in the data . After data cleaning , descriptive analyses , bivariate analyses and logistic regressions was performed . Cross tabulation and χ2 tests were performed for associations amongst variables . Level of statistical significance was set at p <0 . 05 ( two-tailed ) . Variables with significant associations in cross-tabulation were entered into the logistic regression models to test in turn their association with accuracy of GxE knowledge , accuracy of endorsed preventive actions and confidence to explain GxE causes of podoconiosis . We ran Pearson’s correlation to assess the association between accurate GXE knowledge and prevalent misconceptions ( i . e . , inaccurate causal beliefs ) . Odds ratios were computed for the likelihood of having mostly accurate ( 1 ) and mostly inaccurate ( 0 ) understanding about GxE influences and appropriate preventive actions . Odds ratios were also computed for the likelihood of being more confident or less confident to explain podoconiosis-related information .
Half of youth believed that podoconiosis is a contagious disease and over one-third also endorsed several causal factors that constitute misconceptions including evil eye , and snake bites . Eleven percent of participants responded correctly to all four of the GxE knowledge questions related to podoconiosis , 34 . 5% answered three correctly , and 55% answered two or less questions correctly . The highest proportion of correct responses ( 82% ) was agreement with the question “walking barefoot triggers podoconiosis among susceptible individuals” ( See Table 2 ) . Additionally , 61% agreed that susceptibility to podoconiosis is passed from generation to generation . A number of preventive actions were erroneously thought to be protective against podoconiosis , including vaccination ( 87% ) and avoiding walking barefoot in cold weather or dew ( 87% ) . Additionally , one-third of youth incorrectly suggested avoiding personal contact with affected people , and nearly three-quarters endorsed avoiding wearing second-hand shoes , both indicators of belief in contagion . However , a majority of the youth also accurately endorsed regular foot washing ( 86% ) and wearing protective shoes everyday ( 70% ) . Due to their high co-prevalence , we tested the association of accurate GxE knowledge with prevalent misconceptions ( e . g . , endorsement of podoconiosis as a contagious disease ) . We found a modest positive correlation between GxE knowledge and contagion beliefs ( r = 0 . 23 , p<0 . 05 ) . Youth with more accurate understanding of GxE contributors to podoconiosis also were more likely to harbor “other misconceptions” ( r = 0 . 13 , p<0 . 05 ) . Participants with 7th to 12th grade education , who reported having affected family members or a friendship relationship with an affected individual were significantly more likely to be accurate in their understanding ( correct responses to 3 of 4 ) of GxE contributors to podoconiosis than their peers ( See Table 3 ) . Endorsement of contagion as a cause of podoconiosis and interactions with HEWs also were significantly associated with accurate understanding of GxE contributors . Variables found to be significant in bivariate analyses were tested in multivariate logistic regression to predict accurate GxE knowledge ( See Table 4 ) . Affected status , having friendships with affected individuals , reporting any contact with HEWs , having some formal education and having a high contagion inaccuracy score were significantly associated with increased accuracy of GxE understanding . Youth who had attended formal schooling from grade 1st to 6th or 7th to 12th education level had 11 . 8 and 8 . 7 times , respectively ( 95% CIs: 2 . 2–16 . 6; 1 . 7–13 . 4 ) the odds of having accurate GxE knowledge related to podoconiosis as those without formal schooling . Affected youth had 8 . 0 times the odds of being accurate in GxE knowledge as unaffected youth ( 95% CI: 2 . 9–16 . 1 ) . Similarly , youth with affected friends had three times the odds of having accurate understanding as those who did not report having friends with podoconiosis ( 95% CI: 1 . 3–6 . 9 ) . Youth who reported contact with HEWs had 3 . 8 times ( 95% CI: 2 . 0–7 . 2 ) the odds of knowing about the joint contribution of GxE in the development of podoconiosis as those who did not report contact with HEWs . Youth with stronger contagion beliefs had 1 . 2 ( 95% CI 1 . 0–1 . 2 ) times the odds of holding accurate GxE knowledge as those with weaker contagion beliefs . We tested a logistic model that included the factors found above to be associated with accuracy of understanding in a model for appropriate endorsement of preventive actions ( dichotomized as 0 ( 1–3 ) = mostly inaccurate and 1 ( 4–10 ) = mostly accurate ) . Results showed that youth who reported having a friendship with an affected individual had 2 . 4 times ( 95% CI: 1 . 0–5 . 4 ) the odds of of endorsing appropriate preventive actions as those without affected friends ( See Table 4 ) . Youth who reported contact with HEWs had 3 . 8 times ( 95% CI: 2 . 0–7 . 0 ) the odds of endorsing appropriate preventive mechanisms as those who did not report contact with HEWs . Youth who endorsed contagion as a cause of podoconiosis had lower odds of accurately endorsing preventive actions as those who did not endorse contagion beliefs ( OR = 0 . 92 , 95% CI: 0 . 8–1 . 0 ) . Counter to our expectation , accurate understanding about GxE contributors to podoconiosis was not associated with endorsement of appropriate preventive actions ( e . g . , shoe wearing , p = 0 . 92 ) . We tested a logistic model that included the factors found above to be associated with appropriate endorsement of preventive action to test their association with confidence to explain the causes of podoconiosis to others ( dichotomized as 0 ( 1–3 ) = less confidence , 1 ( 4–10 ) = more confidence ) . Results showed that those affected by podoconiosis had lower odds of being confident to explain GxE influences as unaffected youth ( OR = . 98; 95% CI: 0 . 1–1 ) . Accuracy of GxE and preventive action knowledge were significantly associated with confidence to explain the causes of podoconiosis ( See Table 4 ) . For each unit increase in GxE accuracy and endorsement of preventive action , the odds of being confident increased by a factor of 1 . 3 ( 95% CI: 1 . 0–1 . 7 ) and 1 . 3 ( 95% CI: 1 . 1–1 . 5 ) respectively , when compared to the reference group .
Our findings show that youth in podoconiosis-endemic communities have similar misconceptions as those reported previously among adults in neighboring communities [4 , 6 , 25 , 38] . Like adults , youth endorsed multiple causes of podoconiosis that included both contagion and environmental exposures ( e . g . , snake bite , evil eye ) . However , a sizable proportion of youth responded accurately to questions about the joint contributions of heredity and environment in the development of podoconiosis . It is concerning that youth with the most accurate understanding of GxE contributors to podoconiosis also were most likely to harbor misconceptions about the causes of podoconiosis . Thus , endorsement of GxE contributors did not obviate other erroneous causes such as contagion . Our previous qualitative work with youth in these communities that aimed to understand youth’s explanatory mental models documented a similar connection between GxE and contagion beliefs [28] . In our previous report we posited that beliefs that susceptibility to podoconiosis passes from generation to generation ( i . e . , accurate GxE knowledge ) co-occurred with observations of families living in close proximity to each other where bodily fluids could be exchanged . Thus , youth perceived a “both/and” connection in which inherited susceptibility and contagion explanations could co-exist . Generally , health education programs assume that accurate health information can correct inaccurate beliefs . However , our findings suggest that community members can hold numerous , even contradictory , notions of causation simultaneously . The implications of this present challenges for engaging youth as LHWs . Contagion beliefs not only shape priorities given to preventive actions but also fuel stigmatizing beliefs . Endorsement of both GxE and contagion as causes of podoconiosis also may have attenuated the association between related knowledge and endorsement of appropriate preventive behaviors . Thus , any efforts to engage youth as LHAs to promote preventive actions among affected individuals would also need to challenge beliefs about contagion . A large body of literature has shown that contagion beliefs have deep roots and are very difficult to override [28 , 39] . In this study , youth who reported having social ties to affected individuals were more likely to be accurate in their understanding of GxE contributors . Similarly in our prior intervention study , affected adults were less likely to endorse contagion beliefs than unaffected adults [26] . Individuals with social ties to affected individuals , particularly those who have engaged with health extension workers may be more accepting of alternative causal explanations that are linked to preventive actions . Level of education was also associated with GxE knowledge accuracy . Youth without formal education had lower knowledge about podoconiosis , which is consistent with earlier studies among adults [24 , 40 , 41] . Lack of schooling limits youth’s access to health information provided in school curriculums and extracurricular activities . Youth engaged in formal education could be optimal disseminators of accurate GxE information . Indeed , prior studies identified some schooling to be a crucial qualification for LHWs in Peru and the Republic of the Congo [42 , 43] . However , education level was not associated with accurate knowledge of preventive actions or with confidence to explain in this sample . LHA training targeted to school settings would need to strengthen these skills among youth . The study further showed that accurate knowledge about GxE contributors of podoconiosis may not necessarily lead to understanding accurate preventive actions . Experts argue that individuals’ understandings of causes of health conditions have strong influences on what they perceive as best preventive actions to take to lower risk of developing diseases [44 , 45] . The logical assumption is that provided with accurate causes of health conditions , individuals will be more likely to identify preventive actions to lower the risk of developing ailments . The concern is that holding a number of competing beliefs about the causes of podoconiosis could muddle youth’s understanding about what to do to prevent the conditions . LHWs training should not assume that improved understanding and acceptance of GxE causal mechanisms will override beliefs about contagion . Curricula specifying the mechanism through which soil exposure ( e . g . , silica particles ) leads to lymphatic inflammation could be included in school science classes . In this environment , classroom exercises could be used to encourage youth to consider these mechanisms and debate the other causal beliefs using peer discussions and co-teaching . It is heartening that youth with higher GxE accuracy and preventive knowledge also had the highest confidence in their ability to explain the causes of podoconiosis . It has been suggested that effective LHWs have conceptual understanding and confidence to effectively communicate with their constituencies that , in turn , can be capitalized upon to facilitate individual- and community-level health promotion [36] . Similarly , a recent study found confidence to talk about breast and cervical cancer screening information to significant others to be an important individual characteristic for LHWs [36] . However , affected youth had lower self-confidence , despite their greater GxE knowledge compared to unaffected youth . This suggests that confidence-building strategies may need to be different for affected and unaffected youth . Stigma related to being affected by podoconiosis may have lowered self-esteem of these youth and inhibit their confidence to discuss the topic . It also should be noted that further assessment might be needed to identify the factors that motivate young people , both those affected and unaffected , to take part in podoconiosis prevention campaigns . The study had several noteworthy strengths including a high participation rate , and the survey development was informed by extensive qualitative data collection . However , there were a few limitations . The study was conducted in only two rural communities that may not be generalizable to youth in urban areas or other LMIC settings . The survey focused specifically on factors associated with youth’s accuracy of knowledge of podoconiosis causes , GxE influences and preventive actions with the aim of informing health literacy-building activities . Numerous other factors could influence the viability of youth in these settings serving as LHWs . Indeed , studies show that role-related factors , social network and trust , and other characteristics of target communities can influence motivation and performance of LHWs that were not assessed [11 , 36 , 46 , 47] . Moreover , knowledge improvements unto themselves may be necessary for health promotion but are unlikely to be sufficient to influence behavior change . Our findings of coincident endorsement of causal beliefs related to GxE and contagion may have been influenced by social desirability bias , with participants eager to endorse multiple causes and inherently multiple solutions to the problem . These results are preliminary but support the pursuit of further research to consider youth assets and deficits relating to their potential role as LHWs . Additionally , extensive qualitative and quantitative studies will be required to guide identification of youth who may be most willing and competent in this role and to evaluate interventions to help youth acquire skills and assess community impact . The potential to avoid disparities in reach of emerging genomic knowledge warrants these efforts . | This study considers the feasibility of engaging rural Ethiopian youth as lay health workers ( LHWs ) with the objective to improve community understanding of the joint influences of genetics and environment on health . Identifying LHWs to accurately convey contributors to the heritable but preventable neglected tropical disease of podoconiosis provides an optimal context to address this question . Misunderstanding that inherited susceptibility to podoconiosis makes the disease unavoidable has led to numerous negative social consequences ( e . g . , stigma ) and poor uptake of protective footwear . We report data from a pilot study that included a cross-sectional survey of youth ages 15–24 in rural communities with endemic podoconiosis . Results provide preliminary support that a sizeable group of youth hold accurate knowledge about gene x environment influences and self-evaluate as being confident to explain these associations to others . Research to evaluate strategies to engage youth as LHWs and the impact of these approaches on communities’ understanding of the joint influences of genetics and environment in this context is needed . This manuscript fills an important gap in the literature about neglected tropical diseases , as it suggests opportunities to improve the prevention of podoconiosis and reduce misconceptions and stigma through engagement of LHWs in Ethiopia . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"education",
"behavioral",
"and",
"social",
"aspects",
"of",
"health",
"engineering",
"and",
"technology",
"sociology",
"tropical",
"diseases",
"social",
"sciences",
"neuroscience",
"parasitic",
"diseases",
"health",
"care",
"academic",
"skills",
"cognitive",
"psychology",
"global",
"health",
"literacy",
"safety",
"equipment",
"neglected",
"tropical",
"diseases",
"safety",
"public",
"and",
"occupational",
"health",
"protective",
"footwear",
"elephantiasis",
"podoconiosis",
"health",
"systems",
"strengthening",
"schools",
"protective",
"clothing",
"psychology",
"equipment",
"biology",
"and",
"life",
"sciences",
"cognitive",
"science",
"health",
"education",
"and",
"awareness",
"health",
"care",
"policy"
] | 2018 | Rural youths' understanding of gene x environmental contributors to heritable health conditions: The case of podoconiosis in Ethiopia |
Correlative analysis of molecular markers with phenotypic signatures is the simplest model for hypothesis generation . In this paper , a panel of 24 breast cell lines was grown in 3D culture , their morphology was imaged through phase contrast microscopy , and computational methods were developed to segment and represent each colony at multiple dimensions . Subsequently , subpopulations from these morphological responses were identified through consensus clustering to reveal three clusters of round , grape-like , and stellate phenotypes . In some cases , cell lines with particular pathobiological phenotypes clustered together ( e . g . , ERBB2 amplified cell lines sharing the same morphometric properties as the grape-like phenotype ) . Next , associations with molecular features were realized through ( i ) differential analysis within each morphological cluster , and ( ii ) regression analysis across the entire panel of cell lines . In both cases , the dominant genes that are predictive of the morphological signatures were identified . Specifically , PPARγ has been associated with the invasive stellate morphological phenotype , which corresponds to triple-negative pathobiology . PPARγ has been validated through two supporting biological assays .
Genome-wide association studies of expression and clinical data have emerged as a powerful methodology for identifying biomarkers of human diseases . While the literature is rich with supervised or unsupervised clustering of genomic information [1] , methods for studying the relationships between genomic and physiological responses remain limited . This paper contributes to computational protocols for associating morphometric data , collected through phase contrast microscopy , with genome-wide gene expression data . While genome-wide array expression data provide on average a few readouts with structured measurements for an ensemble of colonies , imaging provides one readout per colony and captures the inherent heterogeneity of a population . However , images are composed of unstructured data that require detailed segmentation and representation for the underlying samples . The net result is subtyping , based on computed morphometric features , and a list of associated genes against computed morphometric features for further bioinformatics analysis . This paper also demonstrates that some of the predicted genes are biologically relevant and can be tested through both in vitro and in vivo models . Most of the existing methods for clustering ( e . g . , subtyping ) concentrate on either finding subpopulations for a collection of “OMIC” data or identifying groups of genes that can be associated for each subtype . These methods relate a specific signal across measured conditions , which is appropriate for a focused experiment with a small number of conditions , and for partitioning genes into disjoint sets , thus oversimplifying biological systems . More effective clustering methods have focused on bi-clustering [2]–[4] , where bi-clustering aims to find a subset of genes that behave similarly across a subset of conditions . Still , a more effective method is to correlate expression data with known pathways , because pathways represent higher-level biological functions , where the correlation of real value data with known non-numeric pathway data ( e . g . , KEGG , BRITE ) is generally performed through kernel canonical correlation analysis ( KCCA ) [5] . The original canonical correlation analysis ( CCA ) , developed by Hotelling in 1936 , finds projections from two real-value datasets so that those projections have maximum correlations . The kernelized version extends the CCA to non-numerical values . With respect to the understanding of the mechanism of genome-wide regulation and functions , experiments have to be coordinated with the computational requirements to ensure the robustness of any biological conclusion . This is often met by varying or perturbing experimental conditions ( e . g . , multiple cell lines , different treatment conditions ) . For example , in a recent paper , microarray data were analyzed with the corresponding physiological responses and clinical metadata [6] . The experiment incorporated NCI-60 , a panel of cancer cell lines , that were incubated with Docetaxel , and the impact of the drug was characterized with GI50 ( e . g . , 50% growth inhibition dose concentration in a 48-hour assay ) . Subsequently , genes strongly correlated with GI50 were identified . Experimentally , our method is based on three-dimensional cell culture models , which introduce new computational opportunities , because the assays were imaged with phase contrast microscopy . One primary rationale for designing experiments in 3D cell culture models is that the 3D systems provide a more faithful replication of cell behavior in vivo than 2D substrata systems [7] , [8] . Mammary cells cultured on rigid 2D substrata rapidly lose many aspects of their in vivo phenotype [9] , but the use of 3D extracellular matrix cultures ( which restore the physiological cell-ECM interactions ) allow for a much more faithful replication of in vivo phenomena in culture . For example , mammary epithelial cells form polarized acini that vectorially secrete the milk protein , beta-casein , when cultured within a 3D ECM gel [7] , and breast cancer cells can be readily distinguished phenotypically from non-malignant breast cells simply by observing their aggressive growth in these assays [10] . Our experiment consists of 24 cell lines from a panel of non-malignant and malignant breast cell lines . We have developed a computational protocol that quantifies colony structures through segmentation and multidimensional representations . Such a multidimensional representation enables subsequent associations with expression data , as well as with the identification of subpopulations among all the 24 lines . Our proposed computational protocol consists of five major steps: ( i ) colony segmentation , ( ii ) morphological feature extraction , ( iii ) consensus clustering of morphological features , ( iv ) differential analysis of morphological clusters with gene expression profiles , and ( v ) association of cell-line-specific morphological features and their gene expression signatures . These computational steps are shown in Figure 1 , where colonies in each phase image are segmented from the background based on texture features . Regions containing individual colonies are extracted and subsequently represented by multidimensional indices , such as size and Zernike moments . Such a representation is translation and rotation invariant . At this point , one path allows genes that are predictive of morphogenesis to be identified . The second path identifies subpopulations through a modified consensus clustering , which finally leads to ranking those genes that differentiate each subpopulation . A few of these genes are druggable targets , and one has been selected for biological validation .
Our data set includes phase images from breast cancer cell lines grown in 3D . This data set has produced colonies from all cell lines . Following segmentation and feature extraction , each colony is represented with a multidimensional vector as discussed in the Methods section . This is followed by consensus clustering where the number of clusters is varied from 2 to 7 to examine near optimum partitioning . In order to visualize clustering results , the consensus matrix is traditionally treated as a similarity matrix and reordered using hierarchical clustering . As a result , self similar signatures are placed in close proximity . In this reordered consensus matrix , cell lines with similar morphologies are adjacent to each other , and the darker signal ( in the map ) reflects improved similarity for the purpose of visualization . Ideally , for a perfect consensus matrix , the displayed heat map should have crisp boundaries . These matrices are generated for a number of clusters , ranging from 2 to 7; the results are shown in Figure 2 . The choice of maximum cluster number ( e . g . , 7 ) is arbitrary , and the experiment can be repeated if computed consensus matrices and subsequent analysis suggested a larger number of subtypes , but this is biologically less feasible as one is interested in the simplest partition . Consensus clustering assesses stability for the identification of potential subpopulations , and provides visual feedback as a potential component for the decision-making process . For example , for , the consensus matrix has one large and one small block with crisp boundaries; and for , it appears that the large block for has been partitioned into two other blocks . Therefore , a quantitative method for assigning confidence to the selected number of clusters is needed . This is based on computing consensus distribution [11] . By computing a cumulative distribution from consensus matrices and evaluating proportional increase as a function of the number of clusters , the shape of the concentration distribution can be examined . The cumulative distribution function ( CDF ) is computed from the entire consensus matrix , whose elements are between 0 and 1 . The shape of the CDF and its progression as a function of increase in the number of clusters suggest the presence of desirable subpopulations . An earlier paper by [11] evaluated this method with synthetic and real data , proposed a new measure , a “concentration histogram” computed from the change in the shape of the CDF , and suggested that the peak in the concentration histogram corresponds to an estimate of the number of clusters . The concentration histogram of Figure 3 suggests that three clusters best represent the desired number of subpopulations . Let's examine identification of subpopulations as the number of clusters increases . At , one subpopulation contains three cell lines of , , and , as shown in Figure 4 , where their fingerprints indicate large colony size and complex texture representation displaying aggressive behaviors . At , the larger block of is approximately partitioned into two subpopulations . One subpopulation corresponds to a round symmetrical morphology expected from non-transformed 3D cell culture models . The other population corresponds to a more aggressive line labeled “grape-like” in the literature [12] . In summary , the three clusters of round , grape-like , and stellate , shown in Figure 5 , suggest the best set of subpopulations , based on morphological similarities . At spurious clusters ( not shown here ) are generated that have no clear boundaries . Examining the association between phenotypic signatures and expression data is an exploratory step , which requires molecular diversity in the data set to avoid homogeneity . Two distinct approaches are applied , where each approach brings a unique view to the data . ( I ) In the first approach , expression data associated with each cell line are grouped into their corresponding morphological cluster . As a result , genes that best discriminate between different clusters can be ranked according to their differential strength . ( II ) In the second approach , genes are ranked against each morphological feature through linear or nonlinear regression analysis . As a result , molecular predictors for positive or negative correlation can be inferred .
We compare clustering results with those from interactive methods and provide an interpretation of the morphological similarities based on their known molecular predictors . In an earlier paper [12] , an extended set of similar data was analyzed manually , and four subpopulations – round , mass , grape-like , and stellate– were labeled . However , manual analysis of individual colonies is extremely laborious and prone to user bias . Thus , we have developed a computational protocol to identify subpopulations . In our analysis , round and mass clusters are grouped together , since they have no morphological differences when imaged through phase contrast microscopy . However , the above two phenotypes can be differentiated from each other under fluorescence microscopy . The difference is due to the degree of internal organization in these phenotypes . Round colonies tend to have cells arranged in an approximately radial symmetry , while mass colonies are significantly more disorganized . This can only be visualized at higher magnification and confocal microscopy; however , these data have not been included in our analysis . Otherwise , Figure 5 is consistent with Table 1 in [12] . Results indicate that 8 out of 9 cell lines from the grape-like subpopulation express high levels of ERBB2 as a result of amplification of this gene [12] , which is differentially expressed between grape-like and round/stellate cell lines with p-value of . The exception is MDA-MB-468 , which has a significant amplification of EGFR . Collectively , these data suggest that the deregulation of signaling through the EGFR/ERBB2 signaling axis may make a strong contribution to the grape-like morphology in culture . The stellate colonies are all negative for estrogen receptors , progesterone receptors , and HER2 , a phenotype termed triple negative by pathologists and characterized by a very poor prognosis in cancer patients , as this type of tumor is highly invasive [13] . The invasive nature of the colonies formed by these cells in the 3D culture assay may be reflective of the in vivo invasive capacity of these tumor cells . Previous results for molecular predictors of morphological subpopulations indicate that the gene expression profiles of stellate colonies are the most distinct from the other two morphological classes , which is consistent with their invasive mesenchymal phenotype compared to the more epithelial colonies formed by round and grape-like cells . A brief description of the molecular predictors , listed in the previous tables , and their relevance is provided below . Consistent with the mesenchymal phenotype of these cells , PPAR , the top gene on this list ( Table 1 ) , has been reported to be a potent inducer of EMT in intestinal epithelial cells [14] . Similarly , DAB2 has been reported to be required for TGF-beta induced EMT [15] . PPAR is a nuclear receptor protein , and functions as a transcription factor . It is ( i ) regulated by thiazolidinediones ( TZD ) , a class of oral anti-diabetic drugs , ( ii ) involved in proliferation and differentiation [14] , and ( iii ) shown to be highly expressed in metastasized human breast tissue [16] . FADS1 is involved in the synthesis of highly unsaturated fatty acids such as arachidonic acid [17] , which ( i ) are metabolites that activate PPAR [18] , and ( ii ) can also be converted to prostaglandins , by cyclooxygenases . The Prostaglandin EP4 receptor ( PTGER4 ) was correlated highly with the stellate phenotype and has been implicated in migration of MDA-MB-231 cells in vitro [19] . Inhibition of EP4 has been demonstrated to have anti-metastatic effects in preclinical mouse models [20] . Poly ( ADP-Ribose ) glycohydrolase ( PARG ) was also highly expressed in stellate cells . PARG and PARP have been reported to localize to sites of DNA damage ( reviewed in [21] ) and , intriguingly , mice deficient in PARG are hypersensitive to both -irradiation and alkylating agents [22] , suggesting that high levels of PARG may contribute to resistance to DNA-damaging agents in cancer therapy . Stellate cell lines also expressed relatively high levels of Tissue Factor Pathway Inhibitor ( TFPI ) , which is found at high levels in patients with advanced cancer , yet has been proposed to have anti-angiogenic and anti-metastatic functions [23] . Multiple probes corresponding to PALM2/AKAP2 , which are alternative splicing variants of the same gene [24] , were upregulated in stellate cells . Although the function of PALM2 is not known , PALM1 has been implicated in the filopodia and spine formation during dendritic branching [25] , so it is tempting to speculate that PALM2 may contribute to the production of the stellate processes seen in these cell lines . DCBLD2 is highly expressed by metastatic cells in culture , and in lung cancer tissue at both primary and metastatic sites [26] . PPAR was also the gene most strongly associated with colony size ( Table 2 ) . Also highly associated was INSIG1 , a PPAR target gene [27] , suggesting that the upregulated PPAR is functionally active in these cells . Axl kinase levels also positively correlated with colony size . Consistent with this , Axl activity has been shown to augment MDA-MB-231 xenograft growth in mammary fat pads and subsequent lung metastasis [28] . Of the other genes associated with colony size , TMEM22 has been reported to play a role in cell proliferation in renal cell carcinoma [29] . Among the genes negatively associated with colony size ( Table 3 ) , there are several tumor suppressor genes with roles in normal mammary epithelium . F11R encodes the Junction Adhesion Molecule A ( JAM-A ) gene . This gene is highly expressed in normal mammary epithelium , but down-regulated in invasive breast cancer cells [16] . TNK1 , OVOL2 , and EPHB3 are candidate tumor suppressor genes . Deletion of TNK1 in mice results in spontaneous tumorigenesis in several tissues [30] . OVOL2 is a suppressor of c-MYC , and OVOL2-depletion by siRNA promotes cell proliferation [31] . Overexpression of EPHB3 in colorectal cancer cells inhibited proliferation in monolayer culture and growth in both soft agar assays and as xenografts [32] . Our validation strategy has two supporting components of in vitro and in vivo experiments focusing on PPAR , since it is a druggable target . PPAR is a hub for lipid metabolism and has been suggested as a therapeutic strategy for epithelial tumor types [33] . Figure 6 shows an example of vehicle control , treatment with PPAR inhibitor , and reduction in the proliferation rate , as measured by the rate of metabolism of WST1 . This result is consistent with earlier reports in 2D culture [34] that GW9662 inhibited cell growth and the survival of MDA-MB-231 . In the second case , localization of PPAR was analyzed by immunohistochemistry in normal breast tissue and in sections from triple-negative breast tumors . Other researchers [35] have examined PPAR expression in a large cohort of breast tumors , although they did not specifically analyze triple-negative tumors in their studies . Results are shown in Figure 7 , and details are included in Text S1 .
A panel of 24 breast cancer cell lines was cultured in 3D [12] . HMT-3522 S1 ( S1 ) and HMT-3522 T4-2 ( T4 ) mammary epithelial cells were maintained on tissue culture plastic [36]–[39] . The following human breast cancer cell lines were maintained on tissue culture plastic in the following manners: CAMA-1 , Hs578T , MCF-7 , MDA-MB- 231 , MDA-MB-361 , MDA-MB-415 , MDA-MB-436 , MDA-MB-453 , MDA-MB-468 , MPE-600 , SK-BR-3 , and UACC-812 were propagated in DMEM/H-21 ( Invitrogen ) with fetal bovine serum ( Gemini ) ; AU565 , BT-474 , BT-483 , BT-549 , HCC70 , HCC1569 , T-47D , ZR-75-1 , and ZR-75-B were propagated in RPMI 1640 ( Invitrogen ) with fetal bovine serum; and MCF-12A was propagated in DMEM/F-12 ( Invitrogen ) with ng/ml insulin , ng/ml cholera toxin , ng/ml hydrocortisone , ng/ml EGF ( Sigma ) , and fetal bovine serum . Three-dimensional laminin-rich extracellular matrix ( 3D lrECM ) on-top cultures [40] were prepared by trypsinization of cells from tissue culture plastic , seeding of single cells on top of a thin gel of Engelbreth-Holm-Swarm ( EHS ) tumor extract ( Matrigel: BD Biosciences; Cultrex BME: Trevigen ) , and the addition of a medium containing EHS . Cell lines with round 3D morphology were seeded at a density of cells per ; cell lines with stellate 3D morphology were seeded at cells per ; and all other cell lines were seeded at cells per . All 3D lrECM cell cultures were maintained in H14 medium with fetal bovine serum , with the exception of S1 and T4 , which were maintained in their propagation medium , for 4 days with media change every 2 days . ( I ) Cell lines were grown in 3D , and cultured colonies were imaged with phase contrast microscopy at 10× . Colonies were isolated from 3D cultures by dissolution in PBS/EDTA [40] . ( II ) Purified total cellular RNA was extracted using an RNeasy Mini Kit with on-column DNase digestion ( Qiagen ) . RNA was quantified by measuring optical density at A260 , and quality was verified by agarose gel electrophoresis . Affymetrix microarray analysis was performed using either the Affymetrix high-density oligonucleotide array human HG-U133A chip cartridge system or the Affymetrix High Throughput Array ( HTA ) GeneChip system , in which HG-U133A chips were mounted on pegs arranged in a 96-well format . Robust multi-array analysis ( RMA ) was performed to normalize data collected from different samples . The details can be found in an earlier paper [12] . For gene expression data , the sample size is small , and on the average , there are two samples per cell line . Replicates are either averaged or their medians are selected for representation . On the other hand , the sample size for image-based data is quite large , on the order of thousands . The first step in multivariate profiling is the segmentation of a colony from its immediate background . Segmentation enables the feature-based representation of each colony for subsequent clustering and correlation analysis with expression data . Both linear and nonlinear prediction models are explored to identify molecular predictors . Each model produces a different view of the analysis for subsequent biological validation . In linear regression , the relationship between two variables ( e . g . , morphology index and gene expression ) is given by ( 6 ) where the coefficients and are estimated by minimizing the norm ( e . g . , sum squared error ) : , where is known as the coefficient of determination in statistics and is the proportion of variability in a dataset that can be accounted for by the model . A general definition is given by the ratio of error in the fit ( ) to sample variance ( ) : ( 7 ) where , as before , . In linear regression , the square root of equals the Pearson product-moment correlation coefficient: ( 8 ) where is the sample mean and is the sample standard deviation . Therefore , the Pearson product-moment correlation coefficient measures the quality of least squares fitting to and in Equation ( 6 ) , i . e . , the degree of linear relationship between two variables . A value of indicates a perfect positive linear relationship , and means a perfect negative linear relationship . In the nonlinear case , the relationship is modeled by a logistic function [52]: ( 9 ) where samples are normalized to reside between and . Equation ( 9 ) can be rewritten as ( 10 ) The Pearson product-moment correlation coefficient of the transformed variable measures the fitting quality of and in Equation ( 10 ) , as well as the quality of the logistic fitting to the original data and in Equation ( 9 ) . In all cases , the p-value is computed through permutation . In each permutation step , a subset of the data is used to compute the corresponding Pearson product-moment correlation coefficients based on a higher-level taxonomy for genes being either positively or negatively correlated with morphogenesis . For each gene , from their respective taxonomy , a p-value is then computed by comparing its Pearson product-moment correlation coefficient with values , , from permutated samples . For a gene with positive value , its p-value is: ( 11 ) where is if , and otherwise . For a gene with negative value , its p-value is: ( 12 ) | Cell culture models are an important vehicle for understanding biological processes and evaluation of therapeutic reagents . More importantly , the literature suggests that tumor cells grown in 3D exhibit pronounced drug and radiation resistances that are remarkably similar to that of tumors in vivo . Therefore , the needs for quantifying 3D assays continue to grow . In this paper , we develop robust computational methods to integrate morphometric and molecular information for a panel of breast cancer cell lines that are grown in 3D . Specifically , morphometric traits are imaged through microscopy , and then quantified computationally . We then show that these morphometric traits can identify subtypes within this panel of breast cancer cell lines , and that the subtypes are clinically relevant in terms of being ERBB2 positive or triple negative . These subtypes and their representations are then associated with their molecular data to reveal PPARG as an important marker for triple-negative breast cancer . Finally , we design two independent experiments to show the validity of this marker in both 3D cell culture models and human breast cancer tissue . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"molecular",
"biology/bioinformatics",
"cell",
"biology/cell",
"growth",
"and",
"division",
"computational",
"biology/synthetic",
"biology",
"computer",
"science"
] | 2010 | Molecular Predictors of 3D Morphogenesis by Breast Cancer Cell Lines in 3D Culture |
Previous modeling studies have identified the vaccination coverage level necessary for preventing influenza epidemics , but have not shown whether this critical coverage can be reached . Here we use computational modeling to determine , for the first time , whether the critical coverage for influenza can be achieved by voluntary vaccination . We construct a novel individual-level model of human cognition and behavior; individuals are characterized by two biological attributes ( memory and adaptability ) that they use when making vaccination decisions . We couple this model with a population-level model of influenza that includes vaccination dynamics . The coupled models allow individual-level decisions to influence influenza epidemiology and , conversely , influenza epidemiology to influence individual-level decisions . By including the effects of adaptive decision-making within an epidemic model , we can reproduce two essential characteristics of influenza epidemiology: annual variation in epidemic severity and sporadic occurrence of severe epidemics . We suggest that individual-level adaptive decision-making may be an important ( previously overlooked ) causal factor in driving influenza epidemiology . We find that severe epidemics cannot be prevented unless vaccination programs offer incentives . Frequency of severe epidemics could be reduced if programs provide , as an incentive to be vaccinated , several years of free vaccines to individuals who pay for one year of vaccination . Magnitude of epidemic amelioration will be determined by the number of years of free vaccination , an individuals' adaptability in decision-making , and their memory . This type of incentive program could control epidemics if individuals are very adaptable and have long-term memories . However , incentive-based programs that provide free vaccination for families could increase the frequency of severe epidemics . We conclude that incentive-based vaccination programs are necessary to control influenza , but some may be detrimental . Surprisingly , we find that individuals' memories and flexibility in adaptive decision-making can be extremely important factors in determining the success of influenza vaccination programs . Finally , we discuss the implication of our results for controlling pandemics .
Previously , both complex [1–3] and simple models [4–7] of influenza transmission dynamics have been analyzed to determine what proportion of the population would need to be vaccinated to prevent influenza epidemics and pandemics . However , none of these modeling studies have shown whether this critical coverage can actually be reached . Here we investigate , by modeling vaccination decisions made by individuals , whether the critical coverage can be achieved through voluntary vaccination . We construct an individual-level model of human cognition and behavior and link it to an epidemic model of influenza that includes vaccination dynamics . We assume that the decision of each individual is based upon self-interest such that s/he wishes to avoid catching influenza , preferably without having to be vaccinated . Since protective immunity against influenza lasts less than one year [8] , individuals must decide every year whether or not to participate in a voluntary vaccination program . Individuals who get vaccinated protect themselves from infection , but if they do not get vaccinated they may still avoid infection if sufficient numbers of their peers get vaccinated ( i . e . , through herd immunity ) . This poses a yearly dilemma for the self-interested individual of whether vaccination is necessary . We model each individual's strategy for making yearly vaccination decisions as an adaptive process of trial and error . We track both individual-level decisions and population-level variables ( yearly vaccine coverage level and influenza prevalence; where prevalence is defined as the proportion of the population that is infected ) . We use our model to address the following question: can influenza epidemics be prevented by voluntary vaccination ? Our individual-level adaptive decision-making model is inspired by Minority Game methodology . A Minority Game models how noncommunicating selfish individuals reach a collective behavior with respect to a common dilemma under adaptation of each one's expectations . In the past decade , Minority Games [9] have been used to model inductive reasoning systems [10] and financial markets [11] . Our constructed model consists of a population of N individuals acting in their own self-interest who do not communicate their vaccination decisions to each other . Every year , these individuals independently decide whether or not to get vaccinated against influenza using a risk-free , highly effective vaccine [12] . We assumed that the vaccine presents no real risk and that individuals do not perceive any risk from vaccination . Individuals in the model are characterized by two biological attributes ( adaptability and memory ) that they use when making vaccination decisions . Individuals can adapt their vaccination behavior for the current season on the basis of their memories of the consequences of their past vaccination decisions: i . e . , they use cognition to make decisions . We couple our individual-level model of adaptive decision-making with a model of influenza vaccination dynamics . Our coupled models show the effect of individual-level vaccination decisions on influenza epidemiology and , conversely , the effect of influenza epidemiology on individual-level vaccination decisions . We first use our model to assess whether vaccination programs without incentives could achieve the critical coverage levels necessary to control influenza epidemics . We then assess the potential epidemiological impact of two public heath programs that use incentives to encourage vaccination . There are two major classes of incentive-based public health programs that can be investigated with our coupled models . The first class uses incentives to correlate vaccination decisions for the same individual over many influenza seasons . The second class uses incentives to correlate vaccination decisions amongst individuals in the population in one influenza season . Many additional incentive-based vaccination programs can be formulated by combining the defining characteristics of these two classes . The first public health program that we investigate is an example of the first class of incentive-based programs . This program offers free vaccination for y number of years to an individual who pays for vaccination in the first year . We assume that the individual gets vaccinated each year during the y years of free vaccination , but that s/he also evaluates the necessity of vaccination every year . At the end of y years , each individual in the program then uses their evaluations to decide whether or not to re-enroll in the program . If they choose to re-enroll , they pay for vaccination that season ( i . e . , season y + 1 ) and receive free vaccinations for a further y years . The second public health program that we analyze is an example of the second class of incentive-based programs . This program vaccinates a family for free if the head of the family pays for her/his own vaccination . We assume that the head of the family decides every year whether to re-enroll in the program depending upon how many of her/his family members were infected in the previous season .
We found that influenza epidemics could not be prevented in most seasons if vaccination was voluntary and no incentives were offered ( Figure 1A ) . This result was a consequence of individuals making vaccination decisions each year on the basis of their past experiences . When epidemics occurred , some individuals became infected; this increased the probability that they would get vaccinated in the next influenza season . Thus , the vaccination coverage gradually approached the critical value necessary for prevention ( Figure 1A ) . Eventually , the coverage slightly exceeded the critical coverage level due to the stochastic nature of the individual-level adaptive decision-making process . At this point , an influenza epidemic did not occur; notably , this happened rarely ( approximately once every 35 years; see Figure 1A ) . In the following season , many individuals decided that they did not need to get vaccinated , as an epidemic had not occurred in the previous season; thus vaccination coverage abruptly decreased and a severe epidemic ensued ( Figure 1A ) . The vaccination coverage then repeated a similar cyclic dynamic ( Figure 1A ) . If the initial vaccination coverage was larger than the critical coverage , the coverage dropped to a level below the critical coverage within a few years ( unpublished data ) ; vaccination coverage then followed the same cyclic dynamic as shown in Figure 1A . Since vaccination coverage determines the severity of an influenza epidemic , our results ( as shown in Figure 1A ) revealed that cyclic dynamics of influenza epidemics could simply be caused by individual-level adaptive decision-making . The dynamics of each individual's probability of getting vaccinated each season is more complex than coverage and prevalence dynamics ( Figure 1B ) . Figure 1C shows the distribution containing each of the N individuals' probability of getting vaccinated in one season; two distributions are shown . The first distribution ( black data ) is obtained from a season when an epidemic does not occur . In this season , the N individuals segregate into two groups as has been shown for other inductive reasoning games [13] . Individuals in one group are very likely to get vaccinated whilst individuals in the other group are unlikely to get vaccinated; few are undecided . This segregated distribution results over the course of the years when the coverage is close to but below the critical vaccination coverage . During these years , both vaccination and nonvaccination behaviors are reinforcing with the small exception of a few nonvaccinating individuals who get infected . These infected individuals then begin to get vaccinated and thus increase the coverage closer toward the critical vaccination coverage . The second distribution ( blue data in Figure 1 ) is obtained in successive seasons when severe epidemics occur . In these seasons , the distribution of the vaccination probabilities remains segregated into two groups . However , individuals who were very likely to get vaccinated previously have decreased their vaccination probability ( Figure 1C ) , causing severe epidemics . The distribution shown by the blue data in Figure 1 slowly tended towards the distribution shown by the black data as epidemics decreased in severity . When the critical coverage level is exceeded , the distribution repeats a similar cyclic dynamic . This cyclic dynamic occurred in a homogenous population where every individual had the same memory parameter s and adaptability parameter ɛ . We found that similar cyclic dynamics occur in heterogeneous populations where memory and adaptability are normally distributed , but bounded between 0 and 1 ( unpublished data ) . We note that , using a population-level model with a deductive reasoning game , Reluga et al . [14] have also recently shown that cyclic dynamics in vaccine coverage can occur due to heterogeneity in risk perception . Many individuals are likely to enroll in incentive-based vaccination programs in response to a major epidemic . However , the epidemiological impact of these programs can be complex . We analyzed the potential impact of a commitment-incentive program that offers free vaccination for y years if the individual pays for vaccination in the first year ( Figure 2 ) . A three-year program ( red data ) caused substantially less severe , but more frequent , epidemics than a program without incentives ( black data ) ( Figure 2A ) . In contrast , a fifteen-year program ( green data ) caused more frequent severe epidemics than a program without incentives . Our contrasting results are a consequence of the relationship between the length of the commitment to the program and the time scale of the memory parameter ( s; s = 0 . 7 determines a half life of 1 . 9 years ) . Programs that require only a short-term commitment ( e . g . , y = 3 ) have a high turnover of participants and a time scale comparable to that of the memory parameter . Participants who leave this program become reinfected and therefore quickly re-enroll in the program; this process results in only small frequent epidemics . Programs that require long-term commitment ( e . g . , y = 15 ) have a relatively low turnover of participants and a time scale much longer than that of the memory parameter . Long-term commitment programs prevent epidemics for many years . Thus , at the end of the commitment many individuals do not re-enroll in the program because an epidemic has not occurred for many years . Therefore , vaccination coverage drops and a severe epidemic occurs; severe epidemics occur approximately every fifteen years if y = 15 . To systematically assess the effect of memory , adaptability , and length of commitment on the success of vaccination programs , we conducted an uncertainty analysis for: ( i ) programs without incentives , ( ii ) short-term commitment ( e . g . , y = 3 ) programs , and ( iii ) long-term commitment ( e . g . , y = 15 ) programs ( see Figure 3 ) . We found that the magnitude of epidemic amelioration is determined by the length of commitment to the program , the individuals' adaptability , and their memory . When individuals are very adaptable and have long-term memories , commitment-incentive programs can be very effective in controlling influenza epidemics . As well as a commitment-incentive vaccination program , we also investigated the potential epidemiological impact of a family-incentive program . This program vaccinates a family for free if the head of the family pays for her/his own vaccination . Vaccination coverage dynamics for the family-incentive program appeared fairly similar to the coverage dynamics for the program that does not provide incentives ( Figure 2B ) . However , surprisingly , the family-incentive program increased the frequency of severe epidemics . This result was found because epidemic severity and frequency are a function of the number of individuals who independently decide whether or not to get vaccinated . In the vaccination program without incentives , each member of the population is a decision-maker and decides independently whether to get vaccinated or not . In the family-incentive program , only one member of each family is allowed to make the decision . Therefore , the family-incentive program reduces the number of independent decision-makers from the total number of individuals to the total number of families . Stochastic variation in the coverage ( and hence frequency of severe epidemics ) increases as the number of independent decision-makers decreases . Thus , the family-incentive program increased the frequency of severe epidemics .
The critical vaccination coverage level necessary to eradicate influenza epidemics and pandemics has been calculated by analyzing influenza transmission models [1–7] . However , none of these studies have shown whether it is actually possible to reach the critical coverage level . By coupling a novel individual-level model of human cognition and behavior with an epidemic model , we have determined , for the first time , that this critical level is unlikely to be reached if vaccination is voluntary and no incentives are offered; for mathematical justification see [15] . Our modeling has shown that incentive-based vaccination programs are necessary to control influenza epidemics , but that some of these programs may be detrimental . Hence , incentive-based programs need to be carefully evaluated before they are implemented . Surprisingly , we have found that the epidemiological impact of influenza vaccination programs will depend upon the biological characteristics of individuals as well as the specific incentives that are offered . Influenza evolution and dynamics are driven by genetic changes that can alter strain transmissibility and/or virulence; therefore , influenza epidemics can show seasonal variation in severity . The severity of an epidemic can be defined in terms of the basic reproduction number ( R0 ) ; where R0 represents the average number of secondary cases caused by one infectious case at the beginning of an epidemic . Changes in strain transmissibility and/or virulence may lead to an increase ( or a decrease ) in the value of R0; thus , the value of R0 may show seasonal variation . However , in our analyses we used a constant value of R0 , because we wanted to isolate the impact of individual-level vaccination decisions on influenza dynamics . It is notable that even with a constant R0 , by including individual-level adaptive decision-making , our modeling was able to reproduce two essential characteristics of influenza dynamics and evolution: ( i ) annual variation in epidemic severity , and ( ii ) sporadic occurrence of severe epidemics . Therefore , our results suggest that individual-level adaptive decision-making may be an important ( previously overlooked ) causal factor in driving influenza epidemiology . A pandemic influenza strain will not necessarily have a substantially higher R0 than an interpandemic ( i . e . , seasonal ) strain . In our analyses of interpandemic strains , we used an R0 value of 2 . 5 ( which requires a critical vaccination coverage of 0 . 6 ) ; a similar value for R0 has been quoted and attributed to pandemic strains [5 , 16] . Specifically , in Mills et al . [5] , they calculate that the R0 values for the 1918 pandemic strains were between 2 and 3 . However , it is possible that a pandemic strain may have a substantially higher R0 than an interpandemic strain . Therefore , we also investigated the impact of the sporadic introduction of a pandemic strain with an R0 = 10; this pandemic strain has a critical vaccination coverage level of 0 . 9 . Apart from an occasional substantial increase in coverage in the year after a pandemic , the qualitative behavior of the coverage dynamics are similar to the dynamics for interpandemic influenza ( unpublished data ) . The epidemiological impact of both the commitment-incentive and the family-incentive program on pandemics are also similar to the impact observed previously for interpandemic influenza . Therefore , we conclude that our results can be used to understand , and to predict , the potential impact of vaccination programs for controlling pandemic , as well as interpandemic , influenza . Two previous studies [17 , 18] , based upon a game-theoretic approach using voluntary vaccination programs ( without incentives ) , have shown that it would only be possible to eradicate a vaccine-preventable disease if a risk-free vaccine was used . In contrast , we found that influenza epidemics are unlikely to be prevented by using voluntary vaccination with a risk-free vaccine . The reason that our results are in direct contrast to the previous two theoretical studies are that different pathogens are investigated . The earlier study by Bauch et al . concentrated on smallpox [18] , and in the second study Bauch and Earn [17] analyzed childhood diseases . For both smallpox and childhood diseases , it is necessary to be vaccinated only once [17 , 18]; thus , Bauch and colleagues did not model memory nor individual adaptability . In contrast to these earlier studies , we have modeled influenza , which needs a yearly vaccination . We have assumed that individuals make their decisions both on the outcome of their own previous vaccination decisions , as well as on the basis of the previous seasons' level of herd immunity . Therefore , our model includes greater biological complexity than the previous models , as , when modeling annual influenza vaccination decisions , it is necessary to model a memory effect and to incorporate the possibility of changing behavior ( i . e . , adaptability ) . Hence we found contrasting results to the previous two studies [17 , 18] . The purpose of our analyses was to evaluate the role of memory and adaptation on vaccination decision-making , and also the impact of vaccination decisions on influenza epidemiology . We have presented results for a homogenous population in the memory and adaptability parameters . Similar qualitative dynamics were found for the case where the population was heterogeneous in both memory and adaptability . Many other factors may also influence individuals in making their vaccination decisions [19] . However , memory and adaptation are principle biological attributes of individuals; consequently , including them in models of recurring voluntary vaccination is essential . Our model describes a large population of individuals . We account only for epidemics and we do not consider outbreaks; outbreaks become decreasingly important as the population size N increases . The two central assumptions of our model are that individuals act in their own self-interest and do not communicate their vaccination decisions to each other . If these assumptions are not met , then other outcomes are possible: for example , the public may choose to continue vaccinating even if vaccination does not appear worthwhile for them . This type of behavior may be able to prevent influenza epidemics occurring . However , we stress that even a population of individuals acting in the interest of their own families would not be able to prevent influenza epidemics . Although we do not model the impact of treatment in controlling influenza epidemics , the effects of treatment can be implicitly accounted for in our individual-level model by decreasing the critical vaccination coverage level . We also do not model the economics of vaccination programs . Such analyses could be done using our model in order to assess the most cost-effective vaccination program . In the United States , demand for influenza vaccines is generally met and no major shortages occur . In recent years , vaccination coverage ( based upon voluntary vaccination ) has steadily increased [20 , 21] . One of the national health objectives of the US is to further increase the coverage [20 , 21]; currently , the coverage is below the Healthy People 2010 objectives [20] . Here we have shown computationally , for the first time to the best of our knowledge , that it is unlikely that influenza epidemics will be prevented if vaccination is voluntary and no incentives are offered . We have found that incentive-based vaccination programs will be necessary for controlling influenza . We have also shown that these programs can have surprising effects and sometimes may make epidemics worse . We recommend that public health vaccination programs should be carefully evaluated before they are implemented . By modeling human cognition and behavior , we have shown that the impact of vaccination programs will depend upon both the biological characteristics of individuals as well as the specific incentives that are offered . Surprisingly , we found that individuals' memories and their flexibility in adaptive decision-making can be extremely important factors in determining the success of influenza vaccination programs .
In every influenza season , each individual decides whether or not to get vaccinated , independently from each other . We assume that vaccination completely protects an individual from infection . The probability that individual i chooses to be vaccinated in season n is w ( i ) n . Individual i is assigned a vaccination experience variable V ( i ) n , the value of which changes every year and depends upon whether or not: the individual chose to be vaccinated , they became infected , and an epidemic occurred in the previous season ( Figure 4A ) . V ( i ) n increases each time the individual perceives that there was , or would have been , a benefit to vaccination because ( a ) the individual got vaccinated and there was an epidemic , or ( b ) the individual did not get vaccinated and then became infected ( Figure 4A ) . We model the effect of memory by using a parameter s to discount the previous seasons' vaccination outcome with respect to the outcome of the present season ( 0 < s < 1 ) [22] . Specifically , V ( i ) n+1 = sV ( i ) n + 1 if individual i believes s/he did , or would have , benefited from vaccination in season n . Otherwise , if individual i believes that vaccination was unnecessary in season n ( regardless of whether s/he got vaccinated or not ) , V ( i ) n+1 = sV ( i ) n . We normalize V ( i ) n+1 by ( sn+1 − 1 ) / ( s − 1 ) because this factor is the maximum possible value for V ( i ) n+1 if individual i would have benefited from vaccination in all n influenza seasons . The domain for the V ( i ) n is 0 ≤ V ( i ) < 1/ ( 1 − s ) . Note that if individual i has perfect memory ( i . e . , s = 1 ) , the vaccination experience variable represents the total number of years that this individual would have benefited from being vaccinated divided by the total number of years that vaccination was available . We assume that individuals are adaptable in their decision-making as to whether to be vaccinated or not , and we use a parameter ɛ to describe an individuals' adaptability based upon their past experiences with vaccination ( 0 < ɛ < 1 ) . Thus , the probability that an individual chooses to get vaccinated in the next influenza season is given by w ( i ) n +1 = ( 1 − ɛ ) w ( i ) n + ɛ V ( i ) n+1/[ ( sn+1 − 1 ) / ( s − 1 ) ] . This expression shows that memory s and adaptability ɛ are not interchangeable parameters . An individual may have perfect memory of their vaccination experiences , characterized by s = 1 , but may not use this memory if they have a small adaptability parameter ɛ . In our model , the probability of an unvaccinated individual acquiring influenza q ( p ) decreases linearly as the coverage p increases; Figure 4B . This function is a good approximation of the relationship found for the Susceptible–Infected–Recovered ( SIR ) model as well as the Susceptible–Exposed–Infectious–Recovered ( SEIR ) model that could be used as within-season population-level models; see below . When the vaccination coverage is greater than or equal to the critical vaccination coverage ( i . e . , p ≥ pc ) the probability of an unvaccinated individual getting infected is defined to be zero . At the end of each season , every individual evaluates their vaccination decisions based upon whether vaccination had been necessary to avoid infection . They then modify their probability to get vaccinated the next season to w ( i ) n+1 . Figure 4A shows a diagram of the evaluation tree . Individuals start their first season with no prior experience in decision-making as to whether to be vaccinated or not . The initial condition assigns a random vaccination probability for the first season to every individual . Specifically , V ( i ) 0 = 0 and w ( i ) 0 is a uniformly random variable between 0 and 1 . Our initial conditions were chosen to reflect the fact that the initial public awareness of the benefits of the voluntary vaccination would not be high enough to prevent an epidemic , while at the individual-level the likelihood of vaccination could vary considerably . Our findings reported in Figure 1 are robust . Using the methodology presented in [15] , we found that there exists a considerable region in the parameter space for which our model yields coverage dynamics similar to the results in Figure 1 . The region is given by ɛ − s − 1 < ɛ ( 1 − s ) [1 − q ( 0 ) ( 1 − pc ) /pc] < ɛ − s + 1 and 0 < s ( 1 + ɛ − s ) + ɛ ( 1 − ɛ ) ( 1 − s ) ( 1 − 2s ) . Furthermore , for this parameter region , there exists only one attracting state for the dynamics of the coverage . Thus , our results are independent of the initial conditions . We investigated the potential epidemiological impact of two public health programs that we defined as a voluntary vaccination program coupled with an incentive . The first public health program that we investigated would offer free vaccination for y number of years if the individual paid for vaccination in the first year . We assume that during the y years of free vaccination the individual would continue to get vaccinated each year , but would also evaluate the necessity of influenza vaccination . At the end of their y years of free vaccination , every individual in the program then uses all their evaluations to decide whether to pay for vaccination that season ( i . e . , season y + 1 ) and further receive free vaccinations for a further y years . To model this public health program , the changes that are needed in the model are very few . Namely , if an individual gets vaccinated in year n , then w ( i ) n+r = 1 , with 0 < r < y + 1; thus s/he will also get vaccinated for the next consecutive y years ( when the vaccine is provided for free ) . We also investigated the potential epidemiological impact of a public health program that would vaccinate a family for free if the head of the family paid for her/his own vaccination . This is different from the model without incentives and the one with the first public health program , as in those analyses we assumed that individuals independently decide whether or not to get vaccinated . We assume that the head of the family would make her/his choice on the basis of protecting her/his family against influenza in that particular season that free vaccination is offered . S/he would then modify her/his probability of getting vaccinated in the next season depending upon how many family members became infected . To model this vaccination program we considered a population of N individuals who are grouped into families with C members . The head of the family j updates her/his V ( j ) n value ( where j labels the family , j = 1 . . . N/C ) in the following way: ( a ) V ( i ) n+1 = sV ( i ) n + C if the head of the family had decided to have her/his family vaccinated and there was an epidemic that season; ( b ) V ( i ) n+1 = sV ( i ) n if there was no epidemic that season , regardless of whether or not the family was vaccinated against influenza; ( c ) V ( i ) n+1 = sV ( i ) n + k , if k members of the family were infected in a season where the head of the family had decided not to get her/his family vaccinated and there was an epidemic . We normalized the value of V ( i ) n+1 by a factor of C ( sn+1 − 1 ) / ( s − 1 ) that represents the maximum possible value of V ( i ) n+1 over n seasons . Therefore , the vaccination probabilities are updated as follows w ( i ) n+1 = ( 1 − ɛ ) w ( i ) n + ɛV ( i ) n+1/[C ( sn+1 − 1 ) / ( s − 1 ) ] . To calculate the probability of getting infected with influenza q given a specified vaccination coverage p during one influenza season , we make the following assumptions that are compatible with the vaccination model presented in the main text: a ) we ignore the inflow and outflow of individuals in the study population during a season ( i . e . , we ignore vital dynamics ) ; b ) individuals may get vaccinated against influenza only at the beginning of the influenza season; c ) the vaccine is risk-free and offers perfect protection against infection; d ) individuals who get infected and then recover remain immune to infection until the end of the season . As a result of the above assumptions , we choose to model the epidemic transmission during one season using a SIR model , without vital dynamics , that includes vaccination at the beginning of each influenza season . where S ( t ) , I ( t ) , R ( t ) , and V ( t ) represent the number of susceptible , infected , recovered , and vaccinated individuals , respectively . The total number of individuals N = S ( t ) + I ( t ) + R ( t ) + V ( t ) is constant . β represents the transmissibility in the mass-action term , and γ represents the recovery rate . The initial conditions for the equations are as follows . A fraction p of the population gets vaccinated against influenza leaving only ( 1 − p ) N susceptible individuals . Thus , at the start of the influenza season , S ( 0 ) = ( 1 − p ) N − 1 , I ( 0 ) = 1 , R ( 0 ) = 0 , and V ( 0 ) = pN . The probability of getting infected during an influenza season q ( p ) is given by , where T represents the duration of the influenza season . In Figure 5 we show an illustrative graph of q ( p ) . We note that the featured dependence is approximately piecewise linear . The discontinuity in derivative occurs at p = 1 − 1/N − γ/β ≡ πc which for large N becomes πc ≈ 1 − γ/β . We note that a SEIR model could also be used to model influenza transmission [23] . Using an SEIR model we obtained a dependency of q with p which is similar to that presented in Figure 5 ( unpublished data ) . Therefore , Figure 4B can be used to qualitatively model this dependency for both SIR and SEIR transmission models . | Currently , a major public health concern is the next influenza pandemic; yet it remains unclear how to control such a crisis . By using novel mathematical modeling techniques , here we predict the likely impact of voluntary vaccination programs on controlling influenza epidemics and pandemics . We construct an individual-level model of human cognition and behavior that includes two important biological characteristics: memory and adaptability/flexibility . In each influenza season , each individual in the modeled population decides , using memory and adaptability/flexibility , whether to be vaccinated or not . We combine our individual-level model with an epidemic model to predict the impact of voluntary vaccination programs . We found that severe influenza epidemics cannot be prevented unless vaccination programs offer incentives . Frequency of severe epidemics could be reduced if programs provide , as an incentive to be vaccinated , several years of free vaccines to individuals who pay for one year of vaccination . However , we found that a public health intervention program that focuses on vaccinating families is likely to increase the frequency of severe epidemics . Most importantly , we found that individuals' memories and adaptability/flexibility in decision-making are critical factors in determining the success of influenza vaccination programs . Our results are applicable both for the control of seasonal and pandemic influenza . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"none",
"infectious",
"diseases",
"public",
"health",
"and",
"epidemiology",
"computational",
"biology"
] | 2007 | Can Influenza Epidemics Be Prevented by Voluntary Vaccination? |
Subcellular compartmentation of the ubiquitous second messenger cAMP has been widely proposed as a mechanism to explain unique receptor-dependent functional responses . How exactly compartmentation is achieved , however , has remained a mystery for more than 40 years . In this study , we developed computational and mathematical models to represent a subcellular sarcomeric space in a cardiac myocyte with varying detail . We then used these models to predict the contributions of various mechanisms that establish subcellular cAMP microdomains . We used the models to test the hypothesis that phosphodiesterases act as functional barriers to diffusion , creating discrete cAMP signaling domains . We also used the models to predict the effect of a range of experimentally measured diffusion rates on cAMP compartmentation . Finally , we modeled the anatomical structures in a cardiac myocyte diad , to predict the effects of anatomical diffusion barriers on cAMP compartmentation . When we incorporated experimentally informed model parameters to reconstruct an in silico subcellular sarcomeric space with spatially distinct cAMP production sites linked to caveloar domains , the models predict that under realistic conditions phosphodiesterases alone were insufficient to generate significant cAMP gradients . This prediction persisted even when combined with slow cAMP diffusion . When we additionally considered the effects of anatomic barriers to diffusion that are expected in the cardiac myocyte dyadic space , cAMP compartmentation did occur , but only when diffusion was slow . Our model simulations suggest that additional mechanisms likely contribute to cAMP gradients occurring in submicroscopic domains . The difference between the physiological and pathological effects resulting from the production of cAMP may be a function of appropriate compartmentation of cAMP signaling . Therefore , understanding the contribution of factors that are responsible for coordinating the spatial and temporal distribution of cAMP at the subcellular level could be important for developing new strategies for the prevention or treatment of unfavorable responses associated with different disease states .
For nearly 40 years , subcellular compartmentation has been offered as an explanation for how cAMP , the ubiquitous and diffusible second messenger , can both regulate a multitude of cellular functions and elicit specific and selective responses . Despite widespread recognition of the importance of cAMP compartmentation in tightly controlling local signaling , exactly how compartmentation occurs is still poorly understood . The general definition of compartmentation in this context is when a gradient exists in the concentration of cAMP between two locations . As it relates to cell signaling , the concentration gradient is relevant when it affects the potential for cAMP to activate an effector , such as protein kinase A ( PKA ) , in one location but not another . A number of processes have been suggested to contribute to this phenomenon , but studies have offered conflicting data that differ in their interpretation and assessment of key players . Localized degradation by phosphodiesterases ( PDEs ) has been a prime focus of many studies attempting to understand the basis of cAMP compartmentation [1–5] . Phosphodiesterases are thought to contribute to the generation of cytosolic cAMP gradients either by acting as functional barriers to diffusion that result in lower levels of cAMP distal to its site of production or as sinks that deplete cAMP in localized areas . Evidence clearly demonstrates that PDE activity is an essential factor in cAMP compartmentation . This has been illustrated by employing a number of different experimental approaches , including Jurevcius and Fischmeister who used patch clamp electrophysiology to demonstrate that in frog ventricular myocytes , inhibition of PDE activity allows local stimulation of cAMP by β-adrenergic receptors to enhance distal Ca2+ channel activity [6] . On the other hand , Zaccolo et al . used a genetically encoded FRET-based biosensor to demonstrate that β adrenergic stimulation elicits a localized pattern of cAMP production in neonatal cardiac myocytes that is disrupted by inhibition of PDE activity [7] . However , the question of whether or not PDE activity alone is sufficient to explain the compartmentalized behavior of cAMP signaling is still debated . Computational modeling has proven to be a useful tool in investigating the relative contribution of PDEs to cAMP compartmentation [8] . All modeling studies support the idea that PDE activity is necessary for cAMP compartmentation . At least one study has predicted that it is theoretically possible for artificially high levels of PDE activity alone to explain compartmentation [9] . Other models using more realistic levels of PDE activity suggest that factors such as the shape of the cell and the rate of cAMP diffusion play critical roles in explaining the existence of cAMP gradients within a cell [10–21] . One way that cell shape may be a factor in compartmentalizing cAMP signaling is by affecting the surface-to-volume ratio . Studies using FRET-based biosensors in neurons have found cAMP levels to be higher in dendrites than cell bodies[22] . It was suggested that this could be due to the higher surface-to-volume ratio found in dendrites , resulting in greater membrane bound adenylyl cyclase activity and reduced cytosolic PDE activity . Subsequent modeling supported the feasibility of this hypothesis , without having to assume the involvement of other factors [18] . Feinstein et al . found that the surface-to-volume ratio of a cell can contribute to generation of cAMP gradients , but it was also necessary to assume that the movement of cAMP is slower than the rate of free diffusion [20] . Other models have been able to explain compartmentation independent of cell morphology , as long as it was assumed that cAMP diffusion is somehow restricted [10 , 11 , 13–15 , 17 , 23] . Although the potential effect that the surface-to-volume ratio of a cell has on cAMP compartmentation has been examined , the influence of the actual size and shape of subcellular compartments is less well understood . A major reason is that the physical nature of these microdomains is not well described . Previous modeling studies often circumvented this issue by using loosely defined membrane and cytosolic domains and treating the movement of cAMP between them as fluxes that do not require knowledge of the number , size , or location of these compartments . Previous experimental studies have shown that receptors associated with cholesterol rich lipid rafts , which include caveolae , can elicit cAMP responses that are distinctly different from those produced by extracaveolar receptors found outside of lipid rafts [16 , 24 , 25] . Lipid rafts are liquid-ordered domains of the membrane rich in cholesterol and sphingolipids . Caveolae are a specific subset of lipid rafts that contain caveolins , proteins involved in the formation of signaling complexes that include β1 and β2 adrenergic receptors ( βARs ) as well as adenylyl cyclase isoforms 5 and 6 ( AC5/6 ) [26–29] . In cardiac myocytes , activation of receptors associated with caveolar lipid rafts are involved in local cAMP production and PKA-dependent regulation of L-type Ca2+ channel function [24 , 28 , 29] . There is evidence that these types of compartmentalized cAMP responses also occur in the transverse tubules ( t-tubules ) of cardiac myocytes[30] . T-tubules are invaginations of the plasma membrane that come in close proximity to the junctional sarcoplasmic reticulum ( SR ) forming dyadic junctions [31] . Therefore , it is possible that the size , shape , and distribution of caveolae , especially those associated with the restricted space at cardiac dyadic junctions , may contribute to compartmentation of cAMP signaling in cardiac myocytes . The purpose of the present study was to apply novel computational approaches to predict whether PDE activity alone or in conjunction with restricted diffusion is sufficient to produce cAMP gradients in submicroscopic signaling domains .
Experimental studies in cardiac myocytes have shown that activation of βARs associated with caveolar regions of the plasma membrane produce unique compartmentalized cAMP responses [24 , 25 , 28 , 29 , 32] . Other studies have used computational approaches to investigate the effect that cell morphology has on the generation of cytosolic cAMP gradients [18 , 20 , 23] . However , the importance that the organization and structure of submicroscopic signaling domains has on creating compartmentalized cAMP responses has not been addressed . To investigate how cAMP-mediated responses are localized and prevent initiation of global responses , we developed an idealized , partial differential equation ( partialDE ) -based 2D continuum model of a cardiac myocyte subspace with spatially distinct cAMP microdomains to allow for simulation of cAMP compartmentation and diffusion . A schematic diagram of the longitudinal cross-section of an adult ventricular myocyte with dimensions of 100 μm x 20 μm [14] is shown in Fig 1A . This illustrates the repeating pattern of the sarcomeres , which are spaced 2 μm apart [33] . Fig 1B illustrates the 2D continuum model that we constructed to represent the subcellular sarcomeric space used in the simulations shown in Fig 2 . The model represents the intracellular space between adjacent t-tubules . The t-tubules are lined by caveolar domains spaced 100 nm apart [34 , 35] . Each unit is half the width of the cell ( 10 μm ) . Fig 1C shows a magnified section of the subsarcolemmal space consisting of a single caveolar domain ( blue ) , which is 0 . 1 μm x 0 . 01 μm [34 , 36 , 37] , and an adjacent 0 . 1 μm x 0 . 01 μm extracaveolar space ( green ) . β-adrenergic receptors ( βARs ) and adenylyl cyclases ( ACs ) ( the site of cAMP production ) were distributed equally among the 50 caveolar domains on each side of the sarcomeric space . Using this model we would expect to be able to readily track compartmentation as cAMP gradients . We define a “significant” gradient as one in which the concentration of cAMP drops by more than 15% of its value relative to the site of production . Furthermore , we identify gradients as being relevant to compartmentalized signaling if cAMP concentrations in one compartment reach levels likely to produce PKA activation ( >1 μM ) . Predictions from simulations using this 2D continuum partialDE model to simulate cAMP diffusion are shown as snapshots of cAMP concentration at different points in time across the microdomain space in Fig 2 . The time course of spatial changes in cAMP concentration resulting from activation of βARs is shown when cAMP was allowed to move at a rate approximating free-diffusion ( 300 μm2/s ) under conditions where no phosphodiesterases ( PDEs ) were present ( Fig 2A ) , where PDEs were localized to the caveolar domain at concentrations consistent with those reported experimentally [14 , 15] ( Fig 2B ) , and where PDE concentrations in the caveolar domain were increased 10-fold ( Fig 2C ) . Only miniscule gradients ( sub-nanomolar ) were observed during the 2 . 0 second simulations , even when the concentration of PDE was increased 10-fold . The prediction of the model led to no indicators of significant compartmentation , which we would have expected to observe as gradients of cAMP concentration within the subcellular sarcomeric space . Rather , the monochromatic color maps in Fig 2 at each time point indicates a uniform cAMP concentration . Several studies [10–17 , 20 , 23 , 38] have suggested that for compartmentation to occur , diffusion of cAMP must be substantially slower than the reported value of free diffusion in a dilute aqueous environment , which is 300 to 400 μm2/s [39 , 40] . Assuming that cAMP movement is affected by factors such as cytoplasmic viscosity and molecular crowding , the diffusion coefficient of molecules the size of cAMP has been estimated at 60 μm2/s [41] . However , slowing diffusion in our simulation was still insufficient to generate a spatial gradient , even when the amount of PDE activity was increased 10-fold above the levels believed to exist in cardiac myocytes ( Fig 2D ) . It has also been suggested that buffering of cAMP through its interactions with PKA can decrease the effective diffusion coefficient for cAMP even further , to values closer to 10 μm2/s [41] . Yet , even this marked reduction of cAMP diffusion rate in the simulation was insufficient to generate spatial gradients of cAMP ( Fig 2E ) . In the 2D continuum model , PDEs were contained within the thin caveloar domain ( i . e . all PDE is effectively along the plasma membrane ) . In order to more specifically address the question of whether PDEs can form a “functional barrier” to cAMP diffusion , we developed a 3D stochastic model of cAMP diffusion in a subcellular microdomain ( depicted in Fig 1D ) and implemented the model using MCell [42] . This approach allows for investigation of the contribution of spatial localization of microdomain specific signaling components ( e . g . , PDEs ) to compartmentation . Evidence exists that the localization of signaling complexes is important in producing compartmentalized responses [28 , 43] . Shown in Fig 3 are results from stochastic simulations of cAMP diffusion visualized using CellBlender ( mcell . org ) . Fig 1D illustrates the subcellular compartment model that is used in the 3D stochastic simulations described in Fig 3 through Fig 4 . The space ( 200 x 200 x 1000 nm ) surrounding a single caveolar domain containing 15 βARs is depicted in green in Fig 3A–3E . Activation of βARs generated 120 cAMP molecules/s . Once produced the cAMP molecules diffused freely in the microdomain at a rate of 300 μm2/s . PDE molecules were placed in a plane ( L* ) 100 nm from the inner surface of the plasma membrane ( red plane in Fig 3A–3E ) in order to determine if PDEs could act as a functional barrier to cAMP diffusion . Fig 3A illustrates the distribution of cAMP molecules throughout the microdomain at various time points when the functional barrier consists of 10 PDE molecules . This corresponds to a PDE concentration of 4 . 15 μM . The graph at the top right of Fig 3A illustrates the time course of cAMP accumulation under these conditions . The graph at the bottom right of Fig 3A plots the accumulated concentration of cAMP ( averaged over the first 1 to 10 second time interval along the length of the microdomain . The simulation demonstrates that 10 PDE molecules are not enough to serve as a functional barrier to cAMP diffusion and generate a discernible cAMP gradient . We then evaluated the effect of increasing the number of PDE molecules in the functional barrier by several orders of magnitude ( Fig 3B–3E ) . Only when the number of PDE molecules was increased above 10 , 000 ( Fig 3D and 3E ) did a cAMP gradient become visible . This is illustrated most clearly by the accumulated concentration map at the bottom of each panel . The results described above indicate that PDE activity alone is unlikely to produce significant cAMP gradients by acting as a functional barrier when cAMP was allowed to diffuse freely . We next tested if this was also the case when the rate of cAMP diffusion was decreased to 60 μm2/s , as shown in S3 Fig . This condition reflects the experimentally measured diffusion coefficient of cAMP like molecules that was determined by using fluorescein and the ϕ450 fluorophore , fluorescent molecules about the same size as cAMP that do not bind to PKA . In water , these molecules exhibit rates of free diffusion of ~300 μm2/s , but inside cardiac myocytes the diffusion coefficient decreases to ~60 μm2/s , attributable to collision with other macromolecules in the intracellular environment due to molecular crowding [44] . Despite the slower rate of diffusion , a functional barrier consisting of 10 PDE molecules was still not sufficient to produce a cAMP gradient ( S3 Fig ) . In the setting of slower diffusion , it was necessary to increase the number of PDE molecules to at least 1000 ( S3C–S3E Fig ) before a small gradient was visible . Slowing the rate of cAMP diffusion also increased the concentration of cAMP observed at all levels of PDE activity . We then repeated the simulations using a diffusion coefficient of 10 μm2/s ( Fig 4 ) . This reflects the further slowing of cAMP diffusion due to the effects PKA buffering as suggested experimentally [41] . Interestingly , under these conditions , there is evidence for cAMP compartmentation when the number of PDE molecules in the barrier is at least 100 , which corresponds to a concentration of 41 . 5 μM ( Fig 4B–4E ) . Even with a diffusion coefficient of 10 μm2/s , the diffusion length ( 2Dt ) of cAMP on relevant time scales ( 1–10 seconds ) is much larger than the length scale of the caveolar domain ( 0 . 01–0 . 2μm ) . This leads to a nearly uniform concentration of cAMP in planes parallel to the plasma membrane , and therefore cAMP dynamics can be well-predicted by a 1D continuum model that can be solved analytically to obtain an expression for the steady state concentration of cAMP along the microdomain ( see Methods and S1 Appendix ) . ( Fig 1E shows a schematic of the steady state distribution of cAMP . ) The concentration decreases linearly from a value of A+B at the cAMP production site ( z = 0 ) to a value of B at the location ( L* ) of the PDE molecules that form a barrier to cAMP . Beyond the PDE barrier ( from L* to L ) , the concentration of cAMP is constant at B . The orange curves plotted on the cumulative concentration maps in Figs 3 and 4 show the predictions of the 1D continuum model . In all cases , there is excellent agreement with the full 3D stochastic model . Note that the ratio R = A/ ( A+B ) provides a measure of compartmentation: R = 0 implies that there is a uniform distribution of cAMP throughout the cytosol ( no compartmentation ) , whereas R = 1 implies that all cAMP is trapped behind the functional barrier of PDEs ( complete compartmentation ) . The cAMP compartmentation ratio R for all cases shown in Figs 3 and 4 are provided in the figure captions . Thus far , our modeling results suggest that under physiologically relevant conditions , cAMP diffusion is not sufficiently restricted by the presence of PDE molecules to explain compartmentation . To further test the effects of the model parameter values on cAMP compartmentation , we used the 1D continuum model to perform a parameter sensitivity analysis . The black curves in Fig 5A shows the dependence of the cAMP compartmentation ratio R on PDE concentration for default parameters and a diffusion rate of 300 μm2/s as used in Fig 3 . The red , blue , green , and cyan lines illustrate the sensitivity of R to changes in D/ ( kf L* ) ( a parameter accounting for diffusion , the location of the PDE boundary , and the rate of cAMP association with PDE ) , kb ( rate of cAMP dissociation from PDE ) , kcat ( PDE catalysis rate ) , and JB ( cAMP production rate ) , respectively . Each parameter was adjusted by ±20% , and the results are plotted as a pair of colored lines for each parameter change . Fig 5B and 5C show the sensitivity to the parameters for cAMP diffusion constants of 60 and 10 μm2/s , respectively , as in S3 Fig and Fig 4 . For all cases , the cyan lines are almost entirely obscured by the black lines , indicating that the cAMP compartmentation is insensitive to changes in the cAMP production rate , JB . The compartmentation ratio R was most sensitive to D/ ( kf L* ) and kcat , however no change in R greater than 0 . 06 was observed . It is notable that the highest sensitivity of R ( in terms of change of the absolute magnitude of the ratio ) occurred in the ranges of PDE concentrations that are well above the physiologically relevant range . For PDE concentrations between 1 and 100 μM , R was insensitive to perturbations to all other parameters . The simulations conducted thus far used models incorporating an idealized view of the 3D space between t-tubules in a cardiac myocyte . None of them contained realistic subcellular structures that might , in an actual cell , act as physical barriers to diffusion of cAMP . The cytosolic compartment of a cardiac myocyte is structurally complex and the site of cAMP production in t-tubules likely occurs in close proximity to the junctional SR , forming dyadic clefts . Movement of cAMP out of this space is also likely to be affected by the presence of mitochondria , which make up approximately 30% of the cardiac myocytes volume and are tightly packed around these structures . To examine the possibility that cAMP compartmentation might be observed in this type of restricted space , we created a 3D continuum anatomical barrier model using cryo-TEM images of adult mouse cardiac myocytes , as described previously[45] ( Fig 6A ) . The dimensions of the resulting dyadic cleft were approximately 1040 x 765 x 415 nm . Production of cAMP was generated by 15 AC molecules situated in the center of the dyadic cleft . These were surrounded by a hollow sphere of PDEs 25 nm thick and 200 nm in diameter . The t-tubules , SR , and mitochondria were assumed to be impenetrable barriers to direct diffusion of cAMP throughout the cytosol . We then examined the effects of varying PDE activity , as well as the diffusion coefficient , on cAMP gradients ( Fig 7 ) . When cAMP was produced at the same rate as in the previous simulations ( 120 molecules/s ) , and it was allowed to diffuse at a rate of 200 μm2/s , which is consistent with previous estimates of the diffusion coefficient in intact cells [22 , 46 , 47] , no evidence of a gradient was observed when the number of PDE molecules surrounding the site of production was varied between 100 and 1000 ( Fig 7 ) . This behavior did not change when the diffusion coefficient for cAMP was reduced to 60 μm2/s . However , when the diffusion coefficient was further reduced to 10 μm2/s , a significant gradient was observed at all PDE concentrations .
In this study , we developed several modeling approaches that included discrete cAMP microdomains to carry out simulations to investigate the importance of parameters proposed to be critical to compartmentation , including location and concentration of PDE activity , rates of cAMP diffusion , and anatomical barriers to diffusion . There is a large body of literature demonstrating that the non-uniform distribution of PDE activity in different subcellular compartments plays a critical role in compartmentation of cAMP dependent responses [5] . This is achieved by various mechanisms , including interactions with A kinase anchoring proteins ( AKAPs ) , which create signaling complexes that include not only PKA , but also adenylyl cyclase [48] . In the present study , we used a 2-dimensional continuum model to evaluate the effect of localizing PDE activity near the site of cAMP production . The results demonstrate that concentration of PDE activity in this location did not prevent cAMP diffusion throughout the space representing the area between adjacent t-tubules . Furthermore , a 10-fold increase in PDE activity was still insufficient to restrict cAMP diffusion , rather it reduced cAMP levels throughout the entire compartment . In previous modeling studies using similar levels of PDE activity , cAMP gradients could be generated if was assumed that the movement of cAMP was restricted by some mechanism [10 , 14 , 15 , 20 , 23 , 49] . However , we found that reducing the cAMP free diffusion coefficient to levels similar to those recently described in intact cardiac myocytes [41] , did not promote compartmentation . PDEs were also not enough to prevent the experimentally observed higher concentrations of cytoplasmic cAMP from immediately diffusing into the caveloar spaces , where biosensors have suggested 10-fold lower cAMP concentrations under basal conditions ( S2 Fig ) [14 , 15] . An alternative explanation for how PDE activity creates compartmentalized cAMP responses is based on the common supposition that PDEs act as functional barriers preventing cAMP diffusion [1 , 5 , 50] . To test this hypothesis , we developed a 3D stochastic model of cAMP diffusion . When it was assumed that cAMP moved at rates equal to free diffusion ( 300 μm2/s ) , cAMP gradients could only be observed when the number of PDE molecules in the barrier was unrealistically high ( Fig 3 ) . Furthermore , levels of PDE activity sufficient to produce a gradient resulted in overall cAMP levels that are well below those required for activating cAMP . Decreasing the diffusion coefficient for cAMP reduced the level of PDE activity necessary to produce gradients . However , even with a diffusion coefficient of 10 μm2/s , it was still necessary to use artificially high levels of PDE activity , and the overall level of cAMP was still well below that necessary to activate any downstream signaling . We also implemented a one-dimensional functional barrier model , which was possible because the diffusion length for relevant reaction ( cAMP degradation by PDE ) timescales is much larger than the length scale of the periodic structure of the caveolar-extracaveloar compartments . The one-dimensional functional barrier model allowed for derivation of an analytical expression for the steady-state spatial distribution of cAMP , which we used to identify the dependence of cAMP distribution on model parameters including diffusion coefficients , position of the functional barrier , concentration of PDEs and reaction kinetics . Beyond showing that there is very little compartmentation of cAMP under physiological conditions , the analytical solution indicates that the model results show very little sensitivity to variations in model parameters , an indication of robustness of the models . The final set of simulations examined the role that physical barriers to diffusion might play in the generation of cAMP gradients . A 3D continuum model was implemented , which included subcellular structures , which acted to impede the diffusion of cAMP . It has been postulated that different “compartments” of cAMP can be carved out by creating an area of high AC surrounded by PDEs to form a barrier to prevent the cAMP produced by the AC from affecting the rest of the cell . Although the TEM images of cardiac myocytes do not show anything that would allow such a barrier to exist [51 , 52] , this hypothesis is prominent in the literature [1 , 5 , 50] , and so it is valuable to test its conceptual validity . It is possible that AKAPs might bind a greater proportion of PDEs at the edges of the cleft surrounding the AC molecules , but to date no study has suggested this type of cellular localization . If the PDE activity near the site of production exists in restricted anatomically bounded clefts , this may explain how cAMP levels near the site of production are kept low under basal conditions , preventing activation of PKA by much higher cAMP levels found throughout the rest of the cell [14 , 15] . The present study demonstrates that gradients consistent with those expected to result in compartmentalized responses can be produced , but only in anatomically restricted spaces , and the dimensions of those spaces are below the resolution limit of light microscopy . Therefore , one would not expect to be able to directly visualize these compartmentalized responses with techniques currently available . However , results obtained using FRET based biosensors together with the targeted application of agonists using scanning ion conductance microscopy have shown that activation of beta2-receptors produces evidence of cAMP responses localized specifically to t-tubules in adult ventricular myocytes , and that these cAMP responses do not propagate throughout the cell [30] . This is consistent with our modeling results demonstrating that cAMP production occurring in dyadic clefts along the tubules are compartmentalized . If we assume that PDE concentration of an average cardiac myocyte is ~0 . 1μM [53 , 54] and the volume is 31 , 400 μm3 , this means that there are ~1 . 3 x 106 PDE molecules per cell . If we further assume that there are approximately 13 , 000 dyadic clefts ( [55 , 56] , 10 , 000–50 , 000 ) per myocyte , and all PDE activity in the cell is concentrated in these clefts , this would mean that there are 100 PDE molecules per cleft . At this concentration , we found no evidence of a cAMP gradient across the PDE barrier in the anatomical model when diffusion was set at 200 μm2/s . Even if we assumed that the number of PDE per cleft was 10 fold higher , this did not affect our ability to detect a gradient . However , reducing the cAMP diffusion made a significant difference . With a diffusion coefficient of 10 μm2/s , a cAMP gradient was observed at all PDE concentrations tested . The results of these simulations support the conclusion that PDE activity alone is not sufficient to explain compartmentation , but if diffusion of cAMP is limited by factors such as molecular crowding , PKA buffering , and anatomical barriers combined , then compartmentation may occur . The diffusion coefficient of cAMP was determined by using fluorescein and the ϕ450 fluorophore , fluorescent molecules about the same size as cAMP that do not bind to PKA . In water , these molecules exhibit rates of free diffusion of ~300 μm2/s , but inside cardiac myocytes the diffusion coefficient decreases to ~60 μm2/s . This is consistent with the 4 to 5 fold decrease in mobility typically seen with molecules this size , and it has been attributed primarily to collision with other macromolecules in the intracellular environment due to molecular crowding . It turns out that cytoplasmic viscosity is only believed to be a minor player [44] . PKA can be found in both membrane and soluble cellular fractions of most cells . Our recent data suggest that PKA is targeted specifically to the mitochondrial outer membrane by A kinase anchoring proteins ( AKAPs ) [41] . Our future studies will be aimed at determining the quantitative effects of this anchoring on limiting the diffusion of cAMP . It is worth noting that this work does not preclude gradients of cAMP across entire cells or non-steady state gradients . Several studies of cell motility in non-cardiac cells have shown that a cAMP gradient can exist across the cell [57] . Also , several neuronal studies have pointed to cAMP gradients as a major feature in the turning behavior of neuronal growth cones [58 , 59] . In these cases , the distances under consideration are significantly larger . Also , these systems have different organization of relevant enzymes; for example , one study showed by TEM imaging that significant clusters of AC molecules localize to the synapse in rat neurons [60] . However , these computational experiments show that having 10 , 000 steep gradients around each cleft or each caveolae is infeasible and suggest that another explanation for the observed compartmentalized nature of PKA activity must be considered . In this study , we focused on the example of the dyadic cleft as a restricted space that would be expected to affect the generation of cAMP gradients and compartmentalized responses . Restricted spaces created by other means would be expected to have the same effect . For example , cultured neonatal cardiac myocytes may not have dyadic clefts , but they are flatter , which together with the tight packing of mitochondria beneath the plasma membrane may be another way of creating restricted spaces that contribute to compartmentation . It is also likely that factors yet to be identified contribute to compartmentalized responses in cardiac myocytes as well as other cell types .
We constructed a 2 μm by 10 μm two-dimensional finite difference model representing the sarcomeric space between adjacent t-tubules of an adult ventricular myocyte ( see Fig 1 ) . Cytosolic domains associated with caveolae found in the plasma membrane of the t-tubules were modeled as 0 . 1 μm x 0 . 01 μm spaces . These caveolar domains were flanked on each side by 0 . 1 μm extracaveolar spaces . βARs and AC5/6 were placed in the plasma membrane associated with caveolar domains . All the simulations were encoded in C and run on 48-Core AMD Opteron Processors . The implicit numerical method was used to integrate Eqs 1 & 2 . All parameters used in the model can be found in Iancu-Harvey model [61 , 62] . The time step ( ∆t ) was set to 0 . 001 s . Numerical results were visualized using MATLAB R2014a by The Math Works , Inc . The concentration of cAMP in each compartment was calculated using the following equations: G-protein activation module: RGS= ( Rβ1free×GSfree ) GSfree+KC ( 1 ) LRβ1=Liso× ( Rβ1free−RGS ) Liso+KL ( 2 ) LRGS=LRβ1× ( GSfree−RGS ) ( GSfree−RGS ) + ( KC×KHKL ) +Liso×RGSLiso+KH ( 3 ) Rβ1Total=Rβ1free+LRβ1+LRGS+RGS ( 4 ) ∂GSαGTP∂t=LRGS×kact2+RGS×kact1−GSαGTP×khydr ( 5 ) ∂GSβγ∂t=LRGS×kact2+RGS×kact1−GSαGDP×GSβγ×kreas ( 6 ) ∂GSαGDP∂t=GSαGTP×khydr−GSαGDP×GSβγ×kreas ( 7 ) GTotal=GSfree+GSαGTP+GSαGDP ( 8 ) cAMP produced by AC5/6: kAC5/6= ( 0 . 7+3 . 8234×GSαGTP0 . 97870 . 1986+GSαGTP0 . 9787 ) ×MWAC5/660×10−3 ( 9 ) ∂cAMPAC5/6∂t= ( kAC5/6AC5/6AF5/6 ) ATPK_mATP+ATP ( 10 ) cAMP degraded by PDEs: ∂cAMPPDEx∂t= ( kPDEx×PDEx ) ×cAMPKm_PDEx+cAMP ( 11 ) The general formulation used for each PDE isoform ( PDEx ) . cAMP dynamics: ∂cAMP ( x , z , t ) ∂t=∂cAMPAC5/6∂t- ( ∂cAMPPDE2∂t+∂cAMPPDE3∂t+∂cAMPPDE4∂t ) +D∂2cAMP ( x , z , t ) ∂x2+D∂2cAMP ( x , z , t ) ∂z2 ( 12 ) cAMP dynamics: ∂cAMP ( x , z , t ) ∂t=D∂2cAMP ( x , z , t ) ∂x2+D∂2cAMP ( x , z , t ) ∂z2 ( 13 ) ∂cAMP∂x|x=0=0 , ∂cAMP∂x|x=WL=0 , ∂cAMP∂z|z=0=0 , ∂cAMP∂z|z=LL=0 where D is diffusion coefficient 300 μm2/s ( Fig 2A–2C ) , 60 μm2/s ( Fig 2D ) and 10 μm2/s ( Fig 2E ) , and WL = 2 μm and LL = 10 μm . Definitions and initial values for model parameters were based on experimental data as described in [61 , 62] and shown in Tables 1 and 2 . We also constructed a 3-dimensional stochastic model of cAMP diffusion that was implemented in MCell and visualized using CellBlender ( mcell . org ) . The model consisted of a single caveolar domain ( 100 x 100 nm ) flanked by extra-caveolar space for a total of 200 x 200 x 1000 nm . We define the z direction as orthogonal to the membrane . The boundary conditions were assumed to be no flux boundaries . The caveolar domain contained 15 β1ARs , which generated cAMP at 120 molecules/s . βARs and AC5/6 were placed in the plasma membrane associated with caveolar domains . cAMP freely diffused in space . MCell tracks diffusion in radial coordinates by dividing an octant of a sphere into 16384 directions ( dphi ≈ 0 . 005 degrees ) . PDE molecules were placed in the plane z = L* = 100 nm as functional barriers . The Interaction radius is set to default [63–65] . The time step ( Δt ) was set to 5 . 0 x 10−9 s so that pb=k∙σ2Naπ∙ΔtD<1 , where k is binding rate , σ is surface grid density , Na is the Avogadro constant , and D is diffusion [64] . σ is set according to the following table ( Table 3 ) : The cAMP-PDE reaction was set to cAMP+PDE→Kf←KbPDEcAMP→KcatPDE ( 14 ) where Kf = 1 . 2 x 107 M-1 s-1 , Kb = 58 . 82 s-1 , and Kcat = 14 . 70 s-1 [66] . Even with the smallest cAMP diffusion constant used in this study , the diffusion length for relevant time scales is much larger than the length scale of the caveolar/extracaveolar anatomical microstructure . Therefore , the distribution of cAMP in planes with fixed z are approximately uniform , and the distribution ( effective concentration ) of cAMP as a function of z can be approximated by a 1-dimensional continuum model . ( See S1 Appendix for details ) . This model can be solved to obtain an expression for the steady state distribution of cAMP cAMP ( z ) ={ ( A+B ) −JBDz , 0<z<L*B , L*<z<L , ( 15 ) where A=JBDL*andB=JBKfKcat+KbL* PDE0Kcat-JB , Unless otherwise specified , the rate constants Kf , Kb , and Kcat are as defined above , the flux of cAMP into the domain is JB = 4 . 982 μm μM s-1 , and the location of the functional barrier is L* = 100 nm . PDEtot is the total concentration of bound and unbound PDE and is taken to be 4 . 1514 x 10n μM for n = 0 , 1 , 2 , 3 , and 4 , where the concentration is averaged over the region from the plasma membrane at z = 0 to the PDE barrier at z = L* . ( Note that the product of PDEtot and the cross-sectional area of the microdomain and L* is the total number of PDE molecules . ) For the models that included a physiological geometry , this geometry was developed from cryo-TEM images of adult mouse cardiac myocytes , as described previously [45] ( see Fig 7 ) . Briefly , a tetrahedral surface mesh was imported into Blender for finite element simulations and to smooth the sharp edges from segmentation of the TEM images ( using Blamer ) that would impede numerical modeling and lead to artifacts . BLAMer is a plug-in for the Open Source Blender visualization environment ( http://www . blender . org ) that provides an interactive interface to the GAMer ( Geometry-preserving Adaptive Mesher ) tool ( http://nbcr . ucsd . edu/ ? page_id=1131 ) from the FEtk ( Finite Element ToolKit ) software package ( http://nbcr . ucsd . edu/ ? page_id=495 ) maintained and distributed by the NIH-supported National Biomedical Computation Resource . GAMer produces high-quality simplex meshes of surfaces and volumes and was used via BLAMer by Hake et al . [45] to mesh the myocyte dyadic cleft anatomy from 3D electron tomographic data . This mesh was resegmented and imported into Virtual Cell . This mesh included two t-tubules surrounded by SR as well as two mitochondria . | Subcellular compartmentation of the ubiquitous second messenger cAMP has been widely proposed as a mechanism to explain how this one signaling molecule produces unique receptor-dependent functional responses . But , how exactly compartmentation occurs , is unknown . This is because there has been no way to measure the regulation and movement of cAMP in cells with intact subcellular structures . In this study , we applied novel computational approaches to predict whether PDE activity alone or in conjunction with restricted diffusion is sufficient to produce cAMP gradients in submicroscopic signaling domains . We also used the models to test the effect of a range of experimentally measured diffusion rates on cAMP compartmentation . Our simulations suggest that PDE activity alone is not sufficient to explain compartmentation , but if diffusion of cAMP is limited by potential factors such as molecular crowding , PKA buffering , and anatomical barriers , then compartmentation is predicted to occur . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"medicine",
"and",
"health",
"sciences",
"muscle",
"tissue",
"enzymes",
"enzymology",
"simulation",
"and",
"modeling",
"camp",
"signaling",
"cascade",
"lyases",
"cellular",
"structures",
"and",
"organelles",
"research",
"and",
"analysis",
"methods",
"animal",
"cells",
"proteins",
"coated",
"pits",
"biological",
"tissue",
"chemistry",
"cell",
"membranes",
"muscle",
"cells",
"phosphodiesterases",
"physics",
"biochemistry",
"signal",
"transduction",
"mass",
"diffusivity",
"cell",
"biology",
"anatomy",
"adenylyl",
"cyclase",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"physical",
"sciences",
"chemical",
"physics",
"cell",
"signaling",
"signaling",
"cascades"
] | 2016 | A Computational Modeling and Simulation Approach to Investigate Mechanisms of Subcellular cAMP Compartmentation |
Nature has shaped the make up of proteins since their appearance , 3 . 8 billion years ago . However , the fundamental drivers of structural change responsible for the extraordinary diversity of proteins have yet to be elucidated . Here we explore if protein evolution affects folding speed . We estimated folding times for the present-day catalog of protein domains directly from their size-modified contact order . These values were mapped onto an evolutionary timeline of domain appearance derived from a phylogenomic analysis of protein domains in 989 fully-sequenced genomes . Our results show a clear overall increase of folding speed during evolution , with known ultra-fast downhill folders appearing rather late in the timeline . Remarkably , folding optimization depends on secondary structure . While alpha-folds showed a tendency to fold faster throughout evolution , beta-folds exhibited a trend of folding time increase during the last 1 . 5 billion years that began during the “big bang” of domain combinations . As a consequence , these domain structures are on average slow folders today . Our results suggest that fast and efficient folding of domains shaped the universe of protein structure . This finding supports the hypothesis that optimization of the kinetic and thermodynamic accessibility of the native fold reduces protein aggregation propensities that hamper cellular functions .
The catalog of naturally occurring protein structures [1] exhibits a large disparity of folding times ( from microseconds [2] , to hours [3] ) . This disparity is the result of roughly 3 . 8 billion years of evolution during which new protein structures were created and optimized . The evolutionary processes driving the discovery and optimization of protein topologies is complex and remains to be fully understood . Nature probably uncovers new topologies in order to fulfill new functions , and optimizes existing topologies to increase their performance . Various physical and chemical requirements , from foldability to structural stability , are likely to be additional players shaping protein structure evolution . One indicator for foldability , i . e . the ease of taking up the native protein fold , is a short folding time . Here we propose that foldability is a constraint that crucially contributes to evolutionary history . Optimization of foldability during evolution could explain the existence of a folding funnel [4] , [5] , into which a defined set of folding pathways lead to the native state , as postulated early on by Levinthal [6] . While the biological relevance of efficient folding still needs to be explored , an obvious advantage is the increase of protein availability to the cell . For instance , folding could decrease the time between an external stimulus and the organismal response . However , this increase of accessibility is probably limited by other factors such as protein synthesis , proline isomerization and disulfide formation . A probably more important point to support folding speed as an evolutionary constraint is that fast folding avoids proteins aggregation in the cell [7] . Aggregation avoidance could lead to a selection of topologically simple structures that fold rapidly or exclusion of a large number of geometrically feasible structures that compromise accessiblity . This could have reduced the catalog of naturally occurring folds [8]–[10] . The balance between the need for new structural designs and functions in evolution and the physical requirements imposing pressure on folding has remained elusive . The increasing number of organisms with completely sequenced genomes and experimentally acquired models of protein structures , combined with new techniques to study the folding behavior of proteins now open new avenues of inquiry . A common approach for such studies has been the use of molecular simulations such as lattice or coarse-grained techniques , which are efficient enough to scan sequence space . Simulations generally involve an algorithm that mimics the evolutionary accumulation of mutations . This allows to monitor how proteins are selected and evolve towards specific features that are optimized , including those linked to folding , structure and function [11]–[13] . In contrast , we have uncovered phylogenetic signal in the genomic abundance of protein sequences that match known protein structures . Specifically , phylogenomic trees that describe the history of the protein world are built from a genomic census of known protein domains defined by the Structural Classification of Proteins ( SCOP ) [14] and used to build timelines of domain appearance [15] , [16] that obey a molecular clock [17] . This information revealed for example the early history of proteins [18] , planet oxygenation [17] , and the dynamics of domain organization in proteins [19] . All-atom simulations of denatured proteins folding into their native state [20] , [21] are computationally too demanding to systematically evaluate the folding times of the available structural models of protein domains , currently 100 , 000 in total . A decade ago , Baker and co-workers [22] introduced the concept of contact order , a measure of the non-locality of intermolecular contacts in proteins . Contact order was found to be in good correlation with folding times of two state folders but not multistate proteins . Subsequent studies with extended comparison to experiments led to the definition of the Size-Modified Contact Order ( SMCO ) , ( 1 ) where is the number of contacts , is the total number of aminoacids , and is the number of aminoacids along the chain between residues and forming a native contact . By correcting for protein size , the SMCO showed an improved correlation with experimental folding times , with a correlation coefficient of 0 . 74 [23] . Here , we reveal evolutionary patterns of foldability by mapping the SMCO and thus the folding time onto timelines derived from phylogenomic trees of domain structures ( Figure 1 ) . Remarkably , we find there is selection pressure to improve overall foldability , i . e reduce folding times , during protein history . Interestingly , different topologies such as all- and all- folds show distinct patterns , suggesting folding impacts the evolution of some classes of protein structures more than others .
To trace protein folding in evolution , we determined the SMCO of protein domain structures at the Family ( F ) level of structural organization . Figure 2a shows the folding rate of each F , as measured by its average SMCO , as a function of evolutionary time . Using polynomial regression , we observed a significant decrease ( p-value = 9 . 5e-15 ) in SMCO in proteins appearing between 3 . 8 and 1 . 5 billion years ago ( Gya ) . Trends were maintained when excluding domains from the analysis solved in multi-domain proteins ( Figure S11 ) , and also when studying domain evolution at more or less conserved levels of structural abstraction of the SCOP hierarchy . Namely , we find a significant decrease of SMCO at the level of Superfamily ( SF ) , p-value = 2 . 6e-15 ) , and at the level of domains with less than 95% sequence identity ( p-value< = 2 . 0e-16 , Figure S1a , b ) . Similarly , consistent results were obtained at the F level using linear regression ( p-value = 1 . 0e-06 , Figure S1c ) . Remarkably , even within a smaller data set of only 87 proteins for which folding times have been measured [24] , we find that the experimental folding times exhibit a tendency to decrease early in protein evolution ( Figure S2 ) . As an additional way of validation , we repeated the analysis for 3 million single domain sequences with predicted SMCO [25] , and obtained a decrease again of SMCO up to 1 . 5 Gya ( p-value< = 2 . 0e-16 , Figures S3 , S4 ) . Thus , in this initial evolutionary period , proteins tended to fold faster on average . As suggested by the decrease in SMCO , during evolution , domains diminish long-range and favor short-range interactions , thereby becoming more strongly connected locally . This picture was further corroborated by an analogous analysis of evolutionary trend in tightness , measured by shortest paths in the network of protein contacts [26] . Tightness , and thus the lengths of paths in the interaction network , decreased in evolution until 1 . 5 Gya , followed by an increase , just like the SMCO ( Figure S5 ) . Our results support the hypothesis that folding speed acts as an evolutionary constraint in protein structural evolution . In contrast , we observed an increase in SMCO between 1 . 5 Gya and the present ( Figure 2a ) . Thus , the appearance of many new stuctures by domain rearrangement 1 . 5 Gya , also refered to as the “big bang” [19] of the protein world , affected the evolutionary optimization of protein folding . While a linear regression supports the SMCO increase ( p-value = 2e-16 ) , it was not as observed at the SF level or at the level of domains ( Figure S1a , b ) , and for the analysis of experimentally determined rates ( Figure S2 ) . Given the observed overall evolutionary speed-up of protein folding , we would expect a late evolutionary appearance of so-called downhill proteins , which feature ultra-short folding times on the microsecond scale . We annotated 11 downhill folders [27] by their Fs , namely a . 35 . 1 . 2 , a . 4 . 1 . 1 , a . 8 . 1 . 2 , b . 72 . 1 . 1 , and d . 100 . 1 . 1 , and show their average SMCO per family as black triangles in the timeline of Figure 2a . All of them , unsurprisingly , have an SMCO 2 , and thus fold significantly faster on average than other structures . We find 7% of families to have a lower SMCO ( SMCO 1 . 5 ) than the experimentally identified downhill folders . We predict these Fs will fold even faster than the known downhill folders , rendering them interesting candidates for folding assays . The five Fs containing the fast folders have all appeared no earlier than 2 . 5 Gya , suggesting that they are a result of lengthy evolutionary optimization . According to our predictions , the first fast-folding proteins appeared already 3 . 4 Gya . However , their frequency and optimization of folding speed continue to increase until 1 . 5 Gya . The length of the amino acid chain has been reported to influence the folding kinetics of a protein , with longer chains folding more slowly [23] , [27]–[29] . We therefore ask if the decrease in SMCO we observed from 3 . 8 to 1 . 5 Gya can be explained by a decrease in the chain length of proteins . Figure 2b shows how domain size has varied in evolution . Folding time measured by SMCO and domain size follow a very similar bimodal trend , with a clear decrease occuring prior to 1 . 5 Gya and a slight increase after the “big bang” . As expected , we find domain size , which equals in Equation 1 , and SMCO to be correlated with folding rate in agreement with other studies [8] , [23] ( Figure S6 ) . In line with this correlation , the downhill folders discussed above and shown in Figure 2a as triangles , have a small domain size of less than 100 residues in common . We next eliminated the effect of domain size on the evolutionary trends observed in folding rate to analyze factors other than domain size . To this end , we dissected our dataset according to the amino acid chain length . This analysis was done with all 92 , 000 domains to ensure enough data points for each length . The distributions of chain length are shown in Figure 3a , b for the two time periods before and after the “big bang” ( 1 . 5 Gya ) . The length distribution for proteins appearing before the “big bang” exhibited a peak at around 150 amino acids , and shifted later ( 1 . 5 Gya to the present ) to shorter chains with a peak at around 100 aminoacids , underlining the tendency for a decrease of domain size . We note that the resulting average chain length of three-dimensional structures in SCOP , which have been obtained from X-ray or NMR measurements , is smaller than the average length of sequences in genomes [30] , apparently due to the increasing experimental difficulties when working with large proteins . We then analyzed evolutionary tendencies for every domain length subset by measuring the variation in the end points of a polynomial regression . The color mapping in Figure 3a indicates an increase ( blue ) , a decrease ( yellow-red ) , or a non-significant change ( green ) of SMCO . Overall , 85% of the data returned a significant result according to the F-test . During early protein evolution ( 3 . 8–1 . 5 Gya ) , we found that 54%0 . 3% of all domains in each size subset optimized their foldability during evolution by decreasing their SMCO . Conversely , 37%0 . 4% of domains showed a slow-down in folding , i . e . a significant increase in SMCO . These results confirm the tendencies observed for the full data set ( Figure 1a ) , and hold for different tresholds of identity , namely 95% and 40% ( Figures S7 , S8 ) . As expected , due to the smaller data set , partitioning domains defined at F and SF levels according to size yielded results that were statistically not significant . In summary , even after dissecting the effect of chain length on changes in SMCO , the tendency of proteins to fold faster during evolution is confirmed . After the “big bang” , the SMCO and thus foldability showed a overall increase in evolution ( Figure 3b ) , in agreement with results from the total set ( Figure 2a ) . Apparently , fast folding did not represent a major evolutionary constraint during this period . Instead , other constraints must have been optimized at the expense of foldability . We next discuss secondary structure as one factor influencing the impact of foldability on protein structure evolution . Secondary structure composition is another factor reported to have an influence on folding kinetics [23] , [27] , [28] . We repeated the analysis of domains partitioned by size that was described above for domains in each secondary structure class of SCOP ( all- , all- , / , and + domains ) and thereby revealed differences in the evolution of foldability . As shown in Figure 4a , the tendency of a decreasing SMCO before the “big bang” is reproduced for all classes . This result was confirmed at the level 95% identity and 40% identity ( Figures S9 , S10 ) , though with a significant decrease only for the + and classes at the 40% identity level , i . e . for a much smaller data set . Again , our analysis strongly supports an evolutionary constraint for fast folding of proteins appearing early in evolution , 3 . 8–1 . 5 Gya . Interestingly , we here observe a specialization of protein classes , with all- proteins tending to fold faster and all- proteins tending to fold more slowly , all of which was supported at the 95% domain level ( Figure 4b ) . Why should the all- class be under a stronger fast folding constraint than the all- class ? Figure S12 shows the average SMCO for each secondary structure class . The all- and all- class show the highest and lowest SMCO , respectively , suggesting that all- proteins in general fold slower than all- proteins . This is in line with previous findings that containing all- proteins fold more slowly than all- proteins due to long range interactions between all- strands that increase contact order [23] , [27] , [28] .
Protein aggregation damages cellular components and can lead to a variety of neuronal diseases [31]–[33] . A way of reducing aggregation is to enhance the kinetic and thermodynamic accessibility of the native fold of a protein . Incremental increases in kinetic or thermodynamic stability of a protein might therefore represent an evolutionary trace reflecting optimization of protein foldability [34] . Here , we confirm the hypothesis that foldability exerts a constraint in the evolution of protein domain structures , as we find a tendency of proteins to on average fold faster than their structural ancestors . As expected , shortening of protein chain length during evolution is an important factor leading to faster folding . However , the exclusion of this protein-size effect preserved the trend of decreasing folding times . Thus , faster folding is not a side effect of chain shortening , but likely acts as an evolutionary constraint in itself . An alternative reason for the decrease of folding times in evolution is the need of proteins for flexibility in order to optimize their function such as enzymatic catalysis or allosteric regulation [35] . Folding speed and flexibility are known to correlate , as the formation of the compact state with no or only minor native contacts is much quicker than the arrangement of the native – often long-range – contacts [36] . Fewer native contacts in turn result in lower stability and may increase conformational flexibility as required for some biological functions [37] . Our analysis of protein folding speed on an evolutionary time line can be similarly carried out for measures of flexibility to test this scenario . Evolutionary constraints on folding are apparently not uniformly imposed onto the full repertoire of protein structures and during the entire protein history . Instead , our analysis revealed a bimodal evolutionary pattern , with folding speed increasing and decreasing before and after 1 . 5 Gya , respectively . The speed-up of folding was most pronounced for all- folds . The evolutionary inflexion point coincides with the previously identified protein “big bang” , which features a sudden increase in the number of domain architectures and rearrangements in multi-domain proteins triggered by increased rates of domain fusion and fission . We speculate that the slow down of folding that ensues could be due to cooperative interactions during folding of domains in the emerging multi-domain proteins [38] . Alternatively , the observed slow-down after the “big bang” could be related to the appearance of protein architectures that are known to help proteins to fold , such as chaperones [39] , [40] Moreover , protein architectures specific to eukaryotes appeared at 1 . 5 Gya [16] . The Eukaryotic domain of life has the most elaborate protein synthesis and housekeeping machinery , including enzymes for post-translational modification . This machinery might have mitigated the constraints for fast folding , thereby increasing evolutionary rates of change [34] , while preventing misfolding and aggregation prior to attaining the native fold [41] . Finally , we revealed striking evolutionary diversity in protein folding when comparing all- and all- fold classes from 1 . 5 Gya . Their average folding times diverged after the “big bang” , with the all- class further decreasing and the all- class instead increasing their folding times . This result can support the idea of an optimization of folding that increased the difference in folding time between all- and all- through evolution . As previously shown [22] , all- folds have on average higher SMCO and fold slower than their all--counterparts . This simply results from their different topology and is also the result of our analysis ( Figure S12 ) . We here show that earlier in evolution , however , folding times have been more similar and only diverged from each other as late as after 1 . 8 Gya . But why would all- folds have been relieved from the evolutionary constraint of fast folding ? Since the “big bang” is responsible for the discovery and optimization of many new functions , including an elaborate protein synthesis and folding machinery , we speculate that the divergence of averge folding times of all- and all- folds probably reflects an optimization of function . This optimization happens to be on the expense of foldability for only the all- class , the reasons of which remain unknown . One possible scenario would be that all- have the tendency to carry out functions that require high flexibility , a property that correlates with few long-range contacts , i . e . high foldability . An important experimental study by Baker and colleagues [42] tested the idea that rapid folding of biological sequences to their native states does not require extensive evolutionary optimization . Using a phage display selection strategy , the barrel fold of the SH3 domain protein was reproduced with a reduced alphabet of only five amino-acids without any loss in folding rate . Despite extensive changes to protein sequence , experimental manipulation preserved contact order . While these results should not be generalized to the thousands of other fold topologies that exist in nature , they are revealing . They suggest that stabilizing interactions and sequence complexity can be sufficiently small and still enable evolutionary folding optimization . In other words , optimal folding structures can find their way through the free energy landscape without extensive explorations of sequence space . This property of robustness could be a recent evolutionary development , since the SH3 domain F appears very late in our timeline of protein history . Alternatively , it could represent a general structural property . The fact that we now see clear and consistent foldability patterns along the entire timeline supports the existence of limits to evolutionary optimization of folding that are being actively overcome in protein evolution . We conjecture that these limits were initially imposed by the topologies of the early folds , and that structural rearrangements ( resulting from insertions , tandem duplication , circular permutations , etc [43]–[46] ) offered later on opportunities for fast and robust folding as evolving structures negotiated trade-offs between function and stability . We end by noting that we cannot exclude overlooking effects on folding times from cooperative folding . These could influence trends of folding times . The SMCO is known to show high correlations with folding times only for single-domain proteins [22] . Developing schemes for estimating folding times from structures comprising more than one domain is a challenge [38] but would enable a more general view onto protein foldability as a constraint throughout evolution . Moreover , our analysis is based on the sequence and structural data that is available . Results might therefore be biased by the choice of proteins and their accessibility . However , the structure of most protein folds and families have been acquired and will not exceed those that are expected [47] . Moreover , our approach allow us to steadily test if the predicted evolutionary trends of foldability are maintained upon inclusion of new sequences and protein folds into the analysis . Interestingly , multiple studies have found folding rates to correlate with stability rather than contact order [48] . Analyzing phylogenomic trends of stability might in this light be an important study to further elucidate evolutionary contraints on protein structure .
A most parsimonious phylogenomic tree of F domain structures was reconstructed from a structural genomic census of 3 , 513 Fs ( defined according to SCOP 1 . 73 ) in the proteomes of 989 organisms ( 76 Archaea , 656 Bacteria and 257 Eukarya ) with genomes that have been completely sequenced [49] . Similarly , a most parsimonious phylogenomic tree of SF structures ( 860 , 497 steps; CI = 0 . 0255 , HI = 0 . 9745 , RI = 0 . 780 , RC = 0 . 020; g1 = −0 . 109 ) was derived from a structural genomic census of 1 , 915 SFs ( defined according to SCOP 1 . 73 ) in the proteomes of 1 , 096 organisms ( 78 Archaea , 719 Bacteria and 299 Eukarya ) . The structural census scanned genomic sequences against a library of hidden Markov Models ( HMMs ) in SUPERFAMILY [50] with probability cutoffs E of 10-4 , as described in detail in previous studies [15] , [16] . Data matrices of domain abundances were normalized to genome size , coded as multistage phylogenetic characters with characters states ranging from 0 to 29 , and used to build rooted trees using maximum parsimony ( MP ) as optimality criterion in PAUP* [51] . A combined parsimony ratchet and iterative search approach avoided traps in suboptimal regions of tree space . Finally , the age of each domain ( nd ) was derived directly from its relative position as taxa in reconstructed trees . A PERL script counted the number of nodes from the most ancient domain at the base of the tree to each leaf , providing it in a relative 0-to-1 scale . These relative ages ( in nd units ) were transformed to geological ages ( in Gya ) by using molecular clocks of SFs and Fs derived previously [17] and used to construct an evolutionary timeline of domain appearance . A general finding is a sudden explosion of diversity in protein architectures at 1 . 5 Gya [19] . As a measure for the folding time of each protein architecture , we evaluated the size modified contact order ( SMCO ) of domains indexed in the SCOP database . We used the ASTRAL repositories to download the 92 , 470 three-dimensional structures classified in SCOP 1 . 73 . The phylogenomic tree was built at the F level on the basis of the same protein structures , i . e . the 1 . 73 SCOP version . We note that the SMCO calculations are based on single protein domains from SCOP , while many proteins consist of multiple domains . Some studies showed that interactions between domains might affect folding [52] . To test if the evolutionary trends also hold for the subset of domains excluding those which have been structurally solved in multi-domain proteins , we carried out the following steps . We first downloaded the CathDomainList from the website of CATH ( http://www . cathdb . info/download ) , and removed the PDB chains with two or more CATH domains or NMR structures or obsolete PDB entries . We then eliminated redundancy using the PISCES webserver ( http://dunbrack . fccc . edu/PISCES . php ) [53] using the following cut-offs: Sequence percentage identity: < = 25% , resolution: 0 . 0 3 . 0 , R-factor: 0 . 3 , sequence length: 40 10 , 000 , Non X-ray entries: excluded , C-only entries: excluded , cull PDB by chain . We detected SCOP families using HMMs on the PDB chains and removed chains with long non-domain segments , i . e . the length of a segments without any domain assignment should be less than 30 . Finally , we removed the chains with two or more SCOP families and the chains with two or more CATH entries . Using this dataset , we revealed the same tendencies in SMCO ( Figure S11 ) as those of the whole dataset ( compare Figure 2 ) . We calculated the average SMCO for each F and SF , and mapped these averages , 3 , 513 of them for F , and 1 , 915 for SF , onto timelines derived from corresponding phylogenomic trees . Average SMCO of each F or SF as a function of node distance showed non-linear dependencies that were therefore analyzed using LOESS ( locally weighted polynomial regression ) [54] , [55] to reveal global trends of foldability during evolution . A second-degree polynomial was fitted to the data at each point of the timeline , with a span of 0 . 7 . LOESS resulted in regression function values for each of the 3 , 513 F or 1 , 915 SF data points . The results from LOESS revealed a drastic change in SMCO at 1 . 5 Gya , a time point of evolution that coincides with the “big bang” of protein domain rearrangements and the rise of Eukarya [19] . We therefore also analyzed our data by two independent linear regressions describing SMCO data points before and after the “big bang” . To validate our results , we repeated the phylogenomic analysis of SMCO using two subsets of protein structures , namely only SCOP domains with 40% of sequence identity ( 10 , 570 domains ) , and those with 95% identity ( 16 , 713 domains ) . In addition , we used one subset of single domain sequences ( 3 , 500 , 000 domains ) from the TrEMBL [56] database with predicted SMCO [57] the results of which are shown in Figures S3 , S4 . Only results valid for all four different data sets and thus robust with respect to the selection of protein structures are presented here , if not otherwise noted . For the chain length analysis , we used all 92 , 000 domains to ensure enough data points for each length . The distributions of chain length are shown in Figure 3a , b . The analysis was repeated 100 times with varying data sample and every dataset ( e . g: 95% and 40% ) . We obtained standard errors of the mean , which are included in Figure 3 , 4 and Figures S7 , S8 , S9 , S10 . | Nature has come up with an enormous variety of protein three-dimensional structures , each of which is thought to be optimized for its specific function . A fundamental biological endeavor is to uncover the driving evolutionary forces for discovering and optimizing new folds . A long-standing hypothesis is that fold evolution obeys constraints to properly fold into native structure . We here test this hypothesis by analyzing trends of proteins to fold fast during evolution . Using phylogenomic and structural analyses , we observe an overall decrease in folding times between 3 . 8 and 1 . 5 billion years ago , which can be interpreted as an evolutionary optimization for rapid folding . This trend towards fast folding probably resulted in manifold advantages , including high protein accessibility for the cell and a reduction of protein aggregation during misfolding . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"evolutionary",
"biology",
"protein",
"folding",
"evolutionary",
"modeling",
"biology",
"computational",
"biology",
"evolutionary",
"selection",
"biophysics",
"evolutionary",
"processes"
] | 2013 | Evolutionary Optimization of Protein Folding |
DEAD-box helicase proteins accelerate folding and rearrangements of highly structured RNAs and RNA–protein complexes ( RNPs ) in many essential cellular processes . Although DEAD-box proteins have been shown to use ATP to unwind short RNA helices , it is not known how they disrupt RNA tertiary structure . Here , we use single molecule fluorescence to show that the DEAD-box protein CYT-19 disrupts tertiary structure in a group I intron using a helix capture mechanism . CYT-19 binds to a helix within the structured RNA only after the helix spontaneously loses its tertiary contacts , and then CYT-19 uses ATP to unwind the helix , liberating the product strands . Ded1 , a multifunctional yeast DEAD-box protein , gives analogous results with small but reproducible differences that may reflect its in vivo roles . The requirement for spontaneous dynamics likely targets DEAD-box proteins toward less stable RNA structures , which are likely to experience greater dynamic fluctuations , and provides a satisfying explanation for previous correlations between RNA stability and CYT-19 unfolding efficiency . Biologically , the ability to sense RNA stability probably biases DEAD-box proteins to act preferentially on less stable misfolded structures and thereby to promote native folding while minimizing spurious interactions with stable , natively folded RNAs . In addition , this straightforward mechanism for RNA remodeling does not require any specific structural environment of the helicase core and is likely to be relevant for DEAD-box proteins that promote RNA rearrangements of RNP complexes including the spliceosome and ribosome .
Structured RNAs are involved in many essential biological processes such as pre-mRNA splicing , regulation of gene expression , protein synthesis , and maintenance of chromosome ends [1]–[5] . These functions require the RNAs to fold into specific structures and , for some , to transition between functional conformations . However , RNAs have a strong propensity for misfolding , and because RNA structure is inherently stable , even at the local level , resolution of misfolded RNAs or rearrangements of structured RNAs can be slow on the biological timescale . These properties suggest a general requirement for RNA folding chaperones in vivo [6] , and diverse proteins have been shown to possess ATP-dependent or ATP-independent RNA chaperone activity [7] , [8] . DEAD-box proteins are superfamily 2 RNA helicases that can function as RNA chaperones to promote the formation and remodeling of functional RNAs and RNPs [9] , [10] and are linked to essentially all RNA metabolic processes in all three branches of life [10]–[12] . They use a conserved helicase core of two RecA-like domains to perform a broad range of activities including protein displacement from RNA [13] , RNA structure formation [14] , [15] , and their hallmark activity , ATP-dependent unwinding of short RNA helices [10] , [16] , [17] , including those within structured RNAs [17] . However , in addition to the helical segments that constitute RNA secondary structure , structured RNAs typically include tertiary contacts that must be disrupted during many remodeling processes [18]–[21] . Although it has been proposed that regulated binding to single-stranded RNA ( ssRNA ) might be sufficient to accelerate disruption of tertiary contacts [22] , such disruptions have not been demonstrated for any DEAD-box protein , leaving the mechanisms of these RNA remodeling reactions unclear . CYT-19 , a DEAD-box protein found in the mitochondria of Neurospora crassa , functions as a general RNA chaperone [23] , facilitating correct folding of diverse group I intron RNAs by accelerating unfolding of misfolded intermediates [17] , [19] , [24] . Here , we probe how CYT-19 promotes unfolding of structured intermediates by monitoring changes in the secondary and tertiary structure of the P1 helix within the Tetrahymena thermophila group I intron ribozyme , which has been extensively studied as an isolated tertiary folding event within a globally folded RNA [25] , [26] . The P1 helix forms by base pairing of the ribozyme with an oligonucleotide substrate that mimics the 5′ splice site . This helix docks into tertiary contacts with the ribozyme core , principally through hydrogen bonds between 2′-OH groups within the helix and nucleotides within the core [27] . CYT-19 can unwind the P1 helix , and previous results have shown that the unwinding efficiency depends on the docking stability of the P1 helix , suggesting that unwinding requires loss of the tertiary contacts prior to or during unwinding [17] . However , it was unclear how CYT-19 accomplished the RNA tertiary unfolding and whether it resulted from a known or a novel activity . To dissect this multistep remodeling process , we used a single molecule Förster resonance energy transfer ( smFRET ) approach to observe CYT-19 disruption of the 11-bp P1 helix . We directly monitored changes in both tertiary structure and secondary structure , allowing us to independently resolve and quantify the effects of CYT-19 on each step . Thus , we generated a detailed view of the process by which a DEAD-box protein can promote local unfolding of a structured RNA with disruption of tertiary and secondary contacts . Our results lead to a simple physical model that explains previous results , suggests a general mechanism for directing DEAD-box proteins to misfolded RNA intermediates , and is likely to be used broadly by DEAD-box proteins that remodel structured RNAs .
In the absence of CYT-19 , P1 was predominantly docked in most molecules but underwent cycles of undocking and redocking , as observed previously under similar conditions [25] , [28] . Docking and undocking rate constants were determined from the lifetimes of the P1 helix in the undocked and docked state , respectively , giving a docking rate constant of 120 min−1 and an undocking rate constant of 20 min−1 ( Figure 1B , top trace , Figure 1C–D , Figure S2 , and Table S1 ) . Spontaneous unwinding of the P1 helix , as measured by a loss of Cy3 fluorescence beyond the rate expected for photobleaching , was not detectable . However , addition of CYT-19 and ATP led to robust unwinding ( Figure 1B , second trace , Figure S3 , and Table S1 ) . We found that P1 unwinding occurred primarily from the low FRET state ( see Text S1 , “P1 Unwinding Monitored Using Single Molecule Fluorescence” ) . Thus , the CYT-19–mediated remodeling process occurs in two steps , with tertiary undocking preceding helix unwinding . Strikingly , the rate of P1 undocking was not increased ( Figure 1C ) , even with CYT-19 concentrations that approached saturation ( see below ) and gave substantial increases in the overall unwinding rate ( Figure S3 and Table S1 ) . Thus , CYT-19 apparently “waits” for spontaneous loss of the tertiary contacts and then interacts with the undocked P1 helix to unwind it . Although CYT-19 does not actively disrupt the P1 docking contact , we found that it increased the lifetime of the P1 helix in the undocked state . In the presence of CYT-19 , a substantial fraction of undocked events had long lifetimes , resulting in a slow phase with an observed rate constant for redocking of 20 min−1 ( Figure 1D ) . Other undocked events were followed by rapid redocking with the intrinsic docking rate constant ( 120 min−1 , Figure 1D ) , presumably because CYT-19 was not bound or was not positioned to interact with the P1 helix . Supporting a contribution from incomplete binding , the fraction of undocked events with long lifetimes increased with CYT-19 concentration ( Table S1 ) , and additional experiments indicated that CYT-19 was approaching saturation at these concentrations but not fully saturated ( Figure S3 ) . For the long-lived undocked complexes , we observed a competition between alternative fates . For undocked events that were not truncated by the termination of data collection , the P1 helix was either unwound , resulting in a loss of fluorescence ( 56% of events ) , or it redocked into the ribozyme core ( Figure 1B , middle traces , and Figure S4 ) . We calculated unwinding and docking rate constants from the lifetime distributions of these complexes and the probabilities of the alternative outcomes , and we found that CYT-19 slows P1 docking by ∼20-fold to 5 . 2±2 . 1 min−1 ( Tables S1 and S2 , and see Text S1 , “Determination of P1 Docking and Undocking Kinetics” ) . We considered the possibility that CYT-19 might be able to accelerate tertiary unfolding of a helix that forms tertiary contacts less strongly . Thus , we tested two versions of the P1 helix that include specific 2′-methoxy groups shown previously to weaken docking [25] , [26] , [30] . Although these substitutions increased the rate of undocking in the absence of CYT-19 , as expected [25] , CYT-19 did not accelerate undocking of the helices ( Figure S5 and Tables S1 and S3 ) . Further , CYT-19 retained the ability to capture these helices when they undocked spontaneously , giving decreased rates of redocking that were comparable to that of the standard P1 helix ( Figure S5 and Tables S1 and S3 ) . Together , the results indicate that CYT-19 interferes with P1 docking by binding and capturing the P1 helix after it undocks spontaneously . This “helix capture” mechanism allows CYT-19 to destabilize tertiary docking of the P1 helix , shifting the equilibrium toward the undocked state , without actively disrupting the tertiary contacts . To probe the role of ATP in CYT-19–mediated destabilization of P1 tertiary docking , we monitored P1 docking behavior with ATP analogs and in the absence of nucleotide . We found that upon replacing ATP with the ATP analog AMP–PNP , ADP , or in the absence of nucleotide , CYT-19 does not unwind the P1 helix significantly , but it retains the ability to block tertiary docking ( Figure 1B , bottom trace , Figure 1E , and Table S1 ) . With AMP–PNP , the redocking rate is the same within error as with ATP , whereas the rate is modestly increased with ADP or in the absence of nucleotide ( 2–3-fold , Table S1 ) . Overall , the lack of a requirement for nucleotide binding suggests that helix capture by CYT-19 does not require closure of the two RecA-like domains of the helicase core [31]–[33] and may result primarily from interactions of the helix with domain 2 ( see Discussion ) [34] . When CYT-19 interacts with the 11-bp P1 helix , helix unwinding is partially rate limiting for the overall disruption process , as indicated by the substantial fraction of long-lived undocking events that result in P1 redocking rather than unwinding ( Table S2 ) . Most helical segments in structured RNAs are shorter than 11 bp and correspondingly less stable , such that unwinding of these helices may be fast enough that the overall process is fully rate limited by the intrinsic loss of the tertiary contacts . We tested this idea using a ribozyme construct with a shorter P1 helix of 6 bp , which also displayed extended undocked lifetimes in the presence of CYT-19 and AMP–PNP ( Figure 2A–C ) . This helix was indeed unwound much faster by CYT-19 in the presence of ATP [17] , which precluded generating robust statistics with smFRET ( Table S4 ) . Therefore , we used rapid quench-flow techniques to measure the maximum rate constant for the overall process of P1 unwinding by CYT-19 ( kmax , which includes loss of tertiary structure and secondary structure ) . When binding of CYT-19 is saturated , the 6-bp P1 helix was unwound with a kmax of ∼6 min−1 , which is comparable to the intrinsic undocking rate constant for this helix , suggesting rate-limiting undocking ( Figure 2C–D ) . As expected from the model , the kmax value increased when docking was weakened and decreased when docking was strengthened ( Figure 2D and Figure S6 ) . Thus , unwinding of a short helix is indeed rate limited by loss of the tertiary interactions , and this tertiary disruption is not accelerated by CYT-19 . We next used the CYT-19–dependent destabilization of P1 docking to monitor the lifetime of the DEAD-box protein interaction with the ribozyme , testing whether CYT-19 remains associated with the ribozyme after it releases the P1 helix . We were particularly interested in this question because previous work suggested that CYT-19 can form two distinct interactions with RNA simultaneously: one interaction through the helicase core and a second interaction through a highly basic and unstructured “tail” of 50 amino acids ( the C-tail ) [35] , [36] . Thus , it would be possible that an interaction of the C-tail with the ribozyme could persist when the P1 helix is released from the helicase core of CYT-19 . To measure CYT-19 dissociation , we added CYT-19 and AMP–PNP to immobilized ribozyme , and then we washed CYT-19 out of the sample channel so that its dissociation from the ribozyme would be irreversible . We then monitored the FRET values of ribozyme molecules for which the P1 helix was undocked at the start of the observation period following the washout ( i . e . , those with a low FRET value of ∼0 . 2 ) . From this collection of molecules , we plotted the average FRET value as a function of time . We interpreted the data in the context of the predictions from two models . In the first model , dissociation of the helicase core from P1 results in dissociation of CYT-19 from the ribozyme . This model predicts that the average FRET value would increase back to the value of 0 . 85 , which reflects the “intrinsic” docking equilibrium of the ribozyme , with a rate constant of ∼5 . 2 min−1 , the redocking rate constant for the P1 helix after being captured by CYT-19 ( Figure 1A ) . In the second model , when P1 is released from the helicase core and redocks into the ribozyme core , CYT-19 can remain bound , presumably through its C-tail , so that it can capture P1 when it undocks again . This model would predict a time dependence consisting of at least two exponential phases . An initial increase would reflect the re-equilibration of P1 docking , with CYT-19 remaining bound , and would thus have a rate constant corresponding to the sum of the docking and undocking rate constants with bound CYT-19 ( ∼23 min−1 ) . This phase would be followed by one or more slower phases reflecting CYT-19 dissociation , which would ultimately allow the docking equilibrium to return to its intrinsic state as above . As predicted by both models , the average FRET value of these molecules increased over time , ultimately returning to a value that reflects the intrinsic P1 docking equilibrium . In strong support of the second model described above , the initial increase in FRET in the presence of CYT-19 occurred with a rate constant of ∼30 min−1 , which we infer reflects the re-equilibration of P1 docking , whereas CYT-19 remains bound to the ribozyme . A subsequent increase in the average FRET value gave a rate constant of 0 . 43 min−1 . This slow phase was not present in a control reaction lacking CYT-19 , which gave a single rate constant that reflects rapid P1 redocking ( ∼130 min−1; Figure 3 , black ) . Thus , the slower increase in average FRET value most likely reflects dissociation of CYT-19 from the ribozyme . A very slow third phase was also observed , which most likely reflects slow re-equilibration of ribozyme molecules that form alternative states that dock P1 weakly ( see also Figure S2 ) [28] . In the absence of CYT-19 , we did not collect data at the long observation times necessary to measure this phase , but we infer that it was present because the observed endpoint was lower than the expected value ( 0 . 73 versus 0 . 85 expected; Figure 3 ) . Thus , the key conclusion is that CYT-19 can remain bound to the ribozyme after releasing the captured P1 helix . The continued binding , which is most likely mediated through the C-tail of CYT-19 , is expected to allow CYT-19 to participate in multiple cycles of helix capture and unwinding , with the helicase core likely remaining poised to capture P1 or other helical elements as they become exposed by transient fluctuations . We tested the generality of the helix capture mechanism by using Ded1 , a multifunctional DEAD-box protein from Saccharomyces cerevisiae [37]–[39] . In the presence of ATP or AMP–PNP , we found that Ded1 uses the same basic mechanism to destabilize tertiary docking of the P1 helix . Specifically , Ded1 does not accelerate the loss of tertiary contacts but slows their subsequent formation ( Figure 4 and Table S5 ) , indicating that like CYT-19 , Ded1 captures the P1 helix after spontaneous undocking . There are also some interesting differences . First , long-lived undocking of P1 was observed in the presence of ATP or AMP–PNP but not in the absence of nucleotide ( Figure 4B , right , and Table S5 ) , indicating that helix capture by Ded1 depends on bound nucleotide . Second , the fraction of P1 undocking events that resulted in helix capture is lower than for CYT-19 and did not depend strongly on Ded1 concentration ( Figure 4B , left and center , and Table S5 ) , suggesting that Ded1 is saturating in our experiments for the binding that is responsible for helix capture . However , ensemble unwinding assays display increases in rate constant across the same concentration range ( Figure S7 ) . Previous studies have indicated complexity in RNA binding and unwinding by Ded1 , with participation of multiple Ded1 protomers [40] , [41] , which may contribute to the differences we observe between CYT-19 and Ded1 ( see Discussion ) . Despite these differences , Ded1 shares the basic behaviors delineated for CYT-19 , capturing the transiently exposed RNA helices and preventing re-formation of tertiary contacts .
Although DEAD-box proteins have previously been shown to promote conformational transitions of highly structured RNAs , which can require extensive disruption of tertiary interactions , it was not known how they disrupt RNA tertiary structure . Here , we used single molecule fluorescence to dissect an RNA unfolding process into discrete steps involving losses of tertiary and secondary structure . Together , our results suggest a straightforward mechanism by which DEAD-box helicase proteins can disrupt RNA tertiary structure ( Figure 5 ) . Even if the protein is pre-associated with the RNA , the helicase core does not actively disrupt tertiary contacts . Instead , it captures RNA helices that become exposed transiently by spontaneous fluctuations . For CYT-19 , this helix capture process does not require ATP and may result from RNA binding by just one of the two RecA-like core domains , as closure of the two domains typically requires a bound nucleotide [31]–[33] . Supporting this idea , domain 2 of the S . cerevisiae DEAD-box protein Mss116 can bind double-stranded RNA ( dsRNA ) in the absence of an adenosine nucleotide [34] . Ultimately , closure of the domains and consequent unwinding of the RNA helix permits the ssRNA product strands to form new contacts , allowing refolding to a functional structure or exchange between structures . This helix capture process is reminiscent of a mechanism described for some processive helicases , termed “passive unwinding , ” in which the helicase does not actively disrupt base pairs but instead captures the nucleotides from the terminal base pair upon spontaneous fraying , preventing the base pair from reforming . Processive unwinding can be achieved by this mechanism if the helicase protein repetitively captures the frayed end of the helix while it tracks directionally along one of the strands [42] , [43] . As each frayed base pair is successively captured , the loss of base stacking is expected to weaken the adjacent base pair , accelerating its fraying and therefore accelerating unwinding [43] . In a conceptually analogous manner , when a DEAD-box protein captures a helix from a structured RNA , it will not only destabilize tertiary structure by preventing reformation of tertiary contacts by the captured helix , but it will also weaken additional tertiary contacts within the folded RNA if they form cooperatively [44]–[46] . Thus , despite its passive nature , this helix capture mechanism is expected to accelerate the kinetics of large-scale tertiary unfolding of structured RNAs . This mechanism for unfolding RNA tertiary structure is likely to be used broadly by DEAD-box proteins that function to promote RNA folding , as it relies on their inherent abilities to bind dsRNA and induce ATP-dependent helix unwinding [34] , and does not depend on any specific protein binding site or structural context . Previous work showed that CYT-19 can unfold the Tetrahymena ribozyme with an efficiency that depends on the overall stability of the RNA [24] , and helix capture provides a physical model for this result . Less stable structures are expected to undergo more frequent dynamic fluctuations , allowing for more frequent capture events and therefore more efficient unfolding . Thus , this mechanism allows DEAD-box proteins to sense RNA stability , leading to preferential action on less stable misfolded intermediates , regardless of specific structural features in the misfolded RNAs , while minimizing activity upon stable , natively folded RNA . Consistent with this view , CYT-19 is activated for ATPase activity to a lower extent by the natively folded wild-type Tetrahymena ribozyme than by less stable mutants , suggesting fewer productive interactions with the more stable structure [47] . A corollary of the model is that groups of cellular RNAs that lack stable tertiary structure , such as mRNAs , are potentially subject to unfolding by DEAD-box proteins . Indeed , recent work has shown that cellular mRNAs are continually remodeled , such that they are less structured on average than they are under standard in vitro conditions [48] . Furthermore , this remodeling requires ATP [48] , highlighting the roles of RNA helicase proteins as general manipulators of RNA structure in vivo . To test whether the helix capture mechanism is used by DEAD-box proteins beyond CYT-19 , we monitored P1 helix unwinding by the multifunctional yeast protein Ded1 . Ded1 is implicated in many processes that involve remodeling of mRNAs and mRNPs , including mRNA splicing [49] , transcription initiation [50]–[54] and repression [53] , [54] , ribosome scanning [55] , RNA interference [56] , [57] , and RNA storage and decay [53] , [54] . Our findings that Ded1 does not accelerate P1 undocking and that it slows P1 redocking show that , like CYT-19 , Ded1 captures the P1 helix after it loses tertiary contacts spontaneously , thus relying on the same general mechanism for RNA tertiary structure disruption . There are also two notable differences between the proteins . Most strikingly , helix capture by Ded1 requires nucleotide binding , whereas helix capture by CYT-19 does not . One possibility is that helix capture by Ded1 involves closure of the two core domains , in which case the capture event may occur concomitantly with local strand separation [34] . However , any strand separation must be insufficient to give complete unwinding of the P1 helix , because we observe the completion of unwinding as a second , slower step that results in dissociation of the Cy3-labeled oligonucleotide . Alternatively , the nucleotide requirement may reflect a difference in the RNA binding and unwinding modes of Ded1 . Unlike CYT-19 , which is thought to use its C-tail as a tether for interaction with structured RNA , Ded1 is thought to function as a multimer , with one or more Ded1 monomers interacting with RNA structures or ssRNA extensions to localize an additional Ded1 monomer that performs helix unwinding [11] , [22] , [41] . Importantly , the Ded1 that binds the extension and serves as the landing site most likely associates through its helicase core in a nucleotide-dependent manner [11] , [22] , [41] . Thus , the nucleotide requirement for helix capture may arise not from the Ded1 molecule that interacts directly with P1 but instead from a molecule that binds elsewhere on the ribozyme and recruits the Ded1 protein that binds P1 . A second difference is that Ded1 has a lower helix capture efficiency than CYT-19 , even at protein concentrations that appear to be saturating . It is possible that when the helicase core of Ded1 binds a dsRNA , it forms an initial encounter complex that frequently dissociates and is not detected by our method . It is notable that the in vivo substrates of Ded1 tend to be less structured than the group I intron substrates encountered by CYT-19 and therefore may not require a robust helix capture efficiency . An alternative explanation is that Ded1 is preferentially positioned on the ribozyme in our single molecule experiments , most likely by additional interactions with a second Ded1 monomer as described above , and this positioning is suboptimal for capturing P1 when it undocks transiently ( but close enough to block other Ded1 monomers from solution ) . In this case , the low capture efficiency may not be a general property of Ded1 . Indeed , Ded1 is comparable to CYT-19 in its ability to promote folding transitions of group I introns [58] and at least as active as CYT-19 for overall unwinding of isolated RNA helices [17] , [58] and of the P1 helix within the context of the ribozyme ( [17] and Figure S7 ) . Although further studies focused on Ded1 will be required to determine the origins of the specific behaviors of Ded1 , the work here demonstrates that Ded1 can disrupt RNA tertiary structure using a helix capture mechanism . In addition to DEAD-box proteins that function as general RNA chaperones , the helix capture mechanism may also be important for DEAD-box proteins that function more specifically in processes such as assembly of the ribosome and spliceosome [59]–[61] . In these processes , capture and unwinding of dynamic helices would be expected to promote conformational transitions , whereas formation of a stable , folded surface would indicate that an RNA folding or protein assembly step has proceeded correctly . Thus , this helix capture mechanism is likely to be used widely by DEAD-box proteins , ranging from those that function as general RNA chaperones to those that promote specific RNA structural transitions in complex biological processes .
CYT-19 was purified as previously described ( see Text S2 , “CYT-19 Purification , ” for details ) [24] . For ensemble experiments , the L-21/ScaI form of the T . thermophila group I ribozyme was prepared by in vitro transcription ( >4 h at 37°C with 25 mM MgCl2 ) [17] . For single molecule experiments , L-21/T2 , a form of the ribozyme that is extended at the 3′-end with the tail sequence ACCAAAAUCAACCUAAAACUUACACA , was prepared under the same conditions [29] . L-16/ScaI , a version of the ribozyme with a 5′-extension of GGUUU ( resulting in an 11-bp P1 helix ) , and L-16/T2 , which includes both the 5′- and 3′-extensions , were transcribed in vitro at 30°C for 30 min with 4 mM MgCl2 to minimize self-cleavage [28] . All RNAs were then purified with RNeasy columns ( Qiagen ) and stored in TE buffer at −20°C . Dye-labeled oligonucleotides were purchased from IDT and unlabeled RNA oligonucleotides were purchased from Dharmacon . All oligonucleotides were stored in TE buffer at −20°C . For ensemble experiments , substrate oligonucleotides were 5′-end labeled with [γ32-P]ATP ( PerkinElmer ) using T4 polynucleotide kinase ( New England Biolabs ) . See Table S3 for sequences of all oligonucleotides used . Benchtop and rapid quench-flow experiments monitoring the unwinding activity of CYT-19 or Ded1 were performed at 25°C in 50 mM Na-MOPS ( pH 7 . 0 ) , 10 mM MgCl2 , 50 mM KCl , 2 mM ATP-Mg2+ ( ATP mixed with an equal amount of MgCl2 ) , and 5% glycerol as previously described [17] . Ribozymes were prefolded to the native state in 50 mM Na-MOPS ( pH 7 . 0 ) and 10 mM MgCl2 for 30 min at 50°C [17] , [28] , [29] . Alternatively , the misfolded ribozyme was generated by incubation in 50 mM Na-MOPS ( pH 7 . 0 ) and 10 mM MgCl2 for 5 min at 25°C [17] , [18] . Trace radiolabeled substrate was incubated with prefolded native or misfolded ribozyme for 5 min at 25°C . Unwinding reactions were initiated by adding CYT-19 or Ded1 and at least 25-fold excess unlabeled substrate and quenched to a solution of 33 mM MgCl2 and 1 mg/ml Proteinase K . Bound and unbound substrates were separated on a 20% native polyacrylamide gel at 4°C and quantified using a PhosphorImager and ImageQuant ( GE Healthcare ) . Data were analyzed using Kaleidagraph ( Synergy Software ) . A diode-pumped solid-state green laser ( 532 nm; CrystaLaser GCL-100-M ) and a red laser ( 637 nm; Coherent , maximum power 50 mW ) were directed through a prism at an angle that allows TIR at the surface of the sample channel , which was constructed from a glass cover slip adhered to a quartz slide with double-sided tape . The surfaces of both the cover slip and slide were passivated with a mixture of mPEG and biotin-PEG , allowing for ribozyme immobilization while preventing protein adsorption to the slide surface ( see Text S2 for description of slide preparation ) . Images were collected using a 60× water-immersion Olympus UPlanApo objective ( numerical aperture , 1 . 2 ) , filtered through a 550-nm long-pass filter ( Chroma Technology ) to remove scattered excitation light , separated into “green” and “red” images using dichroic mirrors , and focused onto the two halves of a microchannel plate intensified charge-multiplying charge-coupled device ( CCD ) ( I-PentaMAX , Princeton Instruments , Roper Scientific , Inc . ) . The ribozyme was annealed to biotinylated , Cy5-labeled tether ( ≥10∶1 molar ratio of ribozyme to tether ) in 50 mM Na-MOPS ( pH 7 . 0 ) with 200 mM NaCl by heating at 95°C for 1 min before cooling at 0 . 1°C/s to 50°C . The ribozyme was then folded to its native conformation by adding MgCl2 to a final concentration of 10 mM and incubating the solution at 50°C for 30 min . Cy3-labeled substrate oligonucleotides were then added to the prefolded ribozyme at approximately 7-fold excess and incubated for 5 min at 25°C in ribozyme buffer ( 50 mM MOPS , pH 7 . 0 , 10 mM MgCl2 ) . The ribozyme-substrate-tether complex was then diluted to 10–25 pM in ribozyme buffer and immobilized onto PEG slides via a biotin-streptavidin linkage ( see Text S2 for description of slide preparation ) . To measure P1 docking and unwinding , various concentrations of CYT-19 or Ded1 protein were diluted in CYT-19 buffer solution ( 50 mM Na-MOPS , pH 7 . 0 , 10 mM MgCl2 , 50 mM KCl , 5% glycerol ) . For some experiments , ATP or another nucleotide ( see Table S1 ) was added to a final concentration of 2 mM . The solution was then flowed through the sample channel along with an oxygen scavenging system ( OSS ) of 1 mM Trolox [ ( ± ) -6-hydroxy-2 , 5 , 7 , 8-tetramethylchromane-2-carboxylic acid , Aldrich , >97%] , 500 mM glucose , 0 . 1 mg/ml glucose oxidase , and 0 . 06 mg/ml catalase . Images of the dye-labeled molecules within the sample channel were collected in 40-ms or 100-ms frames for 10–30 s ( fully intensified at ∼1 , 000 V ) . To measure CYT-19 dissociation , slide-immobilized ribozyme was incubated with near-saturating concentrations of CYT-19 ( 1–2 µM ) along with 2 mM AMP–PNP for at least 2 min at 25°C . The sample channel was then washed with a solution of CYT-19 buffer , AMP–PNP , and OSS to remove the protein from solution , preventing CYT-19 from rebinding . After a dead time of ∼30 s , data recordings were acquired at 2-s frames for 5–10 s ( to reduce dye photobleaching ) every 30 s over a period of 30 min . Molecules that were present in the low FRET state at the start of data collection were selected to bias the analysis towards protein-bound ribozymes . This is because the fraction of ribozyme molecules that are undocked at given time is low in the absence of CYT-19 , whereas a fraction of the protein-bound molecules would be expected to persist in the undocked state during the dead time of 30 s . Fluorescence signals were collected under green laser excitation and then under red laser excitation for colocalization of Cy3 with Cy5 . The average signal-to-noise ratio was ∼5 , with green laser intensity averaging ∼15 mW ( measured near the laser aperture ) . All relevant data are within the article and its Supporting Information files , except primary data , including raw intensity values for donor and acceptor fluorophores , which are available from the UT Box database ( https://utexas . box . com/s/t0va9jj9x2xbf3wilxxg ) . | In addition to carrying genetic information from DNA to protein , RNAs function in many essential cellular processes . This often requires the RNA to form a specific three-dimensional structure , and some functions require cycling between multiple structures . However , RNAs have a strong propensity to become trapped in nonfunctional , misfolded structures . Due to the intrinsic stability of local structure for RNA , these misfolded species can be long-lived and therefore accumulate . ATP-dependent RNA chaperone proteins called DEAD-box proteins are known to promote native RNA folding by disrupting misfolded structures . Although these proteins can unwind short RNA helices , the mechanism by which they act upon higher order tertiary contacts is unknown . Our current work shows that DEAD-box proteins capture transiently exposed RNA helices , preventing any tertiary contacts from reforming and potentially destabilizing the global RNA architecture . Helix unwinding by the DEAD-box protein then allows the product RNA strands to form new contacts . This helix capture mechanism for manipulation of RNA tertiary structure does not require a specific binding motif or structural environment and may be general for DEAD-box helicase proteins that act on structured RNAs . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biochemistry",
"rna",
"rna",
"structure",
"proteins",
"enzymes",
"chaperone",
"proteins",
"biology",
"and",
"life",
"sciences",
"enzymology",
"rna",
"stability",
"long",
"non-coding",
"rnas",
"ribozymes"
] | 2014 | DEAD-Box Helicase Proteins Disrupt RNA Tertiary Structure Through Helix Capture |
The dual toxicity/essentiality of copper forces cells to maintain a tightly regulated homeostasis for this metal in all living organisms , from bacteria to humans . Consequently , many genes have previously been reported to participate in copper detoxification in bacteria . Myxococcus xanthus , a prokaryote , encodes many proteins involved in copper homeostasis that are differentially regulated by this metal . A σ factor of the ECF ( extracytoplasmic function ) family , CorE , has been found to regulate the expression of the multicopper oxidase cuoB , the P1B-type ATPases copA and copB , and a gene encoding a protein with a heavy-metal-associated domain . Characterization of CorE has revealed that it requires copper to bind DNA in vitro . Genes regulated by CorE exhibit a characteristic expression profile , with a peak at 2 h after copper addition . Expression rapidly decreases thereafter to basal levels , although the metal is still present in the medium , indicating that the activity of CorE is modulated by a process of activation and inactivation . The use of monovalent and divalent metals to mimic Cu ( I ) and Cu ( II ) , respectively , and of additives that favor the formation of the two redox states of this metal , has revealed that CorE is activated by Cu ( II ) and inactivated by Cu ( I ) . The activation/inactivation properties of CorE reside in a Cys-rich domain located at the C terminus of the protein . Point mutations at these residues have allowed the identification of several Cys involved in the activation and inactivation of CorE . Based on these data , along with comparative genomic studies , a new group of ECF σ factors is proposed , which not only clearly differs mechanistically from the other σ factors so far characterized , but also from other metal regulators .
Myxococcus xanthus is a soil-dwelling δ-proteobacterium of the group of myxobacteria used as a model to study multicellular behavior and differentiation , because it exhibits a complex developmental cycle triggered by starvation [1] . However , M . xanthus cells not only have to adapt their metabolism and behavior to changing nutritional concentrations , but also to other parameters , such as metals . Copper is a transition metal that functions as an ideal biological cofactor due to its ability to alternate between the redox states Cu ( I ) and Cu ( II ) . However , copper also generates reactive oxygen species that cause cell damage [2] . This duality forces organisms to maintain a strict homeostasis for this metal . The most representative examples of the effect of disturbances in copper homeostasis are two inherited human disorders , Wilson disease and Menkes syndrome , which are directly linked to overload and deficiency of this metal , respectively [3] . Copper is required by prokaryotes in trace amounts because it is used as a cofactor by a few proteins . Hence , most bacterial homeostatic mechanisms are devoted to conferring resistance to this metal . The most common mechanisms are copper-transporting P1B-type ATPases , copper chaperones , multicopper oxidases ( MCOs ) , and Cus systems [4] . In bacteria such as Escherichia coli , one of each of these elements is encoded in the genome [4] . In other bacteria , the homeostatic mechanism is even simpler , consisting of two P1B-type ATPases and one chaperone ( Synechocystis PCC6803 , Enterococcus hirae , and Lactococcus lactis ) , or one ATPase and one chaperone ( Bacillus subtilis ) [4] , [5] . In contrast , the large M . xanthus genome encodes a large number of paralogous genes to confer copper tolerance: three MCOs , at least two Cus systems , and three P1B-type ATPases , as well as the genes required for the biosynthesis of carotenoids [6]–[9] . This gene redundancy indicates that copper homeostasis in this myxobacterium is more complex than in other prokaryotes . All of these genes have been shown to be differentially regulated [6]–[9] , suggesting that this sophisticated network must be finely regulated by specific metal sensors . One of the signal transduction mechanisms used by bacteria to direct gene expression at the transcriptional level in response to stress signals is represented by alternative σ factors [10] . The largest group of alternative σ factors is the ECF ( extracytoplasmic function ) family , which corresponds to group 4 of the σ70 proteins [11] . ECF σ factors are small proteins , quite divergent in sequence , that contain only two regions ( σ2 and σ4 ) required for interaction with the RNA polymerase core enzyme and recognition of the promoter [12] . Their ability to promote transcription relies on a protein that is normally cotranscribed with the σ factor , named anti-σ factor . In the absence of external signals , ECF σ factors are sequestered by their cognate anti-σ . After detecting the specific stimulus , the anti-σ factor releases the σ subunit , which can then promote gene expression after recruitment of the core RNA polymerase [13]–[16] . In this report , we identify a novel metal sensor involved in copper homeostasis in M . xanthus named CorE ( for copper-regulated ECF σ factor ) . We demonstrate that CorE requires copper in order to bind to DNA and that its activity is modulated by the redox state of this metal . According to these data , we propose a new group of ECF σ factors , defined by a Cys-rich domain ( CRD ) located at the C terminus of the protein , which is essential for activation and inactivation of the protein .
Most M . xanthus genes involved in copper homeostasis are located in the genome in two clusters [8] . In copper region 2 , and next to the MCO cuoB , a gene encoding a protein with high similarity to ECF σ factors was found ( MXAN_3426 ) , suggesting that it could regulate the expression of genes involved in conferring copper resistance . This σ factor has been designated as CorE . The analysis of the CorE sequence has revealed a domain architecture with the conserved regions σ2 ( sigma70_r2 , PF04542; E-value of 3 . 2e-14 ) and σ4 ( sigma70_r4_2 , PF08281; E-value of 4 . 3e-06 ) typical of this type of σ factors [11] , [12] . To determine the role of CorE in copper homeostasis , a strain harboring a corE-lacZ fusion was constructed , and the analysis of this strain revealed that corE was up-regulated by copper ( Figure 1A ) . Additionally , an in-frame deletion mutant ( ΔcorE ) was also generated , and the phenotypic analysis of this strain confirmed that this regulator conferred copper tolerance ( Figure 1B ) . To identify genes regulated by CorE , plasmids containing fusions between the genes that have so far been involved in copper and/or other metal homeostasis in M . xanthus and lacZ were electroporated into the ΔcorE mutant . When the expression profiles of these genes in the mutant were compared with those exhibited in the wild-type ( WT ) strain , it was observed that only the MCO cuoB and the P1B-type ATPase copB remained undetectable in the ΔcorE background in the presence of copper ( Figure 2A and Figure S1 ) , indicating that they are regulated by this σ factor . Interestingly , these two M . xanthus genes exhibit a characteristic expression profile , with a peak at 2 h after the addition of exogenous copper . As ECF σ factors are usually autoregulated , corE expression was analyzed in the ΔcorE mutant . The results obtained showed that this σ factor is only partially responsible for its own up-regulation by copper , especially in the early stages after metal addition . However , some up-regulation by the metal still remains in the mutant ( Figure 1A , red lines ) , indicating that although cuoB and corE are very close in the genome ( Figure S2 ) , their regulation exhibits some differences . The comparison and analysis of the upstream regions of cuoB and copB has allowed the identification of two similar sequences that could function as the promoter elements recognized by CorE ( Figure S3 ) , one located upstream of copB , and the other upstream of a third gene genetically linked to cuoB and corE which encodes an outer membrane efflux protein ( MXAN_3424 ) . A manual search for homologous sequences to this putative CorE-binding site in the M . xanthus copper regions 1 and 2 [8] revealed the presence of two other matches , one upstream of the gene for the P1B-type ATPase CopA and the other upstream of the gene identifier MXAN_3427 , which encodes a protein with a heavy-metal-associated domain ( PF00403 , with an E-value of 4 . 7e-14 ) . To corroborate that these two genes are regulated by CorE , plasmids harboring fusions between these two genes and lacZ were introduced into the WT and ΔcorE backgrounds and β–gal specific activity was determined in these strains . The results obtained revealed that the gene for the heavy-metal-associated protein exhibits an expression profile in the WT strain very similar to those of cuoB and copB after copper supplementation ( compare Figure 2A and 2B ) . Up-regulation by copper was completely eliminated in the ΔcorE mutant , demonstrating that this gene is also part of the CorE regulon . In the case of copA the result was less clear . The expression profile of copA in the WT strain clearly differs from those exhibited by the CorE-regulated genes ( compare panel C with panels A and B in Figure 2 ) , and instead of a peak at 2 h , a plateau is reached 24 h after copper addition . Accordingly , the expression profile of copA in the ΔcorE mutant is quite similar to that of the WT strain ( Figure 2C ) . However , when the expression level of copA in these two strains was analyzed with greater precision at short intervals ( Figure 2D ) , it could be observed that the rapid induction of this gene obtained in the WT strain was no longer observed in the mutant . This result suggests that copA is subject to double regulation by CorE and another unidentified copper-dependent regulator . Nevertheless , further work will be required to unambiguously demonstrate that copA is regulated by CorE . Finally , using the consensus sequence of the promoters for these four genes , we tried to determine which other genes could also be under control of CorE . By using the approach described in Materials and Methods , another 13 similar sequences were identified in the M . xanthus genome ( Figure S3 ) . However , the fact that only two of them contain the seven invariable residues found in the other promoters , and that none of the proteins encoded by the genes located downstream of these sequences exhibit similarities to other proteins known to be involved in copper handling and trafficking , preventing us from drawing the conclusion that they are indeed regulated by CorE . As the activation of CorE by copper could be caused either by the general oxidative stress induced by this metal or by the direct binding of the protein to copper in either of its two redox states , cuoB expression in the WT strain was tested in the presence of several concentrations of the oxidants hydrogen peroxide and diamide , and the Cu ( II ) mimetic divalent metals Cd2+ , Ni2+ , and Zn2+ . Similarly , Ag+ was used to mimic Cu ( I ) . The results obtained revealed that only Cd2+ and Zn2+ could induce cuoB expression ( Figure 3A and Figure S4 ) . The fact that Ni2+ does not up-regulate cuoB is not surprising , because the same metals cannot always mimic the copper effect . As an example , the M . xanthus P1B-type ATPase copA has been reported to be induced by copper , Ni2+ and Co2+ , but not by Zn2+ [9] . It is notable that the expression levels obtained with Cd2+ and Zn2+ were not only much lower than with copper , but also that the expression profiles were different . In the case of Cd2+ , no peak was observed at 2 h; instead , a plateau was reached 24 h after metal supplementation ( Figure 3B ) . Although Zn2+ also yielded a rapid cuoB induction , the peak at 2 h was not as evident as in the case of copper ( Figure 3C ) . cuoB up-regulation by Cd2+ and Zn2+ is also dependent on CorE ( Figure 3B and 3C ) . These data indicate that Cu ( II ) is the redox state of copper that activates CorE . It should be noted that the Cd2+ and Zn2+ concentrations needed to observe a clear cuoB induction are close to the maxima that M . xanthus cells can tolerate [7] , while 0 . 3 mM copper has almost no effect on myxobacterial growth ( Figure 1B ) . It should also be noted that the addition of metals to the media not only alters the growth rates of the cultures , but also inhibits cell motility , explaining why the morphology of the cell spots is not the same in all of the media tested . Many ECF σ factors function with a cognate anti-σ which is genetically linked to the σ subunit [11]–[16] . Analyses of the genes located in the proximity of corE revealed that they encode either proteins located in the periplasmic space or in the outer membrane , or that they exhibit striking similarities to well-characterized proteins involved in specific functions , suggesting that no anti-σ factor is cotranscribed with corE . However , the possibility remained that it could be encoded in some other region of the M . xanthus genome . To test for the existence of an anti-σ factor , a strategy was designed consisting of the over-expression of corE [17] . If CorE were present in higher quantities than an unidentified anti-σ factor , it would be released from the antagonistic effect of the anti-σ , and cuoB should be expressed even in the absence of any stimulus . To follow this approach , corE was cloned under control of the oar promoter and introduced into the ΔcorE mutant harboring cuoB-lacZ to facilitate the analysis of cuoB expression . The oar promoter allows genes to be expressed constitutively at high levels [18] . As a control , a corE' cuoB-lacZ strain was also constructed , in which the corE gene was under control of its own promoter ( Figure S2 displays the cuoB-lacZ fusions used in this study ) . Quantitative analyses of cuoB expression in both strains reported no expression of this gene in the absence of copper ( Figure 4A ) , indicating that an excess of CorE was not sufficient to activate the transcription of cuoB . To corroborate that CorE expressed under the oar promoter was functional , copper was added to the media . In this case , up-regulation of cuoB was observed in both strains and with similar expression profiles ( Figure 4B ) . Finally , to confirm that corE was over-expressed when cloned under the oar promoter , we constructed the same two strains described above but introducing a His tag at the N terminus of CorE ( hCorE' ) . Western blot analyses using antibodies against the His tag confirmed that corE was indeed expressed at very high levels in the absence as well as in the presence of copper ( Figure 4C and 4D ) . CorE migrates as a double band , which must correspond to different forms of the protein . Activity of the hCorE' protein was further tested by following cuoB expression . The results obtained indicated that the proteins holding the His tag could promote cuoB transcription in the same manner as the native ones ( Figure S5 ) . Although it cannot be completely ruled out that a cognate anti-σ factor for CorE is encoded in the M . xanthus genome , all of these results indicate that CorE functions in a different manner from the one reported for the other characterized ECF σ factors . The fact that the over-expression of corE did not lead to the induction of cuoB unless copper was added to the medium suggested that CorE might require the binding of copper to promote transcription . Hence , the ability of CorE to bind DNA in vitro was tested by using electrophoretic mobility shift assays . CorE was expressed in E . coli with an N-terminal His tag and purified by affinity chromatography . Additionally , a 265-bp fragment containing the copB promoter was amplified and labeled with 32P to be used as a probe . As shown in Figure 5 , an electrophoretic mobility shift was only observed in the reaction mixture containing copper and bathocuproine disulfonic acid ( BCS ) , a specific chelating agent for Cu ( I ) [19] , [20] . These results not only confirm that CorE uses copper as a cofactor , but also suggest that Cu ( I ) prevents CorE from binding to DNA , and hence , that CorE-Cu ( II ) is the active form of this σ factor . This is also supported by the fact that only divalent metals can mimic the effect of copper on cuoB up-regulation . No other σ factor has so far been reported to require any metal to bind DNA . The expression profiles of the CorE-regulated genes exhibit a peak around 2 h after copper addition ( Figure 2A and 2B ) , in spite of the fact that corE expression is maintained for 48 h ( Figure 1A ) . This observation could be explained by proteolysis of the σ factor . To investigate this option , Western blot analyses were carried out using the hcorE' cuoB-lacZ strain . The data shown in Figure 6A demonstrate that CorE was stable for 24 h after copper addition . Another explanation could be that CorE underwent a cycle of activation/inactivation , whereby the regulator would only be in the active form for a limited period of time . As shown in Figure 5 , to obtain binding of CorE to DNA , the reaction mixture must include not only copper , but also a chelating agent for Cu ( I ) . Moreover , only other divalent metals can mimic the copper effect on cuoB up-regulation ( Figure 3 and Figure S4 ) . These data suggest that the redox state of copper could be the key in this process . To investigate this possibility , the expression of cuoB was assayed in vivo in conditions that favor the formation of Cu ( I ) and Cu ( II ) . As shown in Figure 6B and 6C , cuoB up-regulation could only be observed when copper was added to the medium . However , the maximum expression levels were diminished when the reducing agent ascorbate was also included in the medium to favor the formation of Cu ( I ) ( Figure 6C , brown line ) . Similarly , the addition of Ag+ , which mimics Cu ( I ) , also yielded expression levels lower than those obtained with only copper ( Figure 6C , green line ) . In contrast , when copper was added with the Cu ( I ) chelators BCS or bicinchoninic acid ( BCA ) [19] , [20] , cuoB expression was around three times that of the control ( Figure 6C , blue versus red and black lines ) . Moreover , the addition of copper with tetrathiomolybdate ( TTM ) , a chelator of Cu ( I ) and Cu ( II ) [21] , decreased the up-regulation mediated by this metal to a very basal level ( Figure 6C , orange line ) . In contrast , when these three chelators were tested with Zn2+ as the inducer , the expression levels of cuoB were diminished as the concentrations of all the chelators increased ( Figure S6 ) , due to the fact they can also chelate Zn2+ , although to a much lesser extent than copper . According to all these data , CorE requires copper for activation , and it only acquires an active conformation in the presence of Cu ( II ) , while the reduced state of the metal leads to an inactive conformation . This notion agrees well with the lack of a peak when up-regulation of cuoB is achieved by Zn2+ and Cd2+ , which are metals with only one redox state ( Figure 3B and 3C ) . To confirm the results obtained in vivo , the DNA-binding assay was carried out again including Ag+ or TTM in the reaction mixtures . As shown in Figure 7 , these two additives overrode the electrophoretic mobility shift achieved by the addition of copper and BCS . It should be reminded that Ag+ mimics Cu ( I ) and that TTM chelates Cu ( II ) . All of the data presented in this section demonstrate that CorE activity is modulated by the redox state of copper . This mechanism of action implies that Cu ( II ) must be available in the cytosol during the next 2 h after copper supplementation . Although it is assumed that all the copper in the reducing environment of the cytoplasm is present as Cu ( I ) under normal circumstances [2] , [22] , it is also expected that the cytoplasm will become more oxidizing in the presence of agents such as copper [23] , favoring the formation of Cu ( II ) until the reducing conditions are restored by the participation of the elements involved in copper detoxification . Furthermore , as free copper in the cells is estimated to be less than one atom per cell [24] , it is plausible to speculate that CorE functions with an unidentified Cu ( II ) -specific metallochaperone , which would ferry the cupric form through the cytoplasm to activate this σ factor . Such an activator working upstream of CorE would explain why the expression levels of cuoB do not increase when corE is over-expressed ( Figure 4 ) . However , another possible explanation for this observation could be that CorE aggregates when produced in large amounts . One paradox is the fact that CorE is activated by Cu ( II ) and inactivated by Cu ( I ) , while genes under its control encode proteins , such as CopA , CopB , and CuoB , that utilize Cu ( I ) as a substrate [7] , [9] . Nevertheless , this contradiction can be explained by considering two facts: i ) Out of the two redox states of copper , Cu ( I ) is the most toxic form [2] . As CorE-regulated genes represent the first protective barrier against the deleterious effect of copper ( please note that these genes are rapidly up-regulated after copper addition , as shown in Figure 2 and Figure S1 ) and this protein is activated by Cu ( II ) , it is plausible to speculate that copper will initially get into the cytoplasm in the form of Cu ( II ) , activating CorE , and preparing the cells to act on Cu ( I ) as soon as it appears . At this point , the CorE regulon will be inactivated by the presence of Cu ( I ) in the cytoplasm . ii ) If the presence of copper persists in the environment , M . xanthus cells will obtain protection against the metal by means of at least two other mechanisms ( first , by the P1B-type ATPase CopA and the MCO CuoA , and later , by the Cus2 and Cus3 systems ) , which are sequentially induced after copper addition [7]–[9] ( see also Figure 2 and Figure S1 ) . Although many bacterial transcriptional regulators need metal to bind DNA [25] , [26] , none of them have been reported to be modulated by the redox state of the metal . Moreover , those that function with copper show selectivity for Cu ( I ) [24] , [27] . Hence , CorE represents a novel type of bacterial copper sensor . CorE contains a short C-terminal extension after the σ4 domain consisting of 38 residues named CRD . Six of these residues are Cys . As different arrangements of Cys have been proved to be key elements in several metal-binding proteins [22] , [27] , [28] , we tried to determine whether CRD was involved in the activation/inactivation of CorE mediated by copper . An M . xanthus in-frame deletion mutant was constructed in which most of the CRD region was deleted . This strain , designated as ΔcorECRD , encoded a protein containing the two domains σ2 and σ4 of CorE , but none of the six Cys of CRD . To analyze the activity of CorECRD , the two fusions cuoB-lacZ and copB-lacZ were introduced in this mutant and β-gal activity was assayed in the absence and in the presence of copper . The data obtained revealed that neither cuoB or copB were up-regulated by copper ( data not shown ) , a result identical to that shown in Figure 2A , when the entire corE gene was deleted . These data demonstrate that CRD is essential for the copper-dependent transcription of the genes controlled by CorE . To determine which Cys are involved in CorE activity , each residue was individually mutated to an Ala by site-directed mutagenesis . The six mutated genes were introduced into the ΔcorE strain harboring the fusion cuoB-lacZ . The effect of the mutations was evaluated by analyzing the expression of cuoB in the absence and in the presence of copper . The results obtained showed different patterns . Mutations C181A and C206A exhibited transcription profiles very similar to those of the WT ( Figure 8A ) , although some small differences regarding the maximum expression levels and timing were observed , indicating that these residues play a minor role in CorE activity . More severe effects were obtained with the mutations C192A and C194A . In these mutants , the expression levels of cuoB in the absence of copper were higher than in the WT ( Figure 8B , dashed lines ) , suggesting that both Cys play some role in CorE inactivation . Moreover , although cuoB expression was up-regulated by copper in both mutants , the rapid induction and the peak exhibited by the WT at 2 h were not replicated ( Figure 8B , continuous lines ) . The effect of the mutation C189A was even more drastic , because no expression was observed in the absence of copper and the up-regulation by the metal was almost completely non-existent ( Figure 8C and 8D ) . Accordingly , it can be concluded that these three residues are important in the CorE activation process . Cys184 was clearly required for CorE inactivation , because mutation C184A yielded a constitutive expression in the absence of copper ( Figure 8E ) and the addition of copper provoked a rapid up-regulation of cuoB . Interestingly , the expression level did not peak at 2 h , but kept increasing until it reached a plateau at 24 h ( Figure 8F ) . The effect of each point mutation in cuoB expression was also analyzed in cells grown on media containing copper plus BCS or silver ( Figure S7 ) . Mutations C181A and C206 , which in the presence of copper yielded expression profiles similar to that of the WT ( Figure 8A ) , also exhibited higher expression levels in the presence of copper plus BCS , and lower levels with copper plus Ag+ ( Figure S7 ) . In the case of substitutions C189A , C192A , and C194A , BCS and Ag+ barely affected the expression levels obtained with only copper ( Figure S7 ) . However , it should be reminded that these three residues are important for activation , and that cuoB up-regulation by copper was impaired in these mutants ( Figure 8B and 8D ) , suggesting that these three proteins might only have a limited affinity for the metal . Surprisingly , however , the protein with the mutation C184A ( the most important residue in CorE inactivation as shown in Figure 8E and 8F ) , can still be modulated by the two redox states of copper ( Figure S7 ) . This result indicates that some other residues must also be involved in the inactivation of CorE . As mentioned above , two of them could be Cys192 and Cys194 , because mutations C192A and C194A yield constitutive expression of cuoB . However , it is expected that other residue ( s ) of CRD might also be required for the proper functioning of the protein ( see below ) . Taken as a whole , the results demonstrate that at least four of the Cys of CRD form a coordination environment for copper . This domain is able to recognize copper and sense its redox state , allowing the binding of CorE to DNA to activate transcription in those conditions that favor the formation of Cu ( II ) , and inactivating the σ factor in those that favor the formation of Cu ( I ) . However , how exactly CorE distinguishes between Cu ( I ) and Cu ( II ) is not easy to predict , because all of the residues identified so far that modulate the activity of CorE are Cys . Although Cys are able to coordinate Cu ( I ) and Cu ( II ) , they require the presence of other amino acids , such as His , Asp , Glu , or Met to exert this function [22] , [29] . Accordingly , it is expected that other residues also participate in the coordination of copper in either of the two redox states . Moreover , thiols are known to allow different types of modifications in an oxidative environment [30] . The exact modification of each individual Cys might also be crucial in the CorE activation/inactivation process . Further genetic , biochemical , and structural studies will be required to elucidate this intriguing question . The role of CRD in CorE resembles the function of the anti-σ domain present in many anti-σ factors [31] . Anti-σ domains require Zn2+ binding to sequester their cognate σ factor . However , the anti-σ domain and CRD differ in many aspects: i ) CRD is an extra portion of the σ factor; ii ) elimination of CRD does not activate the σ factor; and iii ) CRD senses the redox state of copper to activate or inactivate the σ factor . BLASTP analyses have allowed the identification of 21 ECF σ factors with CRD , which are distributed in only four phyla . Fourteen belong to Proteobacteria ( 9 α and 5 δ ) , four to Acidobacteria , two to Verrucomicrobia , and one to Nitrospira ( Figure 9 ) . As in the case of CorE , anti-σ factors are not linked to any of these σ factors . The alignments of these CRDs have revealed that only 4 Cys ( corresponding to residues 181 , 184 , 192 , and 194 in CorE ) are absolutely conserved among these σ factors ( Figure 9A ) . Surprisingly , Cys181 , whose mutation causes a minor effect on CorE activity , is conserved in all these regulators . In contrast , Cys189 , which is the main residue in CorE activation by copper , is only present in 11 σ factors . Interestingly , however , several of the strains with σ factors that conserve this Cys exhibit some synteny in the regions where they are encoded . The surrounding genes encode proteins with high similarities to others known to be implicated in copper handling and trafficking ( Figure 9B ) . Due to the diversity of the ECF σ factors , Staroń et al . [13] have proposed a classification of this family of regulatory proteins into 44 groups based on sequence similarities and domain architectures . However , CorE did not fit into any of the groups they defined and it was excluded from this classification . The data presented in this report support the notion that a new group should be added to the list , which will include the 21 ECF σ factors that contain CRD . So far , seven families of metal de-repressors , metal co-repressors , and metal activators are known [25] , [26] , [32] , all of which clearly differ mechanistically from CorE . Hence , elucidation of the exact mode of action of CorE will offer new insights into our current knowledge of metal sensors . Moreover , identification of the factor ( s ) working upstream of CorE will also help to elucidate how this type of σ factors works and how the trafficking of metals in the bacterial cytoplasm occurs . Finally , characterization of CorE-like proteins identified in other bacteria will also contribute to understanding the role , mechanism of action , and distribution of this novel type of regulators .
Genotypes of the bacterial strains and plasmids used in this study are listed in Table S1 and Table S2 , respectively . M . xanthus was grown in CTT medium at 30°C , supplemented with the additives indicated in each figure , as previously described [6] . E . coli was grown on Luria-Bertani ( LB ) medium at 37°C [33] . The methodologies used for obtaining the in-frame deletion mutants and the transcriptional lacZ fusion strains used in this study are the same as previously reported [7] . To generate the corresponding plasmids ( listed in Table S2 ) , the desired fragments were amplified by polymerase chain reaction ( PCR ) , using WT chromosomal DNA as a template , the oligonucleotides listed in Table S3 as primers , and the high-fidelity polymerase PrimeSTAR HS ( Takara ) [33] . PCR products were ligated to vectors pBJ113 and pKY481 [34] , [35] to generate in-frame deletion mutants and lacZ fusions , respectively . Plasmids were always introduced into M . xanthus strains by electroporation to obtain integration into the chromosome by homologous recombination . Southern blot analyses were carried out to confirm the proper recombination events . β-gal specific activity in cell extracts obtained by sonication of the strains harboring lacZ fusions was determined as previously described [7] , and it is expressed as nmol of o-nitrophenol produced per min and mg of protein . Measurements shown are the averages of data from triplicate experiments . Appropriate oligonucleotide pairs ( Table S3 ) were used to amplify by PCR an 817-bp fragment upstream of the oar gene ( MXAN_1450 ) using M . xanthus chromosomal DNA as a template [33] . Simultaneously , corE was also amplified by PCR . A BamHI site was introduced at the start codon of oar in frame with another BamHI site introduced at the start codon of corE . Both PCR products were cloned in a vector derived from pUC19 in which the ampicillin-resistance gene was substituted by one that encodes resistance to tetracycline ( Tetr ) . The resulting plasmid , pNG06 , was introduced by electroporation into an M . xanthus strain with the genotype ΔcorE cuoB-lacZ , and several kanamycin-resistant ( Kmr ) and Tetr colonies were analyzed by Southern blot to confirm the proper recombination event . β-gal specific activity was determined to quantify cuoB expression . As a control , the plasmid pNG00 was constructed , in which corE was cloned under control of its own promoter . This plasmid was also electroporated into the ΔcorE cuoB-lacZ to restore corE at its original genomic location ( see Table S1 and Figure S2 ) . To corroborate that CorE was being over-produced under the constitutive oar promoter , we constructed the same strains described above , but introducing an N-6His tag upstream of CorE , to obtain the strains hcorE' cuoB-lacZ and oar-hcorE' cuoB-lacZ ( Table S1 ) . Briefly , the corE gene with an N-6His tag was amplified with appropriate oligonucleotide pairs ( Table S3 ) using pETTOPOCorE plasmid ( see below ) as a template . The PCR product obtained was cloned under control of the oar promoter and its own promoter as above , obtaining plasmids pNG08 and pNG05 , respectively . Plasmids were introduced into the ΔcorE cuoB-lacZ strain , and Kmr Tetr colonies were also analyzed by Southern blot hybridization . The corE gene was amplified by PCR using the M . xanthus chromosome as a template and the primers CorEcTopoR and CorEcTopoF ( Table S3 ) . The PCR product was cloned into pET200/D-TOPO using a Champion pET Directional TOPO Expression Kit supplied by Invitrogen to create pETTOPOCorE . The absence of unwanted mutations in the insert was confirmed by DNA sequencing . As a result , CorE contained an N-6His tag at the N terminus , and this protein has been named hCorE . The resulting plasmid was used to transform the strain E . coli BL21 Star ( DE3 ) . The transformed cells were grown in LB medium containing 25 µg/ml kanamycin at 37°C until the optical density at 600 nm ( OD600 ) reached 0 . 7 to 0 . 8 . Induction was performed by the addition of 1 mM of isopropyl-β-D-thiogalactopyranoside . Induced cultures were incubated with shaking at 37°C for 6 h . One liter of the cell culture was collected by centrifugation and resuspended in 20 mM Tris-HCl ( pH 7 . 5 ) containing the protease inhibitors leupeptine and antipain ( 2 µg/ml each ) , 10 µg/ml DNAse I , and 5 mM MgCl2 . Cells were disrupted in a French pressure cell ( at 9000 psi ) , followed by centrifugation ( 13000×g for 30 min at 4°C ) to remove cell debris . The resulting soluble extract was loaded onto a HisTrapHP column ( bed volume 5 ml; GE Healthcare ) equilibrated with 20 mM Tris-HCl ( pH 7 . 5 ) containing 0 . 5 M NaCl and 30 mM imidazole . Elution was carried out with a linear imidazole gradient ( 30–250 mM ) in the same buffer . Protein fractions were analyzed by using SDS-PAGE . Those fractions containing hCorE were pooled , concentrated by ultrafiltration ( cutoff of 10 kDa ) , and equilibrated to 20 mM Tris-HCl ( pH 7 . 5 ) . The purified hCorE protein content was determined by the Bio-Rad protein assay kit as specified by the manufacturers , using bovine serum albumin as standard . The purity of the samples was higher than 90% . This methodology was used to detect hCorE either in M . xanthus or E . coli extracts . Cells were disrupted by sonication and centrifuged to remove cellular debris . Proteins were separated by SDS-PAGE and transferred onto a membrane of Immobilon-P at 0 . 8 mA/cm2 for 1 . 5 hours . hCorE was detected with an anti-His G-AP antibody ( Invitrogen ) , which is conjugated with alkaline phosphatase , using nitro-blue tetrazolium chloride and 5-bromo-4-chloro-3′-indolyphosphate p-toluidine as substrates , following the instructions specified by the manufacturer . A DNA fragment containing an upstream sequence of the predicted ribosome-binding site of copB was amplified from the M . xanthus genome using primers 3422EMSA265F and 3422EMSA265R ( Table S3 ) . After purification , the 265-bp PCR product was labeled with T4 polynucleotide kinase ( MBI Fermentas ) and [γ-32P]ATP , and purified through a ProbeQuant G-50 Micro Column ( GE Healthcare ) . Binding reactions contained 20 mM Tris-HCl ( pH 7 . 5 ) , 2 mM MgCl2 , 0 . 25 mg/ml of bovine serum albumin , 0 . 5 mM dithiothreitol , 15% glycerol ( v/v ) , 40 mM KCl , 0 . 5 nM of labeled DNA ( 13000 cpm ) , and a 500-fold molar excess of competitor DNA ( polydIdC ) . When indicated , 500 nM of hCorE protein , 0 . 1 mM CuSO4 , 0 . 05 or 0 . 1 mM BCS , 0 . 1 mM AgNO3 , or 0 . 1 mM TTM were also added to the reaction mixtures . After incubation for 10 min at 30°C , the mixtures were loaded onto a pre-run 5% polyacrilamide gel and run at 100 V for 1 h . The gel was dried under vacuum and exposed to an autoradiography film at −80°C . Single amino acid substitutions in the CRD of CorE were generated using the QuikChange II site-directed mutagenesis kit ( Stratagene ) . Plasmid pNG00 ( containing the WT corE sequence ) was used as a template . The primers were designed using the QuikChange Primer Design Program ( http://www . genomics . agilent . com/ ) . Oligonucleotides used to generate the six point mutations are listed in Table S3 . The presence of the desired mutations in the resulting plasmids pNG181 , pNG184 , pNG189 , pNG192 , pNG194 , and pNG206 ( carrying the mutations C181A , C184A , C189A , C192A , C194A , and C206A , respectively ) , and the absence of unwanted mutations in other regions of the inserts were confirmed by DNA sequencing . These plasmids were electroporated into M . xanthus JM51EBZY ( ΔcorE cuoB-lacZ ) to obtain strains SDM181EBZY to SDM206EBZY ( Table S1 ) . Genes encoding ECF σ factors with a CRD in the prokaryotes were identified by BLASTP analysis of all the genome sequences deposited in the database of the National Center for Biotechnology Information ( http://www . ncbi . nlm . nih . gov/genomes/lproks . cgi ) and the DOE Joint Genome Institute ( http://www . jgi . doe . gov/ ) . All the sequences obtained with an E-value<2e-10 that conserved at least 4 Cys in the C terminus were back-searched against Pfam ( http://pfam . sanger . ac . uk/ ) [36] to unequivocally verify that they matched the σ2 ( PF07638 ) and σ4 ( PF08281 ) regions conserved in all the ECF σ factors [11] , [12] . Protein sequence alignments of the 21 σ factors with CRD identified were performed using ClustalX [37] . The graphic representation of the multiple sequence alignment was adjusted and colored manually using the model generated at ESPript . cgi Version 3 . 06 CGI 3 . 05 ( http://espript . ibcp . fr/ESPript/cgi-bin/ESPript . cgi ) . For synteny determinations , two upstream and two downstream predicted proteins from CorE were initially aligned with the other 20 predicted proteomes using the BLASTP program ( E-value<1e-4 ) . We then manually analyzed the gene organization of positive matches . In the case of conservation , we extended the BLASTP searches to other genes within the same region as already described by Pérez et al . [38] . First , the upstream regions of copB and cuoB were manually analyzed to find the sequence AAC , which is well conserved in the −35 regions of other known ECF σ-factor promoters in M . xanthus and other bacteria [39] . Alignment of the sequences found permitted the identification of two regions that could function as the CorE-binding site ( Figure S3 ) . Next , homologous sequences to these ones were manually searched in all the upstream regions of the genes of the copper regions 1 and 2 of the M . xanthus genome [8] . By using this strategy we found , in copper region 2 , two other putative CorE-dependent promoters upstream of MXAN_3427 and MXAN_3415 ( which encodes the P1B-type ATPase CopA ) . Experimental approaches demonstrated that the expression of these two genes is dependent on copper and that they are regulated by CorE . The alignment of the four sequences was used to define the CorE-binding motif . The consensus sequence of the −35 region given in the IUPAC code ( defined by Nomenclature Committee of the International Union of Biochemistry ) was used to analyze the whole M . xanthus genome at the Virtual Footprint server ( http://prodoric . tu-bs . de/vfp/ ) [40] to determine which other genes might be part of the CorE regulon . 754 positive sequences were obtained with a maximum of two mismatches with respect to the defined consensus . All the sequences were manually examined for the proper strand orientation and the conservation of the invariant residues observed in the −35 region . The resulting positive matches were again screened to identify a conserved G in the −10 region , maintaining a distance of 16–18 residues from the AAC of the −35 region . Only 13 positive sequences were finally selected ( Figure S3 ) . | Copper exerts a dual effect on living organisms . It is essential for life , but an excess provokes cell damage , forcing cells to maintain a regulated homeostasis for this metal . These two antagonistic biological effects of copper are clearly illustrated by two human genetic disorders , Menkes syndrome and Wilson disease , caused by deficiency or accumulation of this metal , respectively . Myxococcus xanthus , a soil-dwelling bacterium , also has to cope with changes in copper concentration in its environment . The large genome of this myxobacterium encodes many genes involved in copper homeostasis , all of which are differentially regulated , indicating that many regulators participate in copper homeostasis in this prokaryote . Here , we identify one of these regulators ( CorE ) , which belongs to the family of the extracytoplasmic function ( ECF ) σ factors . We demonstrate that CorE represents a novel group of ECF σ factors and of metal regulators , because its activity is modulated by the redox state of copper . This ability resides in a Cys-rich domain , which has also been found in other σ factors of different bacterial phyla . Therefore , we propose that CorE is the first member of a mechanistically new group of ECF σ factors . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
] | [
"cellular",
"stress",
"responses",
"gene",
"regulation",
"regulatory",
"proteins",
"dna-binding",
"proteins",
"microbiology",
"dna",
"transcription",
"gene",
"function",
"prokaryotic",
"models",
"model",
"organisms",
"molecular",
"genetics",
"applied",
"microbiology",
"cofactors",
"proteins",
"gene",
"expression",
"comparative",
"genomics",
"biology",
"molecular",
"biology",
"microbial",
"ecology",
"biochemistry",
"cell",
"biology",
"genetics",
"genomics",
"molecular",
"cell",
"biology",
"myxococcus",
"xanthus",
"genetics",
"and",
"genomics"
] | 2011 | CorE from Myxococcus xanthus Is a Copper-Dependent RNA Polymerase Sigma Factor |
Successful perception depends on combining sensory input with prior knowledge . However , the underlying mechanism by which these two sources of information are combined is unknown . In speech perception , as in other domains , two functionally distinct coding schemes have been proposed for how expectations influence representation of sensory evidence . Traditional models suggest that expected features of the speech input are enhanced or sharpened via interactive activation ( Sharpened Signals ) . Conversely , Predictive Coding suggests that expected features are suppressed so that unexpected features of the speech input ( Prediction Errors ) are processed further . The present work is aimed at distinguishing between these two accounts of how prior knowledge influences speech perception . By combining behavioural , univariate , and multivariate fMRI measures of how sensory detail and prior expectations influence speech perception with computational modelling , we provide evidence in favour of Prediction Error computations . Increased sensory detail and informative expectations have additive behavioural and univariate neural effects because they both improve the accuracy of word report and reduce the BOLD signal in lateral temporal lobe regions . However , sensory detail and informative expectations have interacting effects on speech representations shown by multivariate fMRI in the posterior superior temporal sulcus . When prior knowledge was absent , increased sensory detail enhanced the amount of speech information measured in superior temporal multivoxel patterns , but with informative expectations , increased sensory detail reduced the amount of measured information . Computational simulations of Sharpened Signals and Prediction Errors during speech perception could both explain these behavioural and univariate fMRI observations . However , the multivariate fMRI observations were uniquely simulated by a Prediction Error and not a Sharpened Signal model . The interaction between prior expectation and sensory detail provides evidence for a Predictive Coding account of speech perception . Our work establishes methods that can be used to distinguish representations of Prediction Error and Sharpened Signals in other perceptual domains .
The observation that our perception of the world not only depends on sensory input but also on our prior knowledge has been of longstanding interest in psychology [1] and neuroscience [2–5] . There is widespread agreement that sensory input and prior knowledge are combined in neural representations; by which we mean the specific patterns of neural activity that are associated with the content of our sensory experiences . However , despite extensive experimental work in many sensory modalities [6–16] , the neural and computational mechanisms by which prior knowledge guides perception are unclear [17 , 18] . One proposal is that neural representations of expected sensory signals are enhanced or tuned [19 , 20] . Critically , in this account , perceptual representations are sharpened by relevant prior expectations in much the same way as if the quality of the sensory input was increased [17 , 18] . Alternatively , Predictive Coding schemes suggest that expected sensory input is explained away and unexpected information is represented in the form of prediction errors ( cf . in engineering [21 , 22] and neuroscience [3 , 23 , 24] ) . One intuitively attractive aspect of Predictive Coding , both for engineering and neuroscience , is its assumption that minimal effort should be invested in further processing of sensory information that is already known or expected . Our goal in this work is to distinguish these two fundamental coding schemes for how prior expectations influence perception . Do neural representations of sensory signals contain only the unexpected parts of the sensory evidence ( from now on we will refer to these as “Prediction Errors” ) ? Or do they contain an enhanced version of the expected sensory evidence ( from now on “Sharpened Signals” ) ? Our approach allows us to test each of these coding schemes against behavioural and fMRI data to determine how expected sensory signals are neurally coded . Sharpening and Predictive Coding schemes have proved hard to distinguish in neuroscience [2 , 5 , 25] . Predictive Coding theories have proposed that each level of a cortical hierarchy contains two functionally distinct subpopulations ( i . e . , prediction and prediction error units [3 , 20 , 24 , 26] ) . In these accounts , the signals that are passed forward from one level of the hierarchy to the next ( i . e . , the feedforward signals ) represent Prediction Error . This Prediction Error signal is also used to update prediction units within the same level of the cortical hierarchy ( through lateral interactions ) , such that prediction units represent a sharpened version of the sensory signal [3] . Therefore , evidence for Sharpened Signal representations has been used to support both Predictive Coding theories [20] as well as pure Sharpening theories without computation of Prediction Errors [27] . However , evidence for Prediction Error representations would be uniquely consistent with Predictive Coding and challenge pure Sharpening accounts . Speech perception provides a biologically significant domain in which prior knowledge has been clearly shown to guide perception ( for review , see [28] ) . Behavioural experiments show that numerous sources of proximal and distal prior knowledge ( including subtitles , lip-reading , lexical constraint , or semantic predictability ) can enhance subjective and objective perceptual outcomes for degraded speech [29–33] . The dominant computational theories of speech perception have included interactive-activation mechanisms that lead to enhanced representations of expected signals ( i . e . , Sharpened Signals ) , most notably in the TRACE model [34] but also in other influential models of speech perception [35–38] . More recent work has proposed Predictive Coding schemes , which use Prediction Error signals [4 , 7 , 39] to explain how prior expectations improve sensory processing . However , evidence to overturn Sharpening accounts has been lacking . One challenge for existing research is that both suggested computational schemes predict reduced neural activity during perception of expected speech signals , either due to suppression of unexpected noise ( in Sharpened Signals ) or suppression of expected signals ( in Prediction Errors ) . Brain regions in and around the left posterior superior temporal sulcus ( STS ) are proposed to support perceptual processing of speech [40 , 41] and integrate expectations from different modalities with speech input [8 , 39 , 42–46] , and activity in this region is proposed to show effects of prior training on speech responses [47–49] . While these studies provide abundant evidence that prior knowledge can influence the magnitude of activity in the posterior STS during speech perception , they do not determine the computational mechanism by which relevant prior knowledge enhances perception of speech . However , multivariate analyses of the representational content of brain responses can differentiate these two accounts by testing whether representations of speech signals are enhanced ( in line with Sharpened Signals ) or suppressed ( Prediction Errors ) when they match prior expectations . Therefore , we used representational similarity analysis [50] on multivoxel response patterns in the posterior STS . This approach is “information based” because it measures how much information about the phonetic form of speech is contained in spatial fMRI activation patterns in each of the experimental conditions that we tested [51 , 52] . We focus on the posterior STS because this is both a region in which effects of prior knowledge on speech processing have been repeatedly shown and also a region in which syllable identity can be decoded from multivariate BOLD signals [53–57] . To guide our interpretation of this data , we constructed two computational simulations based on either Sharpened Signals or Prediction Errors . Both these simulations can explain observations of perceptual enhancement and reduced fMRI responses in the left posterior STS for degraded speech that matches prior expectations . Crucially , however , these simulations make distinct predictions for the results of multivariate representational similarity analysis . In our Sharpened Signal model , simulated neural representations are enhanced for degraded speech that matches prior expectations in the same way as for speech that is presented with more sensory detail ( Fig 1A ) . However , in our Prediction Error model ( Fig 1B ) , these two manipulations have an interactive effect on simulated neural representations: the effect of increasing sensory detail depends on whether or not speech matches prior expectations . Increased sensory detail for expected speech leads to reduced information about the phonetic form of speech in simulated Prediction Errors . In contrast , increased sensory detail for unexpected speech leads to more Prediction Error and , hence , more information in simulated neural representations . In our experimental work , we test both these proposals using representational similarity analysis ( RSA ) fMRI applied to BOLD responses time-locked to the onset of a degraded spoken word . To obtain experimental evidence to differentiate these two computational accounts , we therefore simultaneously manipulated ( 1 ) prior knowledge of speech content by having participants read matching/mismatching written words or neutral text ( “XXXX” ) before spoken words [8 , 33 , 58] and ( 2 ) sensory detail in speech by presenting vocoded spoken words at one of two different levels of acoustic degradation ( Fig 2 ) [59 , 60] . In this way , we could test whether representations of the phonetic form of speech in the posterior STS [55 , 57 , 61] are enhanced similarly by changes in prior knowledge as by changes to sensory detail ( in line with Sharpened Signals ) or whether these two factors interact ( in line with Prediction Errors ) .
First , we confirmed that , consistent with both Predictive Coding and Sharpening , providing informative prior expectations improves perception of degraded speech . Participants’ report of the degraded spoken words was improved by both increased sensory detail and matching prior information from a preceding written word ( Fig 3 ) . A two-way repeated measures ANOVA with the factors sensory detail ( 4- versus 12-channel ) and prior knowledge ( Match versus Neutral ) revealed significant main effects of sensory detail on word report ( 12-channel: 85 . 39% > 4-channel: 57 . 83% correct; F ( 1 , 20 ) = 133 . 419 , p < 0 . 001 , eta squared = 86 . 96 ) and prior knowledge ( Match: 84 . 42% > Neutral: 63 . 49% correct; F ( 1 , 20 ) = 89 . 582 , p < 0 . 001 , eta squared = 81 . 75 ) , and a significant interaction ( F ( 1 , 20 ) = 74 . 997 , p < 0 . 001 , Fig 3A ) . These effects of sensory detail and prior knowledge combined such that 4-channel vocoded speech in the Match condition was reported with equivalent accuracy as 12-channel vocoded speech in the Neutral condition ( 79 . 17% versus 83 . 53% correct , t ( 20 ) = -1 . 427 , p = 0 . 169 ) . Nonetheless , word report was further enhanced in the Match 12-channel condition compared to the Neutral 12-channel condition ( 89 . 68% versus 83 . 53% correct , t ( 20 ) = 3 . 267 , p = 0 . 004 ) and the Match 4-channel condition ( 89 . 68% versus 79 . 17% correct , t ( 20 ) = -4 . 460 , p < 0 . 001 ) . Word report in the Match 12-channel condition was also more accurate than in a condition in which the spoken word was omitted and participants were prompted to report the preceding written word ( 89 . 68 versus 82 . 14% correct in the written only condition , t ( 20 ) = 2 . 348 , p = 0 . 029 ) . These findings confirmed that participants used prior knowledge to enhance perception of degraded speech even when relatively clear 12-channel speech was presented . Behavioural responses in the Mismatch conditions resemble the pattern of results in the Neutral condition ( see S1 Fig ) . Second , we sought to localise the univariate BOLD activity decrease for degraded spoken words that follow matching written words relative to words following neutral cues . These observations replicate previous findings but do not distinguish between accounts in which this effect is due to suppression of unexpected noise ( Sharpened Signals ) or suppression of expected signals ( Prediction Errors ) . Univariate BOLD responses were influenced by both increased sensory detail and matching written text . A two-way repeated measures ANOVA with the factors sensory detail ( 4- versus 12-channel ) and prior knowledge ( Match versus Neutral ) revealed a main effect of matching versus neutral prior knowledge on responses in the left posterior STS , as predicted , and in other regions of the speech processing network ( Fig 3B and 3C , S1 Table: main effect of Match/Neutral , p < 0 . 05 FWE voxel correction ) . Mean beta values extracted from the left posterior STS showed a reduction during Match in contrast to Neutral conditions ( Fig 3; inspection of contrast estimates from all other clusters also revealed less activity for Match than Neutral ) . In addition , there was a main effect of sensory detail in bilateral insula , SMA , left premotor , and orbitofrontal cortex ( S2 Table; main effect of 4/12-channel , p < 0 . 05 FWE ) . Inspection of contrast estimates revealed less activity for 12- than 4-channel in most clusters; the reverse pattern was only observed in the right middle orbitofrontal gyrus ) . The interaction of prior knowledge and sensory detail did not reach corrected significance ( S3 Table ) . Increased BOLD activity for Mismatch > Match resembles the difference in BOLD activity found for Neutral > Match ( see S1 Fig and S4 Table ) . This confirms that our observed effects are not due to differences in attention , anticipation of more difficult trials , or baseline differences between the Match and Neutral conditions ( see S1 Text ) , but rather due to the influence of matching prior knowledge on speech perception . The behavioural and univariate results appear to be in line with both Sharpening and Predictive Coding theories . Although the underlying coding schemes differ , both accounts suggest that increased sensory detail and matching prior information should improve recognition performance and that prior matching knowledge should reduce univariate fMRI responses . To confirm this , we constructed two computational models of spoken word recognition , which only differed by using representations of Sharpened Signals or Prediction Errors to simulate how sensory information and prior knowledge are combined ( see S2 Fig for details ) . In both these models , behavioural performance ( i . e . , word recognition ) was simulated by the model’s ability to identify the correct word presented in degraded speech , and univariate fMRI results ( i . e . , the magnitude of hemodynamic activity in the left posterior STS ) were simulated by the number of processing iterations required for the model to settle . By simulating the univariate fMRI signal with the number of model iterations , we assume that the hemodynamic signal as measured by fMRI integrates over several seconds of neural activity and that a longer duration of neural processing should result in an increased amplitude of the fMRI signal [63] . Six parameters were optimised for each model: the amount of sensory degradation used to simulate 4- and 12-channel vocoded speech ( which influences word report and processing time ) , variability and confidence in behavioural responses ( which influences word report ) , and the rate and duration of model updating ( which primarily influences processing time; see S3 Fig for sensitivity analysis of the optimized parameters ) . We used Akaike weights to compare goodness of fit to word report and univariate hemodynamic responses in the left posterior STS ( see Materials and Methods for details ) . Based on 1 , 000 replications using the best-fitting set of parameters , a probability density function for the predicted outcome of behavioural and univariate results was generated for both model simulations . We then used the evidence ratio of Akaike weights to compare the relative likelihood of the two models given the observed data . The ratio of the Akaike weights revealed a slightly higher likelihood of Sharpened Signal model than of the Prediction Error model for both the behavioural results ( wPE/wSharp = 0 . 9307 ) and the univariate results ( wPE/wSharp = 0 . 8149 ) . Both of these values are close to 1 , indicating that there is a negligible difference between the two models [64] . The good fit observed between these models and behavioural and univariate hemodynamic data from the current experiment suggests that computation of Sharpened Signals and Prediction Errors can explain the effect of increased sensory detail and matching prior information during perception of degraded words ( model simulations and experimental results shown in Fig 3 ) . Although both models can accurately simulate behavioural and univariate fMRI results , they perform different underlying computations and make different assumptions about the effect of matching prior knowledge on neural representations of speech signals . The Sharpened Signal model predicts that degraded speech is better represented in the STS when it matches prior knowledge , because expected sensory features of the speech input are enhanced and unexpected sensory features are suppressed . In contrast , the Prediction Error model assumes that the expected part of the speech input is explained away ( i . e . , reduced ) and only Prediction Errors ( i . e . , the difference between heard and expected speech ) are represented in the STS . To test these two simulations , we assessed the neural representation of speech information by means of RSA [50] . This approach allowed us to quantify the amount of information about the phonetic form of speech that is carried by the spatial pattern of fMRI activity in each of our four critical conditions . We designed our experiment to test for categorical representations of syllable similarity , because previous studies ( in fMRI [55 , 57] and intracranial recordings [61] ) showed that categorical representations of speech , such as vowels and syllables rather than acoustic cues , are decodable from the STS . Neural representational similarity was first measured by computing a representational dissimilarity matrix ( RDM ) for multivoxel fMRI responses for each item and condition ( see Materials and Methods for details ) . To quantify the amount of speech information , we computed the Fisher-z-transformed Spearman correlation between the observed RDM and a hypothesised RDM of interest that tested for increased similarity between pairs of syllables that shared the same vowel and had other segments in common ( e . g . , “sing” and “thing” ) compared to pairs of unrelated words ( e . g . , “sing” and “bath” , see Fig 4A ) . This similarity measure was computed separately for each condition . This analysis targets speech representations in the posterior STS by testing for similarity of words that have similar phonetic forms but different lexical or semantic representations . We did not compare identical words presented in different scanning sessions . To test our two computational simulations of spoken word recognition , we applied the same multivariate analysis to representations of the sensory input in the Sharpened Signal and Prediction Error models for each of our four conditions ( for details , see Materials and Methods ) . As for the multivoxel fMRI RSA , we quantified the difference in pattern similarity between pairs of similar and dissimilar syllables ( e . g . , “sing” and “thing” versus “sing” and “bath;” see Fig 4A ) . The simulation for the Sharpened Signal model showed increased similarity for both increased sensory detail and matching prior information ( Fig 4C ) . In contrast , the simulation for the Prediction Error model showed an interaction between sensory detail and prior information ( Fig 4D ) . Specifically , there was greater pattern similarity for similar syllable pairs in the Neutral 12-channel than in the Neutral 4-channel condition , whereas in the Match 12-channel there was less pattern similarity than in the Match 4-channel condition . This outcome resembles the interaction of sensory detail and prior knowledge shown for multivariate fMRI results in the posterior STS ROI ( Fig 4B ) . In addition , we repeated the cross-subject consistency analysis on representations generated by individual simulated participants . For the Prediction Error but not for the Sharpened Signal model , this showed the same crossover interaction of sensory detail and prior knowledge as in the equivalent fMRI analysis , suggesting a common underlying explanation ( see S1 Text and S6A–S6C Fig ) . Again , we used the evidence ratio of Akaike weights to compare the evidence for both models given the pattern similarity results in the left posterior STS ( see Materials and Methods ) . Importantly , both models used parameters optimised to simulate the behavioural and univariate fMRI results , and no modifications or parameter optimisation were performed when simulating similarity in spatial patterns of fMRI activity . For the multivariate fMRI results , the evidence ratio of the Akaike weights revealed that the multivariate fMRI patterns very strongly supported the Prediction Error model over the Sharpened Signal model ( wPE/wSharp = 1 . 898 x 1011 , tested based on the independent ROI in the posterior STS [57] ) . Hence , computational simulations provided compelling evidence that multivariate fMRI results are more consistent with computation of Prediction Errors than of Sharpened Signals in the posterior STS during the perception of degraded speech .
Current Predictive Coding theories suggest that cortical regions involved in sensory processing contain two subpopulations of neurons: ( 1 ) prediction error units that represent the unexpected part of the incoming sensory information and ( 2 ) prediction units that represent the expected part of the incoming sensory information ( and can be sharpened by matching prior expectations ) [3 , 24 , 75] . These models have thus drawn support from empirical evidence showing either Prediction Errors [26 , 39 , 49 , 76] or Sharpened Signals [20] by attributing neural responses to prediction error and prediction units , respectively . Our goal in this study was to test two functionally distinct coding schemes in isolation by building computational models in which a simulated cortical area passes only one type of information forward ( only Prediction Errors or Sharpened Signals ) . In the context of these simulations , our results provide clear evidence for representations of Prediction Errors . However , our multivariate fMRI findings do not oppose theories of Predictive Coding that propose Sharpened Signals coded by prediction units in addition to Prediction Errors in prediction error units [3 , 23 , 24] . The absence of evidence for Sharpened Signals in our data from the STS could be explained by previous proposals that fMRI measurements are dominated by responses from prediction error units ( as [26 , 77] have argued for visual cortex ) . It could be that other neural measures , such as neurophysiological recordings with depth electrodes [78] or laminar-specific ultra-high field strength fMRI [79 , 80] are better able to detect responses from prediction units and could provide evidence of laminar-specific representations of Prediction Errors and Sharpened Signals . Nonetheless , the interaction observed in the present study favours Predictive Coding theories ( with representations of Prediction Error ) over the traditional view that the brain directly passes forward the sensory input , as hypothesised in a Sharpening scheme without representations of Prediction Error . Our simulations show that in Sharpening schemes , the Match 12-channel condition should contain the clearest representation of speech content . This was not observed in the present data ( compare Fig 4B and 4C ) . Our work not only provides evidence to support the hypothesis that integration of prior expectation and perceptual input for speech is achieved through computation of Prediction Errors or Sharpened Signals , but also introduces a new and critical diagnostic finding for Prediction Error responses: For unexpected stimuli , increased sensory detail should improve the amount of sensory information contained in neural patterns . However , for stimuli that match expectations , increased sensory detail should lead to a reduction in the amount of information represented . Future studies in other sensory modalities and domains might benefit from adopting similar methods . Our work joins a number of recent fMRI and MEG/EEG studies in proposing an important role for Prediction Error computations in speech perception [4 , 7 , 8 , 39 , 81] . In these earlier studies , the observation of decreased activation for expected stimuli in the STG has been interpreted as a neural correlate of reduced Prediction Error and , hence , as evidence for Predictive Coding theories . However , almost all established computational theories of speech perception can also explain this observation . For example , TRACE [34] implements a form of neural sharpening in which prior knowledge enhances the representation of expected sensory signals and suppresses sensory noise , producing a reduced neural response overall . Similar , interactive activation models [35–38] might predict exactly the same decrease in STG activity for expected stimuli , as observed in these previous neuroimaging studies . Thus , existing empirical evidence proposed for Predictive Coding is also largely consistent with Sharpening theories . Even our previous comparison of Predictive Coding and Lexical Competition accounts of spoken word recognition [39] challenged the competitive lexical selection mechanism implemented in TRACE , but did not test the Sharpening mechanism traditionally described as Interactive Activation . In this context , then , the results of our study have important implications for understanding speech perception , a domain in which the presence and function of top-down processes has been much debated [82 , 83] . By directly quantifying the information represented in multivariate signals during perception of correctly expected and unexpected speech , we provided evidence that the neural mechanisms underlying speech perception are in line with Prediction Error simulations . Prior knowledge of speech content is used to explain away sensory evidence such that speech representations encode Prediction Error . The present multivariate interaction of sensory detail and prior information supports a Predictive Coding theory for how matching expectations improve perception of degraded speech . In contrast , enhanced representation of attended compared to unattended speech supports Sharpening mechanisms [84–87] . These findings could be reconciled by theories proposing that expectation ( Prediction Error ) and attention ( Sharpening ) operate in parallel , as suggested in some Predictive Coding theories [3 , 88] . However , more detailed computational specification of attentional mechanisms will be required to test these theories with experimental data . Comparing neural representations of attended and unattended speech signals at varying levels of expectation and degradation may be informative . There are three reasons why our results are of general interest for the study of speech and other domains of perception . One key aspect of our approach is that we assessed the perception of speech presented at varying levels of signal degradation . As in accounts proposing Bayesian perceptual inference [89] , this provides the best opportunity to observe influences of prior knowledge on perception . In doing so , we also test the perception of speech in listening conditions similar to the way that speech is most often heard in the real world [90] . A second form of generality is that prior expectations for speech were derived from written text . Our results may therefore also inform other situations in which prior knowledge and sensory information are combined across different modalities for speech [91–93] and other cross-modal stimuli [94–96] . Third and perhaps most important , however , is that the representations of Prediction Error that we have observed during speech perception might apply to many other sensory domains in which prior knowledge has been shown to influence perception ( such as audition [6 , 7 , 76 , 97] , vision [9–12 , 20 , 98 , 99] , touch [13] , gustation [14 , 100] , olfaction [15] , and pain [16] ) . The interactive effect of prior knowledge and sensory input on neural representation of degraded stimuli provides a stronger test of Predictive Coding theories of perception than has been provided by existing methods , as it offers the potential to challenge alternative views based purely on Sharpening mechanisms . In summary , the present results show that both increased sensory detail and matching prior expectations improved accuracy of word report for degraded speech but had opposite effects on speech coding in the posterior STS . Following neutral text , increased sensory detail enhanced the amount of speech information , whereas matching prior expectations reduced the amount of measured information during presentation of clearer speech . These findings support the view that the brain reduces the expected and , therefore , redundant part of the sensory input during perception , in line with representations of Prediction Error proposed in Predictive Coding theories .
Ethical approval was provided by Cambridge Psychology Research Ethics committee ( CPREC ) under approval number 2009 . 46 . All participants provided their written informed consent . Twenty-five healthy native-English speakers ( aged 18–40 , with self-reported normal hearing and language function ) participated in the experiment . Three participants had to be excluded because they were insufficiently attentive to the written text during the scanning runs ( they reported less than 50% of the written words correctly when prompted ) . One additional participant had to be excluded due to technical problems . The reported analyses are therefore based on 21 participants ( mean age 25 y [range 19 to 38 y] , 9 females ) . Word stimuli consisted of 24 different monosyllabic words , each with a consonant-vowel-consonant structure . The words were selected as eight triples of three similar words , each sharing the same vowel and with offset and onset changes between items ( eight triples: thing/sing/sit , bath/path/pass , deep/peep/peak , pork/fork/fort , doom/tomb/tooth , take/shake/shape , kite/tight/type , zone/moan/mode ) . These stimuli were recorded by a male native speaker of Southern British English and noise-vocoded ( 4- and 12-channel ) using custom scripts written in Matlab [59] . The syllables were filtered into 4 or 12 approximately logarithmically spaced frequency bands from 70 to 5 , 000 Hz [101] , with each pass band 3 dB down with a 16 dB/octave roll off . In each band , envelopes were extracted using half wave rectification , and pitch synchronous oscillations above 30 Hz were removed with a second-order Butterworth filter . The resulting envelopes were multiplied with a broadband noise and then band pass filtered in the same frequency ranges as the source and recombined . To ensure that acoustic intensity was matched across all stimuli , the RMS amplitude of each sound file was equalised . Finally , we applied an additional filter to ensure a flat frequency response when the spoken words were presented via Sensimetrics insert headphones in the scanner ( http://www . sens . com ) . Participants read written words and listened to subsequently presented degraded spoken words ( see Fig 2 ) . There were four conditions containing different pairings of written and spoken words: ( 1 ) matching written text + spoken words ( “SING” + sing ) ; ( 2 ) neutral written text ( “XXXX” ) + spoken words ( sing ) ; ( 3 ) partially mismatching written text + spoken words ( “SIT” + sing ) ; ( 4 ) totally mismatching written text + spoken words ( “SING” + doom ) . In addition , we included a fifth condition in which only written text ( “SING” ) was presented to test whether participants attended to the written words . Only the match and neutral conditions ( condition 1 and 2 ) were repeated sufficiently ( six presentations per item per condition ) to permit multivariate RSA ( see below for details ) . In occasional catch trials , a response cue , which consisted of a visual display of a question mark , was presented 1 , 000 ms after trial onset . This cued participants to say aloud the written or spoken word that they saw or heard previously . This design does not allow the analysis of response times , because participants were cued to respond after a delay . A previous behavioural study in our lab showed that response times for reporting vocoded spoken words are uninformative even when collected in such a way as to permit response time analyses [102] . The partial and total mismatch conditions ( condition 3 and 4 ) were included to make sure that participants paid attention to both the written and the spoken word; these conditions ensured that they could not simply report the preceding written word . Due to the small number of trials , RSA analysis was not possible for neural responses measured in the Mismatch condition . We can , however , report behavioural and univariate fMRI results for the Mismatch condition; this confirms that behavioural and neural enhancement following matching written text is not due to prestimulus attention or anticipation ( because prestimulus processes will be identical following mismatching text but enhanced perception is not typically observed ) [8 , 33] . Trials commenced with presentation of a fixation cross ( 1 , 000 ms ) , followed by presentation of a written word ( 500 ms ) , again followed by a fixation cross ( 500 ms ) , and finally the presentation of a spoken word . Written cues ( i . e . , written words , neutral “XXXX” , and fixation cross ) were presented in grey in the centre of the black screen . Trials were 3 to 9 s long , depending on the number of inserted null events to decorrelate the events within each run ( 76 trials of 3 s without null event , 45 trials of 6 s with a null event of 3 s , and 15 trials of 9 s with a null event of 6 sec , resulting in 211 TRs per run with null events ) . Spoken words were presented after 4- or 12-channel noise-vocoding to produce two different levels of sensory detail in the speech input . Altogether , this resulted in 816 trials , including 1/6 catch trials ( 136 trials ) in which participants had to give their verbal response ( 24 neutral and 24 match words x 6 repetitions x 2 levels of sensory detail = 576 trials , 24 written-only words x 6 repetitions = 144 trials , 24 partial mismatch and 24 total mismatch words x 2 levels of sensory detail without repetition on the word level = 96 trials; i . e . , 11 . 8% of the trials contained mismatching information ) . These trials were split into 6 runs of 136 trials each , ensuring that each word in each condition occurred once in each scanning run . With additional catch trials , each run took 11 . 7 min , and the overall experiment lasted approximately 70 min for all 6 runs . Stimulus delivery was controlled and behavioural responses were recorded with E-Prime 2 . 0 software ( Psychology Software Tools , Inc . ) . Verbal responses recorded in the scanner were transcribed by two independent raters ( the first author and a native English speaker with a PhD in phonetics who was naïve to the stimulus set ) and disagreements adjudicated by a third rater ( the senior author ) . All raters were blind to which word and stimulus condition was presented in each trial . Responses were scored for whole-word accuracy and analysed using Matlab . Because the percent correct performance scores were bound to [0;1] , we applied an arcsine transformation [103] before we computed a two-way repeated measures ANOVA and the corresponding post-hoc pared t tests . Data were analysed using SPM8 ( http://www . fil . ion . ucl . ac . uk/spm ) applying automatic analysis ( aa ) pipelines [104] . The first three volumes of each run were removed to allow for T1 equilibrium effects . Scans were realigned to the first EPI image . The structural image was coregistered to the mean functional image and the parameters from the segmentation of the structural image were used to normalise the functional images , which were resampled to 2 mm isotropic voxels . The realigned normalised images were then smoothed with a Gaussian kernel of 8 mm full width half maximum . Data were analysed using the general linear model with a 128 s high pass filter . We included the onset of 11 event types in the GLM , each convolved with the canonical SPM haemodynamic response: eight conditions come from specifying the onset of spoken words paired with four types of written text ( matching , neutral , partially mismatching , and totally mismatching ) crossed with two types of vocoding ( 4- and 12-channel ) . We also specified onsets for written words and neutral strings ( “XXXX” ) as well as the onset of the visual task cue that instructed participants to say the spoken word . Following parameter estimation of the first level model , we conducted a repeated measures ANOVA with two factors: prior knowledge ( matching versus neutral text ) and level of sensory detail ( 4- versus 12-channel ) to assess the main effects and interaction of these factors . We were interested in the effect of hearing speech that matches prior expectations on BOLD responses in the left posterior STS . To locate these ROIs for the multivoxel RSA ( see below ) , we tested for a main effect of prior knowledge ( F-contrast “Neutral versus Match” ) and identified a cluster at p < 0 . 05 FWE voxel-corrected in the left posterior STS . Multivariate analyses were conducted on realigned data within each participant’s native space without normalisation or spatial smoothing . An additional first-level model was constructed for each participant that contained the same set of regressors as the first level model used for the univariate analysis , except that regressors for individual spoken words were used in each of the four conditions for which there were sufficient numbers of repetitions for item-specific modelling ( 4- and 12-channel vocoded words following neutral or matching text ) . This resulted in 103 conditions per participant per run: 24 words for each of these four conditions and the remaining seven conditions from the univariate model . For each of the 96 item-specific regressors in these four conditions , we estimated single-subject T-statistic images for the contrast of speech onset compared to the unmodelled resting period , averaged over the six scanning runs . We used the resulting single condition and item T-images for RSA [50] using the RSA toolbox [52] . We used T-images so that effect sizes were weighted by their error variance , which reduces the influence of large but variable response estimates for multivariate analyses [105] . RSA involves testing whether the observed similarity of brain responses in specific conditions ( a neural RDM ) corresponds to a hypothetical pattern of similarity between these conditions ( hypothesis RDM ) . We constructed four hypothesis RDMs to test for greater similarity between syllable pairs within the same stimulus triple ( i . e . , syllables that shared the same vowel and had similar onset or offset segments like “sing” and “thing , ” as compared to dissimilar syllables like “sing” and “bath” ) within each of four critical conditions: Match 4-channel , Neutral 4-channel , Match 12-channel , and Neutral 12-channel . The design of our experiment was motivated by previous work that showed that STS encodes vowel and syllable similarity [55 , 61] , rather than spectrotemporal acoustic cues [61] . The comparisons used in our ROI analysis test for global similarity in representations of the phonetic form of similar-sounding spoken words because multiple consonantal features as well as the vowel are preserved within each syllable triple ( e . g . , bath/path/pass ) . We chose to analyse similarity of neural representations for phonetically similar but non-identical words for two reasons: ( 1 ) this approach allowed us to merge all six runs into a single analysis , which reduced the noise in the estimation of the T-images relative to a split-half method , and ( 2 ) comparing similar but non-identical word pairs makes our method insensitive to other forms of lexical or semantic similarity that could lead to similar neural representations for identical word pairs ( e . g . , in regions that code for word meaning [106] ) . Similarity between items in different conditions and between identical items ( i . e . , the main diagonal ) was therefore not included in our hypothesis RDMs ( see Fig 4A ) . We measured multivoxel RDMs by computing the dissimilarity ( 1–Pearson correlation across voxels ) of T-statistics for a specific item and condition . In a searchlight analysis , the sets of voxels were extracted by specifying grey-matter voxels ( voxels with a value > 0 . 33 in a probabilistic grey-matter map ) within an 8-mm radius sphere of each grey matter voxel ( with a voxel size of 3 x 3 x 3 . 75 mm , i . e . , a maximum of 65 voxels per sphere ) . This was repeated for all searchlight locations in the brain . The similarity between the observed RDM and each of the hypothetical RDMs was computed using a Spearman correlation for each searchlight location , and the resulting correlation coefficient returned to the voxel at the centre of the searchlight . This resulted in a Spearman correlation map for each participant in each grey matter voxel . To assess searchlight similarity values across participants at the second level , the Spearman correlation maps for each participant were Fisher-z-transformed to conform to Gaussian assumptions , normalized to MNI space , and spatially smoothed with a 10-mm FWHM Gaussian kernel for group analysis . These second-level analyses used a within-subject analysis of variance similar to those used for the univariate fMRI analysis . We used two computational implementations of Sharpened Signal and Prediction Error models of spoken word recognition ( using update mechanisms based on [75] ) , to simulate observed behavioural performance ( i . e . , word recognition ) , univariate fMRI results ( the magnitude of hemodynamic activity in the STS ) , and RSA fMRI results ( the similarity of representations for word pairs in the left posterior STS ) in each of our four experimental conditions . The sensory representations supplied at the input , the output lexical representations , and the specification of matching or neutral prior knowledge was identical for both simulations . We used a localist lexical representation ( i . e . , a set of 24 units , each of which was activated to represent a single word ) , as in previous models of spoken word recognition such as TRACE [34] or Shortlist [107] . The input to the model was provided as a distributed set of phonetic features ( derived from [108] ) . These are similar to the acoustic/phonetic features supplied as the input to TRACE or in recurrent network simulations such as the Distributed Cohort Model [109] . However , to avoid the complexity of representing temporal information ( and given the slow haemodynamic responses measured by fMRI ) , we assumed that speech information is provided in parallel over three groups of units for the initial consonant , medial vowel , and final consonant of our CVC words . The key difference between the Sharpened Signal and Prediction Error models concerns the computations by which prior knowledge is combined with degraded sensory representations of expected spoken words . In the Sharpened Signal simulation , expected sensory features receive additional activation through increased sensory gain [19 , 20] , whereas in the Prediction Error model , prior expectations contribute to perception by subtracting expected input from sensory representations ( i . e . , computation of Prediction Error [3 , 23 , 24] ) . In both simulations , an iterative settling procedure was used such that feature representations of the input are combined with prior knowledge to generate feature representations that convey Sharpened Signals or Prediction Errors respectively ( hereafter “sharpened features” and “prediction error features” ) . These representations were used to update lexical activations , and updated lexical activations in turn led to modified top-down expectations . This settling procedure continued until a settling criterion was reached or a maximum number of iterations had been performed . | Perception inevitably depends on combining sensory input with prior expectations . This is particularly critical for identifying degraded input . However , the underlying neural mechanism by which expectations influence sensory processing is unclear . Predictive Coding theories suggest that the brain passes forward the unexpected part of the sensory input while expected properties are suppressed ( i . e . , Prediction Error ) . However , evidence to rule out the opposite mechanism in which the expected part of the sensory input is enhanced or sharpened ( i . e . , Sharpening ) has been lacking . In this study , we investigate the neural mechanisms by which sensory clarity and prior knowledge influence the perception of degraded speech . A univariate measure of brain activity obtained from functional magnetic resonance imaging ( fMRI ) is in line with both neural mechanisms ( Prediction Error and Sharpening ) . However , combining multivariate fMRI measures with computational simulations allows us to determine the underlying mechanism . Our key finding was an interaction between sensory input and prior expectations: for unexpected speech , increasing speech clarity increases the amount of information represented in sensory brain areas . In contrast , for speech that matches prior expectations , increasing speech clarity reduces the amount of this information . Our observations are uniquely simulated by a model of speech perception that includes Prediction Errors . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"syllables",
"linguistics",
"medicine",
"and",
"health",
"sciences",
"diagnostic",
"radiology",
"functional",
"magnetic",
"resonance",
"imaging",
"engineering",
"and",
"technology",
"signal",
"processing",
"social",
"sciences",
"neuroscience",
"magnetic",
"resonance",
"imaging",
"simulation",
"and",
"modeling",
"brain",
"mapping",
"computational",
"neuroscience",
"neuroimaging",
"coding",
"mechanisms",
"research",
"and",
"analysis",
"methods",
"imaging",
"techniques",
"speech",
"psychology",
"radiology",
"and",
"imaging",
"speech",
"signal",
"processing",
"diagnostic",
"medicine",
"phonology",
"biology",
"and",
"life",
"sciences",
"sensory",
"perception",
"computational",
"biology"
] | 2016 | Prediction Errors but Not Sharpened Signals Simulate Multivoxel fMRI Patterns during Speech Perception |
Candida albicans has an elaborate , yet efficient , mating system that promotes conjugation between diploid a and α strains . The product of mating is a tetraploid a/α cell that must undergo a reductional division to return to the diploid state . Despite the presence of several “meiosis-specific” genes in the C . albicans genome , a meiotic program has not been observed . Instead , tetraploid products of mating can be induced to undergo efficient , random chromosome loss , often producing strains that are diploid , or close to diploid , in ploidy . Using SNP and comparative genome hybridization arrays we have now analyzed the genotypes of products from the C . albicans parasexual cycle . We show that the parasexual cycle generates progeny strains with shuffled combinations of the eight C . albicans chromosomes . In addition , several isolates had undergone extensive genetic recombination between homologous chromosomes , including multiple gene conversion events . Progeny strains exhibited altered colony morphologies on laboratory media , demonstrating that the parasexual cycle generates phenotypic variants of C . albicans . In several fungi , including Saccharomyces cerevisiae and Schizosaccharomyces pombe , the conserved Spo11 protein is integral to meiotic recombination , where it is required for the formation of DNA double-strand breaks . We show that deletion of SPO11 prevented genetic recombination between homologous chromosomes during the C . albicans parasexual cycle . These findings suggest that at least one meiosis-specific gene has been re-programmed to mediate genetic recombination during the alternative parasexual life cycle of C . albicans . We discuss , in light of the long association of C . albicans with warm-blooded animals , the potential advantages of a parasexual cycle over a conventional sexual cycle .
In most sexually reproducing eukaryotes , meiosis is used to precisely halve the DNA content in the cell , often for the formation of haploid gametes from diploid precursor cells . This specialized form of cell division involves one round of DNA replication followed by two successive rounds of DNA division . Each round of DNA division is unique . During the first meiotic division ( meiosis I ) extensive DNA recombination takes place between maternal and paternal homologous chromosomes , which then are segregated from one another . The second round of DNA division ( meiosis II ) more closely resembles normal mitotic DNA division , in which sister chromatids are segregated to opposite poles . In the case of spores in fungi and spermatozoa in animals , all four haploid nuclei form four different haploid cells , while in the female meioses of animals only one haploid nucleus survives and forms the mature oocyte . The meiotic process has been studied extensively in the model fungi Saccharomyces cerevisiae and Schizosaccharomyces pombe . In S . cerevisiae , mating of haploid MATa and MATα cells normally generates a stable diploid a/α cell that replicates mitotically until subsequently induced to undergo meiosis under conditions of limiting nitrogen availability and the presence of a non-fermentable carbon source [1] . In S . pombe , mating also occurs between haploid cells but the diploid state is often transient , immediately undergoing meiosis to regenerate the haploid form . The sexual program in S . pombe is again controlled by nutritional cues , as mating and meiosis normally occur only under starvation conditions [2] . In both S . cerevisiae and S . pombe , meiosis generates four recombinant haploid spores held together in an ascus . While S . cerevisiae and S . pombe are rarely pathogenic in humans , the related ascomycete C . albicans is an opportunistic pathogen capable of causing both debilitating mucosal infections and potentially life-threatening systemic infections [3] . C . albicans is normally a harmless commensal fungus , existing in the gastrointestinal tract of at least 70% of the healthy population [4] . However , C . albicans is also the most commonly isolated fungal pathogen , particularly targeting individuals with compromised immune systems and leading to death in up to 50% of patients with bloodstream infections [5–7] . Until recently , C . albicans was thought to be asexual , existing only as an obligate diploid organism and thus classified amongst the Fungi imperfecti [8] . However , a robust mating system has now been uncovered in this organism , in which mating occurs between diploid mating type-like ( MTL ) a and α strains to generate an a/α tetraploid strain . Mating occurs both under laboratory conditions and in different in vivo niches in a mammalian host [9–12] . Population studies of clinical isolates are also consistent with C . albicans strains undergoing genetic exchange in their natural environment , albeit at a limited rate [13] . While an efficient mating apparatus has now been identified in C . albicans , the mating cycle differs in several important respects from that of S . cerevisiae and other fungi . For example , mating in C . albicans is regulated by phenotypic switching; MTL homozygous C . albicans cells can reversibly switch between two heritable states termed white and opaque , and only the opaque form is competent for efficient mating [14] . This unusual mode of mating regulation is so far unique to C . albicans ( and the very closely related yeast , Candida dubliniensis [15] ) making it likely that this adaptation has evolved to regulate mating of C . albicans strains in their natural environment—that of a warm-blooded host . Completion of the mating cycle in C . albicans also seems to occur in an atypical manner . Although reductional DNA divisions by a meiotic program have not been observed , tetraploid strains of C . albicans have been shown to return to the diploid state via a parasexual mechanism . During this process , tetraploid cells exposed to certain laboratory media were induced to lose chromosomes in an apparently random , but concerted , fashion , thereby forming cells with a diploid , or very close to diploid , DNA content [16] . The genetic locus responsible for determining C . albicans mating type ( the MTL locus ) segregated randomly in these experiments so that many of the progeny cells were a and α diploid cells that were themselves mating competent . Mating of diploid cells to form tetraploid cells , followed by random chromosome loss to generate diploid progeny cells , thereby constitutes a parasexual mating cycle in C . albicans . In this study , we examined the genetic profile of strains formed by the parasexual mating process in C . albicans using SNP and comparative genome hybridization ( CGH ) techniques . We observed extensive shuffling of the parental configurations of chromosomes by the parasexual cycle , giving rise to many types of recombinant C . albicans progeny . Many of the progeny strains are not true ( euploid ) diploids; rather , they are aneuploid strains that are often trisomic for one or more chromosomes . In addition , we provide the first evidence that tetraploid strains experiencing chromosome instability and subsequent chromosome loss also undergo genetic recombination between homologous chromosomes . We also report that genetic recombination in C . albicans tetraploids was dependent on the presence of Spo11p , a conserved protein that in other eukaryotes initiates meiotic recombination by the introduction of double-strand breaks ( DSBs ) into the DNA [17] . These results suggest that the parasexual pathway in C . albicans has evolved as an alternative pathway to meiosis for promoting a reduction in cell ploidy , and furthermore , that at least one gene that normally functions in meiotic recombination has been co-opted for use in the parasexual mating cycle .
The parasexual cycle of C . albicans , as currently envisaged , is shown in Figure 1A . Note that no meiotic program has been observed in C . albicans , despite the presence of many genes in the genome whose homologues function specifically in meiosis in other fungi [18] . However , C . albicans strains have been found to undergo a parasexual cycle; tetraploid strains become genetically unstable when incubated on certain laboratory media , losing chromosomes and generating diploid ( and aneuploid ) progeny strains that are themselves mating competent . The chromosome loss process is concerted , with loss of one or more chromosomes predisposing the cell to lose additional chromosomes , and the diploid state being the final product [16] . While tetraploids are stable when grown on YPD medium at different temperatures , two culture conditions were identified that induced genetic instability in C . albicans: ( i ) growth of tetraploid strains on S . cerevisiae “pre-sporulation” ( pre-spo ) medium at 37 °C , and ( ii ) growth of tetraploid strains on medium containing L-sorbose at 30 °C . The latter condition was previously shown to also induce chromosome loss in diploid C . albicans strains [19] . More specifically , diploid strains were unable to grow on L-sorbose medium unless they first underwent loss of one copy of Chromosome ( Chr ) 5 , becoming monosomic for this chromosome . In contrast , diploid strains were relatively stable when grown on pre-spo medium , indicating that diploid and tetraploid strains exhibit very different selective pressures when cultured on this medium . To monitor changes in ploidy in tetraploid strains of C . albicans , we exploited a genetically marked tetraploid strain , RBY18 , containing markers on Chr 1 and 5 . The strain was constructed by mating a/Δα and Δa/α cell types , as shown in Figure 1B [16] . Strain RBY18 is heterozygous for the GAL1 gene 1 on Chr 1 , which is counterselectable . Strains carrying wild-type GAL1 are unable to grow on medium containing 2-deoxygalactose ( 2-DOG ) as the carbon source , while derivative strains that have lost both copies of the GAL1 gene are able to grow on 2-DOG medium [20] . In most cases , it is expected that loss of GAL1 function in RBY18 will occur by loss of both chromosomes carrying the GAL1 allele , although GAL1 function also can be lost by mutation or genetic recombination . The RBY18 tetraploid strain is also heterozygous for all four MTL alleles on Chr 5: WTa , WTα , Δa1/a2 , and Δα1/α2 , which are easily distinguishable using whole cell PCR and oligonucleotides specific to each MTL allele [14] . To generate progeny strains that have undergone the parasexual mating cycle , the marked tetraploid strain RBY18 was induced to undergo chromosome loss on either pre-spo or sorbose medium and gal1− strains were selected by growth on 2-DOG medium . These 2-DOG resistant ( DOGR ) strains were subsequently analyzed by PCR to confirm that loss of MTL alleles on Chr 5 had accompanied loss of GAL1 alleles on Chr 1 , an indication that cells had undergone a reduction in overall cell ploidy ( unpublished data ) . PCR of the MTL loci was also used to detect possible jackpot effects , where several gal1− progeny might have been derived from a single cell having undergone a chromosome loss event . Where possible , progeny cells with different combinations of MTL alleles were used for subsequent analysis . Selected progeny strains were grown in YPD medium at 30 °C and analyzed by flow cytometry to determine the overall ploidy of each strain , as shown in Figure 2 . Flow cytometric analyses confirmed that each strain was diploid , or close to diploid , in DNA content , as judged by staining of the DNA with sytox green [9] . Seven strains ( P1 to P7 ) were derived from RBY18 by growth on pre-spo medium , and six strains ( S1 to S6 ) were derived from RBY18 by growth on sorbose medium ( Figure 2 ) . Subtle differences were observed in the flow cytometry DNA profiles between isolates , where distinct peaks were evident representing non-replicated ( G1 phase ) and replicated ( G2 phase ) DNA . In some strains the majority of the cells contained replicated DNA ( e . g . , S5 and S6 , Figure 2 , panels N and O ) , while others had an almost equal distribution of cells with unreplicated and replicated DNA ( e . g . , P4 , panel F ) . However , there was no obvious correlation between DNA profiles analyzed by flow cytometry and cell growth rates . To further characterize the strains generated by parasexual chromosome reduction , progeny were plated for single colonies on rich ( YPD ) medium to examine colony growth . After incubation at 30 °C for 7 d , colonies were compared for overall size and morphology ( Figure 3 ) . A wide range of phenotypes was observed , including smaller colony sizes relative to diploid and tetraploid parental strains and altered colony morphologies . Some of the isolates produced hyper-filamentous morphologies , as evidenced by increased surface wrinkling of the colonies ( e . g . , progeny strains P3 , P4 , and P6; Figure 3 , panels E , F , and H ) . Normally , C . albicans cells grow as budding yeast , pseudohyphal , or true hyphal cells . Examination of cells from the wrinkled colonies by microscopy confirmed that these colonies contained many filamentous ( pseudohyphal and true hyphal ) cells , while the unwrinkled colonies ( including control strains ) contained very few filamentous cells ( unpublished data ) . Some progeny strains also exhibited reduced filamentation on medium that normally induces hyphae formation ( Spider medium and serum-containing medium , KA and RJB , unpublished data ) . Thus , the parasexual cycle of C . albicans can generate variant strains with diverse colony morphologies . Changes in the ability to undergo the yeast-hyphal transition have been closely linked with the pathogenic potential of C . albicans strains [21–24] . It is therefore likely that many of these variant strains will exhibit reduced virulence in models of candidiasis; but it is also possible that some of these isolates could have increased fitness under particular selective conditions , leading to improved colonization of defined in vivo niches in the host . SNP and CGH microarrays are powerful approaches for examining genetic recombination and genome structure in C . albicans [25–28] . SNP arrays were designed to exploit the sequence diversity between chromosome homologues in the diploid C . albicans genome . The genome-wide SNP arrays used here included 152 SNPs , distributed across all eight chromosomes of C . albicans . As each SNP is specific for one of the parental chromosome homologues , each homologue can be distinguished in progeny from the parasexual mating cycle . In addition , loss of heterozygosity ( LOH ) at SNPs on otherwise heterozygous chromosomes can be used as a marker for genetic recombination . Quantitative SNP analysis can also be used to determine the relative copy number of each homologue in a sample ( see Materials and Methods ) . CGH analysis provides a complementary approach to SNP arrays for the determination of the copy number of each gene on each chromosome in the sample . Labeled genomic DNA from experimental samples ( Cy3 labeled ) and labeled DNA from a reference diploid SC5314 strain ( Cy5 labeled ) were hybridized to whole genome arrays containing >6 , 000 C . albicans ORFs [27 , 28] . CGH data provides information on the copy number of every chromosome , as well as indicating large-scale aneuploidies . In this study , we used both CGH and SNP approaches to obtain a detailed picture of the products of the C . albicans parasexual cycle following concerted chromosome loss . SNP and CGH arrays were first used to analyze RBY18 and the diploid parental strains that had been used to construct this tetraploid strain . SNP analysis confirmed that MTLa and MTLα parental diploid strains were heterozygous for most of the SNPs on the array , although in the parental MTLα strain Chr 2 was homozygous for all markers ( Table S4 ) . CGH array data confirmed that the parental strains were euploid diploids and RBY18 was a euploid tetraploid , as they contained two and four copies of each of the eight C . albicans chromosomes , respectively . We then analyzed 13 progeny strains produced by concerted chromosome loss from RBY18 using SNP and CGH arrays ( see Figures 4 and S4 , and Tables S1 and S4 ) . Only three of the 13 strains were true diploids ( P2 , P5 , and P6 ) . The majority ( 10/13 ) of the progeny strains contained at least one extra chromosome: four of the seven strains derived from growth of the tetraploid on pre-spo medium were trisomic for one to three chromosomes and all six strains derived from growth on sorbose were also trisomic for up to three of the eight C . albicans chromosomes ( Figure 4 ) . Thus , concerted chromosome loss was often incomplete and did not immediately result in true diploid strains . Curiously , there was a strong bias towards trisomy of Chr 4 in the progeny strains; all strains carrying at least one trisomic chromosome ( four pre-spo-selected strains and all sorbose-selected strains ) were trisomic for Chr 4 . Trisomies of Chr R , 2 , 5 , 6 , or 7 were also detected in at least one of the progeny . As expected , Chr 1 was always present in the disomic parental configuration ( one copy of each homologue ) because selection of DOGR progeny requires that the strains lose both Chr 1 homologues from the MTLα mating parent ( Figure 1B ) . The most striking feature of the progeny genetic profiles was that three strains contained a number of short LOH tracts ( six or seven LOH tracts were observed in each strain ) , evidence of multiple recombination events between homologous chromosomes . Isolates P1 , S3 , and S4 exhibited recombination events that included LOH at SNPs on multiple chromosomes ( including Chr R , 1 , 2 , 4 , 5 , 6 , and 7 , see Figure 4 ) . While selection on 2-DOG required inheritance of the gal1Δ alleles on Chr 1 , the LOH events detected here are independent of the GAL1 locus . Moreover , these events did not involve homozygosis of all of Chr 5 , which might be expected to occur in response to sorbose selection . Instead , the recombination events we observed appear to be selection independent . Overall , the appearance of multiple gene conversion tracts within several strains , and the general absence of gene conversion tracts in other strains , suggests that some cells become generally competent for recombination at more than one locus , while other strains do not undergo such recombination events at all . In at least one example ( Chr 2 in strain P1 ) one complete chromosome arm ( Chr 2L ) became homozygous ( Figure 4 ) . This recombination event may have arisen in one of two ways: ( i ) A cross-over between chromosomes led to reciprocal recombination between homologues , as commonly occurs during meiosis in other fungi . In this case , the partner DNA involved in the reciprocal exchange was lost during the process of concerted chromosome loss . ( ii ) A break-induced replication event occurred . In this case , a DSB in one chromosome was repaired by DNA replication that copied the template strand from the break near the centromere all the way to the telomere in the homologous chromosome . Break-induced replication is a non-reciprocal recombination event and in S . cerevisiae is often restricted to repair of DNA DSBs where only one end of the break shares homology with the template [29] . Potential hotspots for recombination were identified in the three strains that had undergone inter-homologue recombination . For example , SNPs HST3 and 2340/2493 on Chr 5 underwent LOH in P1 , S3 , and S4 recombinant strains . Additional experiments are necessary to fully document hotspots for recombination . However , our results indicate that recombination events are not uniform across the C . albicans genome during the parasexual cycle . Natural isolates of C . albicans are diploid , and it has been proposed that haploid forms cannot exist because of the presence of recessive lethal alleles in the genome . Evidence supporting this idea came from classical mitotic recombination studies [30 , 31]; however , no systematic investigation of possible recessive lethal alleles in the C . albicans genome has been reported . Using the present dataset , we can rule out the presence of recessive lethal alleles on some chromosomes . For example , it was already known that Chr 5 does not harbor recessive lethal alleles: loss of either homologue can be induced in diploid cells by growth on sorbose medium [19] . The SNP data presented here supports this finding , as both AA and BB configurations of Chr 5 homologues were observed in the progeny strains P2 and P6 , respectively ( this nomenclature assigns the parental configuration of chromosome homologues as AB ) . Similarly , several other chromosomes did not carry recessive lethal alleles , as their homologues could be lost during the parasexual cycle . Chr R , 2 , 3 , 5 , 6 , and 7 were all found to be homozygous in at least one independent isolate . However , only one homozygous configuration was observed for each chromosome ( either AA or BB ) , leaving open the possibility that the other chromosome homologue carries recessive lethal alleles . We will revisit the issue of recessive lethal alleles below . The C . albicans parasexual cycle provides an alternative mechanism to meiosis for a reduction in cell ploidy . Although no experimental evidence for a meiotic pathway in C . albicans currently exists , the genome contains homologues of many genes that function specifically in meiosis in the related yeast S . cerevisiae [18] . Some of the meiosis genes from C . albicans even complement for meiotic function in S . cerevisiae , demonstrating they encode a conserved protein activity [32] . It seems likely that either ( i ) C . albicans has a cryptic meiotic program still to be discovered , or ( ii ) meiotic genes have been adapted to other processes in C . albicans , perhaps some in the parasexual pathway . To address the latter possibility , we investigated the potential role of the Spo11 protein in genetic recombination during the parasexual cycle . In fungi such as S . cerevisiae and S . pombe and in higher eukaryotes , Spo11p makes meiosis-specific DSBs in DNA via a topoisomerase-like mechanism of DNA cleavage [33 , 34] . C . albicans ORF19 . 11071 on Chr 2 encodes a potential homolog of S . cerevisiae SPO11 ( http://www . candidagenome . org ) . An alignment of this ORF with SPO11 genes from diverse species including S . pombe , S . cerevisiae , Kluyveromyces lactis , and Drosophila reveals that several of the critical conserved residues identified for DNA strand cleavage are present in the C . albicans sequence ( Figure S1 ) . In particular , the conserved active site tyrosine residue , required for breakage of the DNA and formation of a phosphotyrosine bond , is present in the C . albicans protein . Similarly , Glu-233 and Asp-288 residues that are required in S . cerevisiae Spo11p for meiotic recombination [35] are conserved in the C . albicans protein . ORF19 . 11071 is a homologue of the Spo11 family and will therefore be referred to as C . albicans Spo11p in the rest of this study . Attempts to complement S . cerevisiae Spo11 function with C . albicans Spo11p , as measured by rates of meiotic recombination in return-to-growth experiments , were unsuccessful ( Table S2 ) . This result is perhaps not surprising as SPO11 sequences from diverged species are poorly conserved outside of the core catalytic residues [36] ( Figure S1 ) . It is also worth noting that meiotic proteins in general are faster evolving than most cellular proteins [37 , 38] , an issue that is taken up again in the Discussion . To investigate whether C . albicans Spo11p is expressed in mitotically dividing cells , a Spo11-13myc fusion protein was constructed in diploid C . albicans strains . Western blots show that the Spo11-13myc protein was detectable in mitotic extracts of diploid cells grown in YPD medium , although the level of expression was relatively low ( see comparison of protein levels with that of the mitotic spindle protein Kar3-13myc ) ( Figure 5 ) . Thus , in C . albicans , the Spo11 protein is expressed in mitotically dividing cells . The observation that C . albicans Spo11p is expressed during mitotic growth is consistent with it having a function outside of meiosis . To examine if C . albicans Spo11p is required for genetic recombination in the parasexual mating cycle , we deleted all four copies of the SPO11 gene in genetically marked tetraploid strains ( RBY176/RBY177 ) that were heterozygous for GAL1 on Chr 1 . The strains were induced to undergo concerted chromosome loss on pre-spo or sorbose medium and were then exposed to 2-DOG to select for strains that had lost both copies of GAL1 . Eighteen DOGR colonies were selected from tetraploid growth on pre-spo ( eight colonies ) or sorbose ( ten colonies ) and subsequently analyzed by flow cytometry to determine if they were diploid , or near diploid , strains ( Figure S2 ) . Indeed , we detected diploid Δspo11 progeny strains , indicating that Spo11p is not necessary for the process of concerted chromosome loss in tetraploid C . albicans strains . We next analyzed the colony morphologies of the Δspo11 diploid progeny . As was seen with progeny from wild-type tetraploids ( Figure 4 ) , many of the Δspo11 progeny strains exhibited altered colony morphologies on YPD medium ( Figure 6 ) . Genomic profiles of the Δspo11 diploid progeny ( along with the parental diploid and tetraploid strains ) were generated using SNP and CGH microarrays ( see Figure 7 , as well as Tables S5 and S6 , and Figures S1 and S4 ) . One of the diploid parents ( RBY79 , MTLα parent ) was initially homozygous for Chr 2 , and the other parent ( RBY77 , MTLa parent ) carried a long tract of LOH on Chr 2 ( Figure 7 ) . This is reflected in the patterns of Chr 2 inheritance in the diploid progeny which either received only one type of Chr 2 homologue ( Ps2 , Ps3 , Ps4 , Ps5 , Ps6 , Ss1 , Ss2 , Ss3 , Ss4 , Ss8 , and Ss10 ) or received two homologues that only differ near the Chr 2R telomere ( Ps1 , Ps7 , Ps8 , Ss5 , Ss6 , Ss7 , and Ss9 ) . Similarly , one of the gal1Δ Chr 1 homologues in the parental MTLa strain had undergone LOH of a single SNP near the telomere of Chr 1L and this LOH tract was retained in all of the progeny . As in the wild-type ( SPO11+ ) progeny that were close to diploid , a majority ( 11/18 ) of the strains carried at least one and up to three trisomies , and Chr 4 was often one of the trisomic chromosomes ( 5/11 strains ) . Other chromosomes that became trisomic were Chr R , Chr 1 , Chr 2 , Chr 5 , Chr 6 , and Chr 7 . The only chromosome that did not become trisomic in these strains or in the wild-type diploid progeny strains was Chr 3 . Concerted chromosome loss did result in homozygosis of Chr R in nine strains ( and trisomy in one strain ) with the same homologue always being retained ( the blue-colored “A” homologue in Figure 7 ) . Interestingly , while no trisomies of Chr 3 were found , Chr 3 underwent LOH in ten strains , with seven of them retaining homologue B ( colored pink ) and three retaining the A homologue ( Figure 7 ) . The most striking feature of the Δspo11 progeny strains was that they did not undergo any detectable genetic recombination events . No single LOH events ( gene conversion events ) or chromosome crossing over events ( long-range LOH ) events were observed ( although we note that if reciprocal recombination events occurred , in which both recombinant chromosomes were retained , these would not be detected by SNP analysis ) . In contrast , progeny derived from SPO11+ strains exhibited multiple recombination events in three out of 13 strains ( Figure 4 ) , a difference that is statistically significant ( p < 0 . 05 ) . Taken together , the SNP and CGH experiments indicate that genetic recombination takes place in wild-type cells during the parasexual mating cycle , generating recombinant C . albicans strains . These recombination events are dependent on Spo11p , a conserved protein that normally acts specifically in meiosis in a wide range of eukaryotes . We suggest that Spo11p function has been adapted in C . albicans for mediating genetic recombination in the alternative parasexual mating cycle . The Spo11 experiment more than doubled the data on chromosome loss from C . albicans tetraploids , and we used this expanded dataset to re-evaluate patterns of chromosome loss during the parasexual cycle . We pooled genomic profiling data for all 31 progeny strains derived from tetraploids by concerted chromosome loss ( 13 from wild-type SPO11+ tetraploids and 18 from Δspo11 tetraploids ) ( Table 1 ) . We determined , for each chromosome in each strain , whether they existed as the parental configuration of homologues ( AB ) , a homozygous configuration ( AA or BB ) , or a trisomic configuration ( AAB or ABB ) . We excluded Chr 1 from this analysis , as selection for the loss of the GAL1 gene required the AB configuration be retained for Chr 1 in all isolates ( see Figure 1B ) . While the number of strains analyzed in this study is relatively small , several trends are apparent . First , if two of the four copies of each chromosome in tetraploid strains were lost with equal probability , it would be expected that 67% of chromosomes would consist of AB homologues , while 33% of chromosomes would exhibit either AA or BB configurations . Isolates selected from pre-spo medium contained a chromosomal distribution very close to this , with 72% of disomic chromosomes being AB homologues , and 28% of chromosomes being AA or BB homologues . In contrast , isolates derived from sorbose medium were biased towards a homozygous chromosome configuration ( 45% were AA or BB with only 55% exhibiting the AB configuration ) . Sorbose-selected strains were also more likely to contain trisomic chromosomes than were strains selected on pre-spo medium . Trisomic chromosomes were present for 24 . 3% of chromosomes selected on sorbose medium , while only 12 . 4% of chromosomes were trisomic in isolates selected on pre-spo medium . Both of these differences between pre-spo and sorbose media were significant ( p < 0 . 05 ) and provide evidence that strains undergoing chromosome loss on these media either experience different patterns of chromosome loss or different selective pressures . Previous studies have also observed differences between pre-spo and sorbose medium in that diploid C . albicans strains were stable on pre-spo medium but exhibited chromosome instability ( particularly that of Chr 5 but also of other chromosomes ) on sorbose medium [16 , 19 , 39] . One possibility for the higher fraction of homozygous AA/BB chromosomes in tetraploids exposed to sorbose medium is that these conditions generate monosomic chromosomes that then undergo re-duplication to form homozygous disomic chromosomes . At least for Chr 5 , this possibility was ruled out by PCR typing of MTL alleles on this chromosome , as all four MTL alleles in the tetraploid are distinct ( MTLa , MTLα , MTLΔa , MTLΔα; Figure 1B ) . PCR analysis revealed that strains that were homozygous for Chr 5 by SNP analysis always contained two distinct MTL alleles , indicating that monosomy and reduplication had not occurred ( unpublished data ) . These experiments demonstrate that , at least for tetraploid strains , the formation of viable progeny on sorbose medium does not require monosomy of Chr 5 at any stage . Trisomy was more common in strains derived from sorbose medium than pre-spo medium and could be due either to chromosome loss of one homologue from tetraploids or to re-duplication of one chromosome homologue in a disomic strain . Curiously , at least in a subset of cases , trisomy was a result of reduplication of one chromosome homologue in sorbose-derived strains . For example , three strains , S5 , Ss1 , and Ss9 , were shown to be trisomic for Chr 5 by SNP analysis and yet each strain contained only two types of MTL allele by PCR genotyping . This indicates that trisomy of Chr 5 arose by re-duplication of one homologue of Chr 5 in a disomic strain . In addition , one isolate ( strain Ss2 derived from sorbose medium ) was trisomic for Chr 1 , but was clearly gal1− by PCR ( lacked the GAL1 ORF ) and was also 2-DOG resistant . Thus , where trisomies can be distinguished in sorbose-derived strains , they were due to re-duplication of chromosome homologues for Chr 1 and Chr 5 . In contrast , strain Ps8 derived from pre-spo medium was trisomic for Chr 5 and also tri-allelic at the MTL locus , indicating that trisomy occurred via loss of one homologue of Chr 5 and not chromosome re-duplication . Overall , our results suggest that growth of tetraploids on sorbose medium may apply more selective pressure to the cells than growth on pre-spo medium , causing them to produce progeny with increased trisomies by chromosome re-duplication and more bias in the distribution of whole chromosome LOH events . A further conclusion from our analysis is that every chromosome ( excluding Chr 1 because of the selection for gal1Δ/Δ ) can exist in a homozygous form . Because of the limited number of strains analyzed , we only detected one homozygous configuration ( either the AA or BB configuration ) for most chromosomes , however both Chr 3 and Chr 5 were found in both the AA and BB configurations . This implies that Chr 3 , like Chr 5 , does not contain recessive lethal alleles on either chromosome homologue . One possible caveat to this conclusion is that undetected recombination events may have repaired recessive lethal alleles on these chromosome homologues . However , this seems unlikely given that the Δspo11 progeny produced a significant number of strains that were homozygous for both homologues of Chr 3; seven progeny were homozygous for the B homologue and three were homozygous for the A homologue . We detected trisomies for all chromosomes except Chr 3 in at least one progeny strain . This suggests that one or more genes on Chr 3 may not be well tolerated at higher than euploid copy number under these conditions . Recent studies in S . cerevisiae have shown that increased copy numbers of certain chromosomes can be lethal , as haploid cells disomic for Chromosome VI were inviable [40] . However , at least in the majority of cases , C . albicans strains trisomic for one or more chromosomes were viable and produced apparently stable karyotypes . Finally , a comparison of the growth rates of the progeny strains from the parasexual cycle was revealing . While most of the euploid progeny grew at rates very similar to that of a control diploid strain , aneuploid strains grew at increasingly slower rates as the number of trisomic chromosomes increased ( Figure S3 ) . Thus , euploid progeny grew on average 7 . 4% slower than a control SC5314 strain , while strains containing one trisomic chromosome grew 9 . 5% slower , strains containing two trisomies 16 . 3% slower , and strains with three trisomies 23 . 4% slower . Thus , as the number of additional chromosomes increased , so , in general , did the doubling time of the cell . In S . cerevisiae a similar observation has been made , where aneuploid chromosomes ( disomies in haploids or trisomies in diploids ) were found to cause a proliferative disadvantage , and this disadvantage generally increased as the number of extra chromosomes increased [40 , 41] . Aneuploidy therefore appears to confer a proliferative disadvantage in multiple yeast species .
The products of the C . albicans parasexual cycle were diploid and aneuploid progeny that exhibited altered colony morphology phenotypes . In particular , many strains had an increased tendency to form hyphal filaments on solid medium , evident either by increased surface wrinkling of the colony or by increased peripheral filamentation at the edge of the colony . Since the yeast-hyphal transition is closely associated with virulence of C . albicans strains , it is likely that many of these progeny strains will show altered virulence in animal models of candidiasis . Several of the progeny strains exhibited growth defects relative to control diploid strains , to control tetraploid strains , and to other diploid progeny . In some cases this was likely due to chromosomal aneuploidies , as many of these strains carried extra copies of up to three of the eight chromosomes of C . albicans . Indeed , being trisomic for two or three chromosomes increased cell doubling times by 16% and 23% , respectively , over a diploid control strain . Recent studies in S . cerevisiae have found that aneuploidy due to the presence of one or more additional chromosomes resulted in compromised growth rates [40] . Aneuploidy of large chromosomes or of multiple chromosomes correlated with the most significant cell cycle delays in S . cerevisiae [40] . We observed the same phenomenon in C . albicans strains , that the more aneuploid chromosomes a strain carries , the greater the proliferative disadavantage . Compromised growth was also observed in a subset of euploid progeny from the parasexual mating cycle . In general , it appeared that LOH on 1–2 chromosomes did not typically compromise growth rates , but that LOH across multiple chromosomes did ( e . g . , strains Ps7 and Ss10 ) . In these cases , it is likely that LOH at multiple genes led to the reduced fitness of these strains . Consistent with this idea , most clinical isolates , including the SC5314 strain whose genome was sequenced , show extensive heterozygosity . Previous studies have shown that allelic differences between C . albicans genes from different chromosome homologues can result in altered protein expression and altered protein function [51–54] . In addition , a recent study analyzed Chr 5 heterozygosity in multiple clinical isolates and found that LOH at multiple genes along Chr 5 reduced the virulence of strains in a model of systemic candidiasis [55] . Our work is also consistent with the idea that heterozygosity of multiple chromosomes provides C . albicans strains with a fitness advantage , at least for growth on laboratory media . Taken together , these results indicate that being heterozygous for genes on multiple chromosomes can improve both the fitness of mitotically dividing cells in vitro and the virulence of strains in vivo . However , the parasexual cycle ( including the recombination events described in this paper ) generates a great deal of genetic diversity , and it seems likely that conditions exist where strains that show reduced fitness in the laboratory have a selective advantage elsewhere . Recent work has revealed that C . albicans , like other prevalent human fungal pathogens such as Cryptococcus neoformans and Aspergillus fumigatus , has access to a sexual mating program , but that under most conditions it propagates primarily in an asexual manner [56–58] . Recent studies in the model yeast S . cerevisiae found that both asexual and sexual modes of reproduction can be advantageous under the right experimental conditions . Under constant environmental conditions , the asexual mode of propagation was favored , but under stressful conditions the sexual strain had the competitive advantage [59 , 60] . In the case of C . albicans strains , population genetics on clinical isolates first suggested that the predominant mode of reproduction was clonal , with only limited evidence for genetic recombination between strains [61–64] . A recent study , however , found evidence for a high frequency of recombination events amongst clinical isolates , consistent with C . albicans strains undergoing sexual or parasexual recombination in their natural environment [13] . These studies are also consistent with the results presented here: the parasexual mating cycle can generate variant genotypes , including a subset of strains that have undergone extensive genetic recombination between chromosomes . For C . albicans , the parasexual mechanism may provide two significant benefits over a conventional sexual pathway . First , the parasexual mechanism is imprecise , generating many aneuploid strains as well as euploid progeny strains . Common aneuploidies included diploid strains harboring trisomic chromosomes , and this karyotypic variation led to greater genetic and phenotypic diversity in the progeny population . Consistent with this observation , changes in chromosome copy number have previously been linked to phenotypic changes in C . albicans , including increased resistance to antifungal azoles [28 , 65] . Thus , karyotypic variation appears to be an important mechanism utilized by C . albicans to regulate physiologically important genes [66] . In S . cerevisiae , aneuploid strains are similarly at a competitive disadvantage with euploid strains , unless there is a strong selective pressure that favors growth of the aneuploid form [40 , 67 , 68] . A second potential benefit of the parasexual cycle is that it bypasses the process of sporulation common to the sexual cycle of most ascomycetes . Ascospores are thought to be highly antigenic , and , given that C . albicans strains normally exist as commensals within warm-blooded hosts , the absence of spore formation may facilitate the generation of genetic diversity without compromising cell survival [56] . In summary , the parasexual cycle in C . albicans provides an alternative to a sexual reproductive cycle . Concerted chromosome loss reduces the ploidy of the cell from tetraploid to approximately diploid , generating recombinant progeny strains with variant phenotypes . Genetic recombination between homologous chromosomes , dependent on the Spo11 protein , takes place during these reductive mitotic divisions , further contributing to genetic diversity . We propose that at least some of the meiotic recombination machinery has been re-programmed to function in parasexual recombination in C . albicans . Finally , we note that as C . albicans thrives only in warm-blooded animals , the parasexual cycle provides a number of potential advantages over a conventional sexual cycle .
Standard laboratory media were prepared as previously described [69] . Construction of the genetically marked tetraploid strain , RBY18 , was previously described [16] . A tetraploid Δspo11 strain was constructed by first deleting the SPO11 gene in the diploid strains RBY16 and CHY477 [16] . Both copies of SPO11 were sequentially disrupted using a modified Ura blaster method [70 , 71] . A SPO11 gene disruption construct was made by PCR amplifying the HisG-URA3-HisG cassette using oligonucleotides SPO11 KO-5′ and SPO11 KO-3′ from plasmid pDDB57 ( see Figure S5 ) [71] . Heterozygous strains were then constructed by replacing SPO11 coding sequences with the URA3 selectable marker flanked by HisG repeats . Ura+ strains that were deleted for one copy of SPO11 were grown on nonselective medium and subsequently plated on SCD medium containing 5-fluoroorotic acid ( 5-FOA ) and uridine medium to select for loss of the URA3 gene [70] . The HisG-URA3-HisG cassette was then used to delete the second copy of the SPO11 gene . The construction of Δspo11 mutants was confirmed using PCR to check both 5′ and 3′ junctions following integration of the Ura blaster cassette and also to confirm the loss of the SPO11 ORF following the second round of transformation . Deletion of SPO11 in the diploid strains RBY16 and CHY477 generated strains RBY77 and RBY79 , respectively . The diploid strains were mated as previously described [16] to form the tetraploid Δspo11 strains RBY176 and RBY177 . To follow expression of the Spo11 protein the gene sequence was fused to that encoding a 13 × myc epitope tag . The Spo11 gene and promoter were first amplified by PCR using oligonucleotides Spo11 ( myc ) for , 5′-cccaatatgaagcactaaactc-3′ and Spo11 ( myc ) rev , 5′-ggcgcgcccggggatccgtttcgtatagctagccgttcc-3′ . The amplified sequence was then digested with HindIII and SmaI enzymes and ligated into a pMYC-HIS1 vector . The resulting plasmid contains the SPO11 gene sequence fused to 13 copies of the myc epitope . The plasmid was then linearized by digestion with BstBI and used to transform strain RBY1118 ( a diploid a-type mating strain ) to generate CAY126 . RBY1118 itself was derived from a/α strain SNY87 [72] by growth on sorbose medium to select for a and α derivatives , as previously described [19] . PCR was used to confirm that the vector had inserted at the endogenous SPO11 allele . To induce chromosome instability in tetraploid strains , the SPO11+ tetraploid strain RBY18 , or Δspo11 tetraploid strains RBY176/177 , were incubated on S . cerevisiae pre-sporulation ( pre-spo ) medium ( 0 . 8% yeast extract , 0 . 3% peptone , 10% dextrose , and 2% agar ) at 37 °C for 10 d . Alternatively , tetraploid strains were incubated on L-sorbose medium ( 0 . 7% yeast nitrogen base ( without amino acids ) , 2% L-sorbose , and 2% agar ) at 30 °C for 10 d . Following incubation , cells that had undergone loss of Chromosome 1 and become gal1− were selected by growth on 2-deoxygalactose ( 2-DOG ) medium for 2 d , as previously described [16] . 2-DOG+ colonies were patched onto YPD and subsequently frozen ( in a 1:1 solution of 50% glycerol and YPD ) . Subsequent culturing of progeny strains was kept to a minimum ( less than 1 wk ) . We have not attempted to reintegrate SPO11 for three major reasons . First , the phenotype being tested is subtle; RBY18 ( SPO11+ ) tetraploid strains exhibited recombination events in three out of 13 progeny , while Δspo11 mutants exhibited no observable recombination events . Second , because we are studying a phenomenon in tetraploid strains , it is not clear how many copies of the SPO11 gene would need to be reintegrated into the tetraploid to generate a significant difference from the mutant . And third , reconstituted strains often exhibit a range of complementation efficiencies , with multiple strains having to be analyzed to confirm restoration of the wild-type phenotype . PCR analysis of the MTL alleles was used as an indicator of the copy number of Chr 5 in each sample . PCR primers unique to MTLa1 , MTLα1 , Δa1 , and Δα2 were used to distinguish MTL alleles in tetraploid cells and progeny cells derived from tetraploids . The oligonucleotides used for MTL analysis have been previously described [14] . To generate cells for flow cytometric analysis , test strains were grown in YPD medium at 30 °C and harvested when the OD was between 1 and 2 . Samples were then prepared for analysis as previously described [16] . Previously , we described the development of a SNP microarray to determine genotypes at 123 SNP loci across the genome of C . albicans ( Forche et al . , unpublished data . ) . For this study the microarray was expanded to include an additional 29 SNP loci , giving a total of 152 ( Table S3 ) . Fifteen of the 29 new SNP loci were adapted from Wu et al . [73] ( Table S3 ) . Since clinical isolates were used by Wu and co-workers and not derivatives of strain SC5314 ( the SNP microarray is based on SNPs from SC5314 ) , the presence of reported SNPs was confirmed by sequencing , as described previously [25] . New primer pairs were developed to allow for the amplification of small PCR products suitable for SNP microarray analysis . Design of allele-specific oligonucleotides , probe generation , slide preparation/hybridization , data analysis , and sequence confirmation of LOH events were conducted as described elsewhere ( [26]; Forche et al . , unpublished data ) . CGH that has been adapted for C . albicans was carried out as described previously [27] . A two-tailed t test was performed to indicate if changes in karyotype were statistically significant . A p-value of < 0 . 05 in the two-tailed t test was interpreted as a significant difference , while p-values >= 0 . 05 were insignificant . Cultures of strains CAY126 ( Spo11-13myc ) , RSY84 ( Kar3-13myc ) , and the untagged RBY1118 strain were grown to logarithmic phase in YPD medium at 30 °C and cells harvested . Whole-cell extracts from these strains were prepared by resuspending cell pellets in lysis buffer ( 10 mM Tris-HCl [pH 7 . 5] , 50 mM NaCl , 1 mM dithiothreitol ) containing protease inhibitors ( pepstatin A , leupeptin , phenylmethyl sulfonyl chloride , and aprotinin ) and lysis achieved by bead beating for 12–15 cycles ( 30 s vortexing following by 30–60 s on ice ) . An aliquot from each sample was separated by SDS-PAGE and analyzed by western blotting . The myc-tagged proteins were detected using an anti-myc antibody at 1/2 , 000 dilution ( 4a6 antibody; Millipore ) followed by an anti-mouse HRP ( horseradish peroxidase ) -conjugated antibody at 1/1 , 000 dilution ( Jackson Laboratories ) . Antibody binding was visualized using the SuperSignal West Pico Chemiluminescent Substrate ( Pierce ) and exposure to autoradiography film . | Candida albicans is an important human fungal pathogen that has an unconventional sexual cycle . Efficient mating requires that diploid cells of opposite mating type first switch from the more common “white” phase to the “opaque” phase and then undergo cell fusion . The resulting tetraploid strains can return to the diploid state via a non-meiotic parasexual program of concerted chromosome loss . We used SNP and comparative genome hybridization to analyze the progeny resulting from this parasexual cycle and found a range of genetically diverse strains with altered phenotypes . In addition , in a subset of these strains , genetic recombination was found to have taken place between homologous chromosomes . This recombination was dependent on Spo11 , a conserved protein required for the introduction of DNA double-strand breaks in the chromosomes of eukaryotes that undergo conventional meiosis . Thus , Spo11 is required for genetic recombination and the generation of increased genetic diversity during the C . albicans parasexual cycle . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"infectious",
"diseases",
"microbiology",
"genetics",
"and",
"genomics"
] | 2008 | The Parasexual Cycle in Candida albicans Provides an Alternative Pathway to Meiosis for the Formation of Recombinant Strains |
Natural populations often grow , shrink , and migrate over time . Such demographic processes can affect genome-wide levels of genetic diversity . Additionally , genetic variation in functional regions of the genome can be altered by natural selection , which drives adaptive mutations to higher frequencies or purges deleterious ones . Such selective processes affect not only the sites directly under selection but also nearby neutral variation through genetic linkage via processes referred to as genetic hitchhiking in the context of positive selection and background selection ( BGS ) in the context of purifying selection . While there is extensive literature examining the consequences of selection at linked sites at demographic equilibrium , less is known about how non-equilibrium demographic processes influence the effects of hitchhiking and BGS . Utilizing a global sample of human whole-genome sequences from the Thousand Genomes Project and extensive simulations , we investigate how non-equilibrium demographic processes magnify and dampen the consequences of selection at linked sites across the human genome . When binning the genome by inferred strength of BGS , we observe that , compared to Africans , non-African populations have experienced larger proportional decreases in neutral genetic diversity in strong BGS regions . We replicate these findings in admixed populations by showing that non-African ancestral components of the genome have also been affected more severely in these regions . We attribute these differences to the strong , sustained/recurrent population bottlenecks that non-Africans experienced as they migrated out of Africa and throughout the globe . Furthermore , we observe a strong correlation between FST and the inferred strength of BGS , suggesting a stronger rate of genetic drift . Forward simulations of human demographic history with a model of BGS support these observations . Our results show that non-equilibrium demography significantly alters the consequences of selection at linked sites and support the need for more work investigating the dynamic process of multiple evolutionary forces operating in concert .
Genetic diversity within a species is shaped by the complex interplay of mutation , demography , genetic drift , and natural selection . These evolutionary forces operate in concert to shape patterns of diversity at both the local scale and genome-wide scale . For example , in recombining species , levels of genetic diversity are distributed heterogeneously across the genome as peaks and valleys that are often correlated with recombination rate and generated by past or ongoing events of natural selection [1] . But at the genome-wide scale , average levels of genetic diversity are primarily shaped by population size changes , yielding patterns of diversity that are a function of a population’s demographic history [2] . These patterns of diversity may also yield information for inferring past events of natural selection and population history , giving valuable insight into how populations have evolved over time [3–8] . With recent advances in sequencing technology yielding whole-genome data from thousands of individuals from species with complex evolutionary histories [9 , 10] , formal inquiry into the interplay of demography and natural selection and testing whether demographic effects act uniformly across the genome as a function of natural selection is now possible . In the past decade , population genetic studies have shed light on the pervasiveness of dynamic population histories in shaping overall levels of genetic diversity across different biological species . For example , multiple populations have experienced major population bottlenecks and founder events that have resulted in decreased levels of genome-wide diversity . Evidence for population bottlenecks exists in domesticated species such as cattle [11] , dogs [12] , and rice [13] , and in natural populations such as Drosophila melanogaster [14–16] , rhesus macaque [17] , and humans [18 , 19] . Notably , population bottlenecks leave long lasting signatures of decreased diversity , which may be depressed even after a population has recovered to , or surpassed , its ancestral size [20 , 21] . Such examples are evident in humans , where non-African populations exhibit a lower amount of genetic diversity compared to Africans [9] , despite the fact that they have been inferred to have undergone a greater population expansion in recent times [22 , 23] . Locally ( i . e . , regionally ) across the genome , the action of natural selection can also lead to distinct signatures of decreased genetic diversity ( although some forms of selection , such as balancing selection , can increase genetic diversity [24] ) . For example , mutations with functional effects may be removed from the population due to purifying selection or become fixed due to positive selection , thereby resulting in the elimination of genetic diversity at the site . But while sites under direct natural selection in the genome represent only a small fraction of all sites genome-wide , the action of natural selection on these selected sites can have far-reaching effects across neutral sites in the genome due to linkage . Under positive selection , genetic hitchhiking [25] causes variants lying on the same haplotype as the selected allele to rise to high frequency during the selection process ( note that we will use the term “genetic hitchhiking” here only in the positive selection context of selection at linked sites ) . Conversely , under purifying selection , background selection ( BGS ) [26] causes linked neutral variants to decrease in frequency or be removed from the population . Both of these processes of selection at linked sites result in decreased neutral genetic diversity around the selected site . Recombination can decouple neutral sites from selected sites in both cases and neutral diversity tends to increase toward its neutral expectation as genetic distance from selected sites increases [27] . Evidence for genetic hitchhiking and BGS has been obtained from the genomes of several species , including Drosophila melanogaster [28–33] , wild and domesticated rice [34 , 35] , nematodes [36 , 37] , humans [3 , 6 , 38–42] , and others ( see [1] for a review ) . While the relative contributions of genetic hitchhiking and BGS to shaping patterns of human genomic diversity have been actively debated [40 , 43–45] , the data strongly support the large role of BGS in shaping genome-wide patterns of neutral genetic variation [41 , 42] . Indeed , recent arguments have been made in favor of BGS being treated as the null model when investigating the effect of selection at linked sites across recombining genomes [1 , 32 , 45–48] , with one study in humans showing that BGS has reduced genetic diversity by 19–26% if other modes of selection at linked sites are assumed to be minor [6] . Although the effects of selection at linked sites across the genome have been described in a multitude of studies , it is still less obvious whether populations that have experienced different demographic histories , such as African and non-African human populations , should exhibit similar relative effects in those regions . Much of the theory developed in the context of BGS has been developed under the assumption that the population is at equilibrium , and recent work has demonstrated that this assumption likely holds under changing demography if selection is strong enough ( or populations are large enough ) such that mutation-selection balance is maintained [49 , 50] . However , strong , sustained population bottlenecks may lead to violations of that assumption , and the effect of genetic drift may dominate the influence of selection at linked sites on determining patterns of genetic variation . Finally , the effect of demography on influencing patterns of diversity in regions experiencing selection at linked sites through time has also been underappreciated ( although see Ref . [51] for a recent study in maize ) . Since most , if not all , natural populations are in a state of changing demography , differences in neutral diversity between populations within regions experiencing selection at linked sites should not only be expected , they should also be expected to change temporally as a function of each population’s specific demographic history . While little attention has been given to the potential consequences of demography on patterns of neutral variation in regions experiencing selection at linked sites ( but see [52 , 53] for how selection at linked sites may affect the inference of demography itself ) , recent studies have suggested that alleles directly under natural selection experience non-linear dynamics in the context of non-equilibrium demography . For the case of purifying selection , the equilibrium frequency of an allele is dependent on its fitness effect , with deleterious alleles having lower equilibrium frequencies than neutral alleles . After a population size change , deleterious alleles tend to change frequency faster than neutral alleles , allowing them to reach their new equilibrium frequency at a faster rate [54 , 55] . This can result in relative differences in deleterious allele frequencies among populations with different demographic histories . Such effects are especially apparent in populations suffering bottlenecks [56] and have been tested and observed between different human populations with founder populations exhibiting a greater proportion of non-synonymous variants relative to synonymous variants [57–59] . We hypothesized that these non-equilibrium dynamics could also perturb nearby neutral variants due to linkage . In support of our hypothesis , a recent simulation study modeling Drosophila observed that population bottlenecks can result in different rates of recovery of neutral genetic diversity depending on the strength of BGS [48] . Another recent study [51] analyzed neutral diversity surrounding putatively deleterious loci in domesticated versus wild maize . They found that the extreme domestication bottleneck of maize reduced the efficiency of purifying selection , which has resulted in higher diversity in regions experiencing BGS relative to neutral regions in the domesticated population compared to the wild population ( which has likely experienced a much more stable demographic history ) . Together , these studies provide further evidence that non-equilibrium demography should have a strong effect on patterns of diversity in the presence of selection at linked sites . To investigate the effect of non-equilibrium dynamics in regions experiencing selection at linked sites , we measure patterns of average pairwise neutral genetic diversity ( π ) as a function of the strength of BGS , B ( background selection coefficient; inferred by Ref . [6] ) , within a global set of human populations from phase 3 of the Thousand Genomes Project ( TGP ) [9] . We focus on the ratio of neutral diversity in regions of strong BGS ( low B ) to regions of weak BGS ( high B; the closest proxy available for neutral variation in humans ) , which we term “relative diversity . ” Due to the inference procedure used to infer specific B values in Ref . [6] , there are many caveats that may plague their direct interpretation ( e . g . , positive selection is not modeled , the distribution of fitness effects are inconsistent with other studies , and the deleterious mutation rate exceeds the per base pair mutation rate of other studies ) . However , we argue that the inferred B values nevertheless provide a decent proxy for ranking sites from most closely linked to deleterious loci ( low B ) to most unlinked from deleterious loci ( high B ) in humans since the key parameters used to infer B , namely recombination rate and local density of selected sites , are fundamental for defining regions of the genome most susceptible to selection at linked sites . We find substantial differences in relative diversity between populations , which we attribute to their non-equilibrium demographics . We confirm that the interplay of demography and selection at linked sites can explain the differences of relative diversity across human populations using simulations incorporating a parametric demographic model of human history [7] with and without a model of BGS . We also investigate how genetic differentiation between TGP populations is shaped by selection at linked sites by measuring FST as a function of B . Finally , we demonstrate that back migration from Europeans and Asians into Africa re-introduces sufficient deleterious variation to affect patterns of BGS , leading to decreased relative diversity in Africans . Our results demonstrate the strong effect that changing demography has on perturbing levels of diversity in regions experiencing selection at linked sites and have implications for population genetic studies seeking to characterize selection at linked sites across any species or population that is not at demographic equilibrium .
We measured mean pairwise genetic diversity ( π ) in the autosomes ( we ignore the sex chromosomes and the mitochondrial genome for all analyses ) among the 20 non-admixed populations from the phase 3 TGP data set , consisting of 5 populations each from 4 continental groups: Africa ( AFR ) , Europe ( EUR ) , South Asia ( SASN ) , and East Asia ( EASN; population labels and groupings reported in Table L in S1 Text ) . A set of stringent filters , including the masking of sites inferred to be under selective sweeps , were first applied to all 20 populations to identify a high-quality set of putatively neutral sites in the genome ( see Materials and Methods ) . Sites were then divided into quantile bins based on estimates of B [6] . For our initial set of analyses , we focused on the bins corresponding to the 1% of sites inferred to be under the strongest amount of BGS ( i . e . , sites having the lowest inferred B values ) and the 1% of sites inferred to be under the weakest amount BGS ( i . e . , sites having the highest inferred B values ) . Mean diversity was normalized by divergence from rhesus macaque within these bins for each population and is shown in Fig 1A and 1B . As expected , normalized diversity was highest in African populations and lowest in East Asian populations across both 1% B quantile bins . To estimate the effect that selection at linked sites has had on neutral diversity , we calculated a statistic called “relative diversity” for each population . We define relative diversity as the ratio of normalized diversity in the lowest 1% B bin to normalized diversity in the highest 1% B bin , which should capture the relative consequences of selection at linked sites within the genome . While this statistic is analogous to “π/π0” in the BGS literature [26 , 60] , we caution that this interpretation is not completely accurate in the context of observed data since even regions estimated to have the highest B values in the human genome may still experience a minimal effect of selection at linked sites . We will use “π/πmin” in the context of observed relative diversity to make clear that we are attempting to minimize selection at linked sites . Fig 1C shows that observed relative diversity was lower in non-African populations ( 0 . 348–0 . 365 for non-Africans , 0 . 396–0 . 408 for Africans ) , demonstrating that these populations have experienced a greater reduction in diversity in regions with strong selection at linked sites and also suggesting that demography may have contributed to these patterns . To characterize these effects across a broader distribution of sites experiencing selection at linked sites , we grouped populations together according to their continental group ( i . e . , African , European , South Asian , and East Asian , see Table L in S1 Text for a detailed description ) and estimated relative diversity at neutral sites for each of the continental groups in bins corresponding to the lowest 1% , 5% , 10% , and 25% quantiles of B ( note these partitions were not disjoint ) . As expected , relative diversity increased for all continental groups as the bins became more inclusive ( Fig 2B ) , reflecting a reduced effect on the reduction of diversity caused by selection at linked sites . We also observed that non-African continental groups consistently had a lower relative diversity compared to African groups , demonstrating that the patterns we observed in the most extreme regions experiencing selection at linked sites also held for broader regions . Interestingly , we observed a consistent trend of rank order for relative diversity between the different continental groups for each quantile bin , with the East Asian group experiencing the greatest reduction of relative diversity , followed by the South Asian , European , and African groups . This result further suggested an effect of demography on the diversity-reducing effect of selection at linked sites , with the strongest effects for those populations experiencing the strongest bottlenecks . However , the observed differences in relative diversity between non-African and African continental groups became less pronounced as the bins became more inclusive ( Fig 2B ) . These effects remained even after we controlled for the effects of GC-biased gene conversion and recombination hotspots ( S2 and S4 Figs in S1 Text ) or if we did not normalize diversity by divergence ( S3 and S5 Figs in S1 Text ) . Patterns of relative diversity in regions of local ancestry ( i . e . , African , European , or Native American ) across admixed TGP populations also largely recapitulated the patterns observed in their continental group counterparts across B quantile bins , with the largest reductions in relative diversity occurring for the Native American and European ancestral segments ( S11 Fig , S1 Text ) . To test if demography has influenced selection at linked sites more recently in time , we also calculated the number of singletons observed per site ( normalizing by divergence and using the same set of neutral filters as was used for the calculations of π ) across the lowest and highest 1% B quantile bins ( S13 Fig in S1 Text ) . While it has been shown theoretically and observed empirically that selection at linked sites skews the site-frequency spectrum towards a higher proportion of singleton variants among segregating sites , the absolute number of singletons among all sites should be lower in regions of strong selection at linked sites when compared to neutral regions . In addition , since singletons are , on average , the youngest variants within the genome , they should better capture signals about very recent population history . Thus , we took the ratio of singletons observed per-site across these extreme B quantile bins to create a statistic called relative singleton density , which we term “ψ/ψmin . ” We accounted for differences in population sample size by first projecting down all populations to 2N = 170 ( Materials and Methods ) . Qualitatively , our measurements of ψ/ψmin showed patterns in the opposite direction to our estimates of π/πmin , with Africans exhibiting a lower ratio of ψ/ψmin when compared to non-Africans ( 0 . 665–0 . 695 for Africans , 0 . 733–0 . 804 for non-Africans; Fig 3 ) . These patterns suggest that the effect of demography on regions experiencing selection at linked sites is transient , with patterns of relative diversity between populations dependent on the time frame in which they are captured ( see Discussion ) . Our results described above offered evidence that demography can affect patterns of neutral diversity in regions of selection at linked sites . Such patterns may be caused by accelerated drift in these regions , which can be amplified by demographic changes , thus leading to accelerated population differentiation . An increase in population differentiation is obvious in the context of hitchhiking ( where linked neutral loci sweep to high frequency ) but is also expected with BGS [61 , 62] . Here we quantified the magnitude of the effect of BGS on population differentiation in humans and found that population differentiation at neutral loci is indeed highly correlated with B ( the inferred strength of BGS; Fig 4 and Table 1 ) . Specifically , we divided the genome into 2% quantile bins based on the genome-wide distribution of B and measured FST in each bin for all pairs of populations from different continental groups [63] . We then performed simple linear regression using B as an explanatory variable and FST as our dependent variable with the linear model FST = β0 + β1B + ε . We found that across all 150 population comparisons ( i . e . , the “Global” estimate in Table 1 ) , B explained 26 . 9% of the change in FST across the most extreme B values . This result was robust to outliers [64] ( Table F in S1 Text ) and dominated the effects of local recombination rate ( see S1 Text ) . One consequence of BGS and hitchhiking in driving patterns of neutral variation within and between human populations is that demographic inference could be substantially biased [52 , 53 , 65] . To assess the degree of bias in the context of human data , we fit a 13-parameter demographic model of African , European , and East Asian demography using only putatively neutral regions of the genome under the weakest effects of selection at linked sites ( B ≥ 0 . 994 ) from a subset of TGP individuals with high coverage whole genome sequence data ( see Materials and Methods ) . Our demographic model followed that of Gutenkunst et al . [7] , with an ancient human expansion in Africa and a single out-of-Africa bottleneck followed by European- and East Asian-specific bottlenecks , as well as exponential growth in both non-African populations , and migration among all populations . To make comparisons to previous studies that have used sequence data from coding regions or genes [7 , 22 , 23] , which may be under strong BGS or hitchhiking effects , we also inferred demographic parameters using coding four-fold degenerate synonymous sites . Our inferred parameters for human demography were strikingly different between the two sets of sequence data ( S1 Fig , Table A in S1 Text ) . Notably , inferred effective population size parameters were larger for contemporary population sizes when using four-fold degenerate synonymous sites versus ascertained neutral regions with B ≥ 0 . 994 , with Ne inferred to be 22% , 23% , and 29% larger for AFR , EUR , and EASN populations , respectively . This is despite the fact that the ancestral Ne was inferred to be lower for four-fold degenerate synonymous sites ( Ne = 18 , 449 and 17 , 118 , for neutral regions with B ≥ 0 . 994 and four-fold degenerate sites , respectively ) . This result may stem from the expected decrease in Ne going into the past in regions of strong BGS , which can lead to inflated estimates of recent population growth [53] and has been found in simulation studies of synonymous sites under BGS [65] . Put more simply , the skew of the site-frequency spectrum towards rare variants in regions experiencing selection at linked sites [66–68] mimics a population expansion , thus leading to erroneous inference . Using the demographic parameters inferred from neutral regions where B ≥ 0 . 994 , we simulated patterns of neutral diversity with and without the effects of BGS ( see Materials and Methods ) . To measure the relative effect of BGS for each population , we took the ratio of neutral diversity from BGS simulations ( π ) and neutral diversity from simulations without BGS ( π0 ) to calculate relative diversity ( π/π0 ) . As expected , we found that BGS reduced relative diversity ( π/π0 < 1 ) for all three populations in our simulations . However , non-African populations experienced a proportionally larger decrease in π/π0 compared to the African population ( π/π0 = 0 . 43 , 0 . 42 , 0 . 41 in AFR , EUR , and EASN respectively ) . These results are comparable to , but not quite as extreme as , the effects we observed in the regions of the genome with the strongest effects of BGS for these population groups ( Fig 1C ) and may therefore reflect the weaker signatures of BGS shown in Fig 2B . To understand how this dynamic process occurs , we sampled all simulated populations every 100 generations through time to observe the effect of population size change on π , π0 , and the ratio π/π0 ( Fig 5 ) . We observed that there is a distinct drop in π and π0 at each population bottleneck experienced by non-Africans , with East Asians ( who had a more severe bottleneck ) experiencing a larger drop than Europeans . Fig 5C shows that the population bottlenecks experienced by non-African populations also reduces π/π0 . Surprisingly , Africans also experienced a large drop in π/π0 ( but less than non-Africans ) even though they did not experience any bottlenecks . This was attributable to migration between non-Africans and Africans and this pattern disappeared when we ran simulations using an identical demographic model with BGS but without migration between populations ( S7 Fig in S1 Text ) . This finding highlights an evolutionary role that deleterious alleles can play when they are transferred across populations through migration ( see Discussion ) . We also observed the effects of demography and BGS on singleton density by calculating ψ/ψ0 ( i . e . , the ratio of singletons observed among all sites in simulations with BGS relative to simulations without BGS ) and again qualitatively observed patterns similar to , but not as extreme as , our empirical estimates of ψ/ψmin ( S12 Fig A in S1 Text ) . Calculating ψ and ψ0 through time showed that the population bottlenecks experienced by non-Africans led to strong decreases in both ψ and ψ0 , with recent expansion in these populations then leading to large , rapid recoveries . Strong decreases in ψ/ψ0 after each population bottleneck were also observed , including a slight decrease in ψ/ψ0 in Africans that disappeared in the simulations without migration ( S12 Fig B in S1 Text ) . While ψ/ψ0 for the European/East Asian ancestral population in the simulations with migration remained below that of Africans during the course of the Out-of-Africa bottleneck , we observed a rapid recovery in ψ/ψ0 for this population in the simulations without migration ( compare bottoms panels , S12 Fig A and B in S1 Text ) . This suggests that for populations experiencing a sustained population bottleneck , the response of singletons to the weakened intensity of BGS is quite rapid , especially when compared to patterns of π/π0 ( compare S7 Fig C to S12 Fig B bottom panel in S1 Text ) . However , population migration mitigates this pattern . Regardless of whether migration between populations was simulated , BGS had little effect on singleton density recovery in Europeans and Asians once population expansion occurred . Our simulations were based on the functional density found in a 2 Mb region of the human genome with the lowest B values and , thus , where BGS was inferred to be strongest ( chr3: 48 , 600 , 000–50 , 600 , 000 ) . There , 20 . 46% of sites were either coding or conserved non-coding ( see Materials and Methods ) which is why the fraction of the genome experiencing deleterious mutation in our simulations of strong BGS was 0 . 2046 . Our simulations were intended to represent the strongest effect of BGS inferred for humans . However , we did not model the specific genomic locations of coding and conserved non-coding sites in our simulations ( since the structure would be specific to each region of the genome ) , so while the patterns we simulated are qualitatively similar to the patterns we observed in real data , there were slight quantitative differences . Since the strength of BGS is dependent upon the density of sites experiencing deleterious mutation within a given region ( or more formally , U , which is the product of the per-site deleterious mutation rate and the number of sites experiencing deleterious mutation [69] ) , we simulated weaker effects of BGS by reducing the fraction of sites experiencing purifying selection while keeping the distribution of selective effects constant ( see Materials and Methods ) . When the fraction of sites experiencing selection was decreased 2–4 fold in our simulations , we continued to observe a stepwise decrease in π/π0 while maintaining the specific rank order of African , followed by European , and then East Asian populations ( S8 Fig in S1 Text ) . As expected , π/π0 increased for all populations as the fraction of sites that were simulated as deleterious decreased ( π/π0 = 0 . 641 vs . 0 . 802 , 0 . 62 vs . 0 . 777 , and 0 . 611 vs . 0 . 777 for AFR , EUR , and EASN when the fraction of sites experiencing selection was reduced to 0 . 1023 and 0 . 05115 , respectively ) . These simulations resulted in π/π0 values much larger than the observed values of π/πmin ( Figs 1C and 2B ) .
In our analyses of thousands of genomes from globally distributed human populations , we have confirmed that the processes of demography and selection at linked sites influence neutral variation across the genome . While this observation is not unexpected , we have characterized the dynamic consequence of non-equilibrium demographic processes in regions experiencing selection at linked sites in humans . We find that demography ( particularly population bottlenecks ) can amplify the consequences of selection at linked sites . To remove any possible biases that would influence our results , we controlled for functional effects of mutations , variability in mutation along the genome , potential sequencing artifacts , GC-biased gene conversion , and the potential mutagenic effects of recombination hotspots . None of these factors qualitatively affected our results . However , because divergence itself is not independent of BGS [70] , biases may arise when using divergence to control for variation in mutation rate along the genome . This is because the rate of coalescence in the ancestral population of two groups will be faster in regions of strong BGS compared to regions of weak BGS due to the lower Ne of the former , thereby leading to a decrease in overall divergence in those regions . While we attempt to limit the contribution of such biases by using a more diverged primate species ( rhesus macaque ) , our calculations of π/πmin show that our results are actually conservative when normalizing by divergence ( π/πmin for AFR is 0 . 373 without the divergence step and 0 . 402 with the divergence step ) . Moreover , the population comparisons we make should be robust to such biases since all human populations are equally diverged from rhesus macaque and estimates of B are constant across populations . We also note that the estimates of B by McVicker et al . [6] may be biased by model assumptions concerning mutation rates and the specific sites subject to purifying selection , with the exact values of B unlikely to be precisely inferred . In fact , the B values provided by McVicker et al . range from 0 to 1 , suggesting that some regions of the genome should be essentially devoid of diversity ( but we do not observe this to be the case ) . Since our own analyses show that relative diversity has a lower bound at only ~0 . 35 in humans , the exact value of B itself should not be taken at face value . Rather , our primary motivation for using B was to ascertain regions that should be on the extreme ends of the genome-wide distribution of regions experiencing selection at linked sites , for which B should provide a good assessment . A study by Comeron [32] that investigated BGS in Drosophila and utilized the same model of BGS as McVicker et al . found that biases presented by model assumptions or mis-inference on the exact value of B do not significantly change the overall rank order for the inferred strength of BGS across the genome . Thus we , expect McVicker et al . ’s inference of B to provide good separation between the regions experiencing the weakest and strongest effects of selection at linked sites within the human genome , with model misspecification unlikely to change our empirical results . While the effects of selection at linked sites captured in our analyses could in principle include the consequences of positive selection ( such as soft-sweeps and classic selective sweeps ) , we applied stringent filters to remove any such regions before our analyses ( Materials and Methods , S1 Appendix ) . Nonetheless , we cannot rule out all contributions from hitchhiking to our results . In fact , our simulations of BGS fail to capture the complete effects of selection at linked sites on reducing π/π0 in different human populations ( compare Figs 1C and 5C ) , and the additional contribution of hitchhiking to humans may explain this discrepancy ( though proper modeling of linkage among deleterious loci could also improve our quantitative results ) . Further investigation will be needed to in order to more fully characterize the effect demography has on influencing the various modes of selection at linked sites , including BGS , selective sweeps , and interference selection [67] . Non-equilibrium demography has also been of recent interest in regards to its effect on patterns of deleterious variation across human populations ( often referred to as genetic load ) , with initial work showing that non-African populations have a greater proportion of segregating non-synonymous deleterious variants compared to synonymous variants [57] . Similar results in human founder populations [58 , 71] , Arabadopsis [72] , and domesticated species such as dogs [12] and sunflowers [73] further demonstrate the pervasive effect that demography has on influencing the relative amount of deleterious variation across a variety of populations and species . Since BGS is a function of deleterious variation , it is not surprising that we also witness differences in π/πmin across human populations that have experienced different demographic histories . These effects are probably ubiquitous across other species as well . However , there has been recent contention about whether the previously described patterns of increased deleterious variants are driven by a decrease in the efficacy of natural selection ( thus resulting in increased genetic load ) or are solely artifacts of the response of deleterious variation to demographic change [59 , 74–77] . Recently , Koch et al . [56] investigated the temporal dynamics of demography on selected sites within humans and observed that after a population contraction , heterozygosity at selected sites can undershoot its expected value at equilibrium as low-frequency variants are lost at a quicker rate before the recovery of intermediate frequency variants can occur . In the context of both BGS and hitchhiking , which skew the site frequency spectrum of linked neutral mutations towards rare variants [26 , 69 , 78 , 79] , we also expect a transient decrease in diversity as low-frequency variants are lost quickly during a population contraction . Indeed , as evident from our simulations of BGS and demography , immediately after a population bottleneck , rapid losses in singleton density can occur , leading to transient decreases in ψ/ψ0 . However , the recovery in singleton density is also quite rapid , while the recovery in π and π/π0 is quite slow . This is due to the fact that higher frequency variants , which contribute a greater amount to π , take a longer amount of time to recover after a population contraction compared to lower-frequency variants such as singletons . Furthermore , Koch et al . also demonstrated that the effect of demography on diversity is only temporary and that long-term diversity at selected sites approaches greater values once equilibrium is reached . The temporal effects of non-equilbrium demographics on patterns of π/πmin and ψ/ψmin may also explain the conflicting results obtained in a similar study of selection at linked sites in teosinte and its domesticated counterpart , maize [51] . In that study , the authors observed that π/πmin was higher in maize , which underwent a population bottleneck during domestication ( no bottleneck event was inferred for the teosinte population ) but that ψ/ψmin was lower . This result is contrary to what we observed qualitatively between non-African and African human populations . However , the demographic models that have been inferred for maize and humans are quite different . Maize is inferred to have had a recent , major domestication bottleneck that was essentially instantaneous and followed by rapid exponential growth [51] . In contrast , demographic models for non-African humans suggest a much more distant bottleneck that was sustained over a longer period of time , and only recently have non-African populations experienced rampant growth ( coinciding with the advent of agriculture ) . Thus , depending on how far in the past a particular demographic event occurred and how strong the population size change was , different qualitative observations of π/πmin and ψ/ψmin will result . Importantly , our simulations show changing values of these statistics through time ( Fig 5 , S12 Fig in S1 Text ) , which can lead to different qualitive results that are dependent on the time frame in which populations are observed . Broadly , our results show that contemporary patterns of neutral diversity cannot easily be attributable to contemporary forces of selection but instead may be exhibiting signatures that are still dominated by older demographic events . Interestingly though , our simulations reveal an additional factor that can influence the effect of BGS within populations–migration between populations . We observe that the exchange of deleterious variants from populations that have experienced extensive bottlenecks to populations with a more stable demography can magnify the strength of selection at linked sites . In particular , our simulations show that both π/π0 and ψ/ψ0 decrease in Africans despite the fact that they are inferred to have been constant in size in their recent evolutionary history ( Fig 5B ) . These patterns disappear when migration is removed ( S7 Fig , S12 Fig B in S1 Text ) ; however , more work is needed to definitively test this . While we describe here the differential effects of non-equilibrium demography on neutral diversity in regions under strong and weak BGS , it is worth mentioning that differences in the reduction of neutral diversity in the genome between different populations have also been investigated at the level of entire chromosomes . In particular , analyses of neutral diversity comparing autosomes to non-autosomes ( i . e . , sex chromosomes and the mitochondrial genome [mtDNA] ) have been conducted . These studies have shown that population contractions have affected the relative reduction of neutral diversity between non-autosomes and autosomes in a similar fashion to what we have observed between regions of strong BGS and weak BGS , with the greatest losses occurring in bottlenecked populations . This was demonstrated in humans [80] and later modeled and shown in other species [81] , with the explanation that stronger genetic drift due to the lower Ne of non-autosomes causes diversity to be lost more quickly in response to population size reductions . Recent work in humans has confirmed such predictions by showing that relative losses of neutral diversity in the non-autosomes are greatest for non-Africans [82–84] . These studies , plus others [85] , have also shown that there is strong evidence for a more dominant effect of selection at linked sites on the sex chromosomes relative to the autosomes in humans . Since selection at linked sites is a pervasive force in shaping patterns of diversity across the genomes in a range of biological species [1] , it has been provided as an argument for why neutral diversity and estimates of Ne are relatively constrained across species in spite of the large variance in census population sizes that exist [47 , 86] . However , since population bottlenecks are common among species and have an inordinate influence on Ne [20] , demography has also been argued as a major culprit for constrained diversity [2 , 86–88] . Yet , as we show in humans , it is likely that patterns of neutral diversity are in fact jointly affected by both of these forces , magnifying one another to deplete levels of diversity beyond what is expected by either one independently . This may play an even larger role in higher Ne species such as Drosophila , where the overall distribution of B was inferred to be even smaller ( i . e . , exhibiting stronger BGS ) than in humans [32] . In our work , we also identify a potentially substantial role for migration from smaller populations that harbor more strongly deleterious alleles on patterns of linked neutral diversity in large populations . Together , these combined effects may help provide additional clues for the puzzling lack of disparity in genetic diversity among different species [89] . Finally , our results also have implications for medical genetics research , since selection may be acting on functional regions contributing to disease susceptibility . Since different populations will have experienced different demographic histories , the action of selection at linked sites may result in disparate patterns of genetic variation ( with elevated levels of drift ) near causal loci . Recent work has already demonstrated that BGS’s consequence of lowering diversity affects power for disease association tests [90] . Our results indicate that this may be even further exacerbated by demography in bottlenecked populations , leading to potentially larger discrepancies in power between different populations . Overall , this should encourage further scrutiny for tests and SNP panels optimized for one population since they may not be easily translatable to other populations [91] . It should also further motivate investigators to simultaneously account for demography and selection at linked sites when performing tests to uncover disease variants within the genome [90 , 92 , 93] .
2 , 504 samples from 26 populations in phase 3 of the Thousand Genomes Project ( TGP ) [9] were downloaded from ftp://ftp . 1000genomes . ebi . ac . uk/vol1/ftp/release/20130502/ . vcftools ( v0 . 1 . 12a ) [94] and custom python scripts were used to gather all bi-allelic SNP sites from the autosomes of the entire sample set . A subset of TGP samples that were sequenced to high coverage ( ~45X ) by Complete Genomics ( CG ) were downloaded from ftp://ftp . 1000genomes . ebi . ac . uk/vol1/ftp/phase3/data/ . After filtering out related individuals via pedigree analyses , we analyzed 53 YRI , 64 CEU , and 62 CHS samples ( Table B in S1 Text ) . The cgatools ( v1 . 8 . 0 ) listvariants program was first used to gather all SNPs from the 179 samples using their CG ASM “Variations Files” ( CG format version 2 . 2 ) . Within each population , the number of reference and alternate allele counts for each SNP was then calculated using the cgatools testvariants program and custom python scripts . Only allele counts across high quality sites ( i . e . , those classified as VQHIGH variant quality by CG ) were included . Low quality sites ( i . e . , those with VQLOW variant quality ) were treated as missing data . Only autosomes were kept . Non-bi-allelic SNPs and sites violating Hardy-Weinberg equilibrium ( HWE ) ( p-value < 0 . 05 with a Bonferroni correction for multiple SNP testing ) were also removed . We collected 13 whole-genome sequenced KhoeSan samples ( sequence-coverage: 2 . 5-50X , see Table C in S1 Text ) from 3 studies [95–97] and used the processed vcf files from each of those respective studies to gather all bi-allelic polymorphic SNPs ( i . e . , the union of variants across all vcf files ) . SNPs were only retained if they were polymorphic within the 13 samples ( i . e . , sites called as alternate only within the sample set were ignored ) . Positions in the genome were annotated for background selection by using the background selection coefficient , B , which was inferred by McVicker et al . [6] and downloaded from http://www . phrap . org/othersoftware . html . B was inferred by applying a classical model of BGS [60] , which treats its effects as a simple reduction in Ne at neutral sites as a function of their recombination distance from conserved and exonic loci , the strength of purifying selection at those loci , and the deleterious mutation rate . B can be interpreted as the reduced fraction of neutral genetic diversity at a particular site along the genome that is caused by BGS , with a value of 0 indicating a near complete removal of neutral genetic diversity due to BGS and a B value of 1 indicating little to no effect of BGS on neutral genetic diversity ( B = π/π0 = Ne/N0 ) . Positions for B were lifted over from hg18 to hg19 using the UCSC liftOver tool . Sites that failed to uniquely map from hg18 to hg19 or failed to uniquely map in the reciprocal direction were excluded . Sites lacking a B value were also ignored . We focused our analyses on those regions of the genome within the lowest 1% , 5% , 10% , and 25% of the genome-wide distribution of B and within the highest1% of the genome-wide distribution of B . These quantiles correspond to the B values 0 . 095 , 0 . 317 , 0 . 463 , 0 . 691 , and 0 . 994 , respectively . A set of 13 filters ( referred to as the “13-filter set” ) were used to limit errors from sequencing and misalignments with rhesus macaque and to remove regions potentially under the direct effects of natural selection and putative selective sweeps . These filters were applied to all samples in phase 3 TGP ( all filters are in build hg19 ) for all sets of analyses ( see Table D in S1 Text for the total number of Mb that passed the described filters below for each particular B quantile ) : Additionally , an extra set of filters was applied , but only for those estimates of diversity that controlled for GC-biased gene conversion and recombination hotspots: To generate a set of four-fold degenerate synonymous sites , all polymorphic sites that we retained from the high-coverage Complete Genomic samples were annotated using the program ANNOVAR [106] with Gencode V19 annotations . ANNOVAR and Gencode V19 annotations were also used to gather an autosome-wide set of four-fold degenerate sites ( i . e . , all possible sites , regardless of being polymorphic ) , resulting in 5 , 188 , 972 total sites . The inference tool dadi ( v1 . 6 . 3 ) [7] was used to fit , via maximum likelihood , the 3-population 13-parameter demographic model of Gutenkunst et al . [7] to the 179 YRI , CEU , and CHS samples from the high coverage CG dataset of TGP . This sample set consisted of 53 YRI ( African ) , 64 CEU ( European ) , and 62 CHS ( East Asian ) samples . The demographic model incorporates an ancient human expansion in Africa and a single out-of-Africa bottleneck followed by European- and East Asian-specific bottlenecks , as well as exponential growth in both non-African populations and migration between populations . During the inference procedure , each population was projected down to 106 chromosomes , corresponding to the maximum number of chromosomes available in the CG YRI population . Sites were polarized with chimpanzee to identify putative ancestral/derived alleles using the chain and netted alignments of hg19 with panTro4 ( http://hgdownload . soe . ucsc . edu/goldenPath/hg19/vsPanTro4/axtNet/ ) , and the correction for ancestral misidentification [107] option in dadi was used . The 13-filter set described previously was applied to the CG data set , and an additional filter keeping only the autosomal sites in the top 1% of B ( B ≥ 0 . 994 ) was also applied in order to mitigate potential biases in inference due to BGS [53 , 65] or other forms of selection at linked sites [52] . After site filtering and correction for ancestral misidentification , a total of 110 , 582 segregating sites were utilized by dadi for the inference procedure . For optimization , grid points of 120 , 130 , and 140 were used , and 15 independent optimization runs were conducted from different initial parameter points to ensure convergence upon a global optimum . An effective sequence length ( L ) of 7 . 15 Mb was calculated from the input sequence data after accounting for the fraction of total sites removed due to filtering . In addition to the 13-filter set , this filtering included sites violating HWE , sites without B value information , sites that did not have at least 106 sampled chromosomes in each population , sites with more than two alleles , sites that did not have tri-nucleotide information for the correction for ancestral misidentification step , and sites treated as missing data . For calculating the reference effective population size , a mutation rate ( μ ) of 1 . 66 x 10−8 ( inferred from Ref . [108] ) was used . Using the optimized θ from dadi after parameter fitting , the equation θ = 4NeμL was solved for Ne to generate the reference effective population size , from which all other population Ne’s were calculated . This same procedure was also used to infer demographic parameters from four-fold degenerate synonymous sites across the same set of samples . After site filtering ( note that B and the 13-filter set were not included in the filtering step for four-fold degenerate synonymous sites ) , 41 , 260 segregating sites were utilized by dadi for the inference procedure , and an effective sequence length of 2 . 37 Mb was used for calculating the reference effective population size . Forward simulations incorporating the results from the demographic inference procedure described above and a model of background selection were conducted using SFS_CODE [109] . For the model of background selection , the recombination rate , ρ , and the fraction of the genome experiencing deleterious mutation were calculated using the 2 Mb region of chr3: 48 , 600 , 000–50 , 600 , 000 , which has been subject to the strongest amount of BGS in the human genome ( mean B = 0 . 002 ) . A population-scaled recombination rate ( ρ ) of 6 . 0443 x 10−5 ( raw recombination rate of 8 . 19 x 10−10 ) was calculated for this region using the HapMap II GRCh37 genetic map [104] . For ascertaining the fraction of sites experiencing deleterious mutation , the number of non-coding “functional” sites in this region was first calculated by taking the union of all phastCons sites and phyloP sites with scores > 1 . 2 ( indicating conservation ) that did not intersect with any coding exons . This amount totaled to 270 , 348 base pairs . Additionally , the number of coding sites was calculated by summing all coding exons within this region from GENCODE v19 , which totaled to 138 , 923 base pairs . From these totals , the total fraction of deleterious sites , 0 . 2046 , was generated . The background selection model was simulated using a middle 30 kb neutral region flanked by two 1 Mb regions under purifying selection . From the calculated fraction of deleterious sites described above , 20 . 46% of sites in the two 1 Mb flanking regions were simulated as being deleterious . The mutation rate in our simulations for the deleterious sites and for neutral sites were both set to 1 . 66 x 10−8 [108] . Two distributions of fitness effects were used for the deleterious sites , with 66 . 06% of deleterious sites using the gamma distribution ( parameters: mean = α/β , variance = α/β2 ) of fitness effects inferred across conserved non-coding regions by Ref . [110] ( β = 0 . 0415 , α = 0 . 00515625 ) and 33 . 94% of deleterious sites using the gamma distribution of fitness effects inferred across coding regions by Ref . [5] ( β = 0 . 184 , α = 0 . 00040244 ) . Gamma distribution parameters were scaled to the ancestral population size of the demographic models used in Refs . [5 , 110] . Their unscaled values are ( β = 0 . 0415 , α = 80 . 11 ) and ( β = 0 . 184 , α = 6 . 25 ) for conserved non-coding regions and coding regions , respectively . The relative number of non-coding “functional” sites and coding exons described above determined the relative number of sites receiving each distribution of fitness effects in our simulations . An example of the SFS_CODE command for our simulations is in S1 Text . To simulate varying levels of background selection strength , different total fractions of our original calculated deleterious fraction of 0 . 2046 were used ( i . e . , 5% , 10% , 25% , 50% , and 100% of 0 . 2046 ) . However , the same relative percentage of non-coding and coding sites and mutation rate were used . These different simulated fractions of deleterious sites resulted in a reduced total deleterious mutation rate , U , which is the product of the per-site deleterious mutation rate and the total number of sites experiencing deleterious mutation [69] . Thus , weaker effects of BGS were simulated . To simulate only the effects of demography without background selection , only the 30 kb neutral region was simulated . 2 , 000 independent simulations were conducted for each particular set of the deleterious site fraction ( 2 , 000 x 6 = 12 , 000 total ) . Simulations output population genetic information for 100 samples every 100 generations and also at each generation experiencing a population size change ( 22 , 117 total generations were simulated ) , from which mean pairwise nucleotide diversity ( π ) and singleton density ( ψ ) was calculated across the 2 , 000 simulations . Mean pairwise genetic diversity ( π ) and singleton density ( ψ ) was calculated as a function of the B quantile bins described in “Filtering and ascertainment scheme” for each of the 20 non-admixed populations in phase 3 TGP and , for π , across 4 broad populations that grouped the 20 non-admixed populations together by continent ( African , European , South Asian , and East Asian , see Table L in S1 Text ) . Additionally , only regions of the genome passing the 13-filter set were used in the calculations of π and ψ ( see Table D in S1 Text for total number of Mb used in diversity calculations for each B quantile ) . When calculating ψ for each non-admixed phase 3 TGP population , the site-frequency spectrum was first projected down to 2N = 170 samples ( the number of chromosomes in MSL , the smallest phase 3 population sample ) using a hypergeometric distribution [7] from each population’s full ( unfolded ) site-frequency spectrum . This allowed for unbiased comparisons of singleton density between all populations . Additionally , when identifying singletons for calculating ψ , only sites annotated with high confidence calls for polarizing ancestral and derived states were used when creating the unfolded site-frequency spectrum . These high confidence sites were ascertained from the GRCh37 ancestral sequence ( downloaded from ftp://ftp . ensembl . org/pub/release-71/fasta/ancestral_alleles/homo_sapiens_ancestor_GRCh37_e71 . tar . bz2 ) . For estimates of diversity controlling for gBGC or recombination hotspots , the additional corresponding filters described in “Filtering and ascertainment scheme” were also used . Only 100 kb regions of the genome with at least 10 kb of divergence information with Rhesus macaque were used in π and ψ calculations ( see “Normalization of diversity and divergence calculations with Rhesus macaque” below ) . To calculate human divergence with Rhesus macaque , we downloaded the syntenic net alignments between hg19 and rheMac2 that were generated by blastz from http://hgdownload . cse . ucsc . edu/goldenpath/hg19/vsRheMac2/syntenicNet/ . We binned the human genome into non-overlapping 100 kb bins and calculated divergence within each bin by taking the proportion of base pair differences between human and Rhesus macaque . Gaps between human and Rhesus macaque , positions lacking alignment information , and positions that did not pass the 13-filter set described in “Filtering and ascertainment scheme” were ignored in the divergence estimate . Additionally , a separate set of divergence estimates were also made using the additional set of filtering criteria that removed those regions under gBGC or in recombination hotspots and were used for normalizing diversity in those measurements that controlled for gBGC and hotspots . When normalizing diversity and singleton density by divergence , only 100 kb bins that had at least 10 kb of divergence information were used ( 21 , 100 bins total for 13-filter set; 20 , 935 bins total for the 13-filter set plus the additional gBGC and hotspot filters ) . Bins with less than 10 kb of divergence information were ignored . To make estimates comparable , in those measurements of diversity that did not normalize by divergence , diversity was still calculated using the same set of 100 kb bins that had at least 10 kb for estimating divergence . FST calculations were performed as a function of B between every pair of non-admixed phase 3 TGP populations not belonging to the same continental group ( 150 pairs total ) . We followed the recommendations in Bhatia et al . [63] to limit biases in FST due to 1 ) type of estimator used , 2 ) averaging over SNPs , and 3 ) SNP ascertainment . Specifically , we 1 ) used the Hudson-based FST estimator [111] , 2 ) used a ratio of averages for combining FST estimated across different SNPs , and 3 ) ascertained SNPs based on being polymorphic in an outgroup ( i . e . , the KhoeSan ) . For ascertaining SNPs in the KhoeSan , we also performed filtering according to the filtering scheme described under “Filtering and ascertainment scheme . ” For a position to be considered polymorphic in the KhoeSan , at least one alternate allele and one reference allele had to be called across the 13 genomes we utilized ( see “Data” ) . These criteria left 3 , 497 , 105 total sites in the genome in the phase 3 dataset for FST to be estimated across . FST was calculated across 2% quantile bins of B ( based on the genome-wide distribution of B ) for all pairwise comparisons of populations between a specific pair of continental groups ( 25 pairs total ) or across all pairwise comparisons using all continental groups ( 150 pairs total ) . Simple linear regression was performed with the model FST = β0 + β1B + ε . The mean of the bounds defining each quantile bin was used when defining the explanatory variables for the regression . Linear regression , robust linear regression [64] , and simple correlation were performed using the lm ( ) , rlm ( ) , and cor ( ) functions , respectively , in the R programming language ( www . r-project . org ) . To generate standard errors of the mean , this same procedure was performed on FST results generated from each of 1 , 000 bootstrapped iterations of the data . | Patterns of genetic diversity within a species are affected at broad and fine scales by population size changes ( “demography” ) and natural selection . From both population genetics theory and observation on genomic sequence data , it is known that demography can alter genome-wide average neutral genetic diversity . Additionally , natural selection can affect neutral genetic diversity regionally across the genome via selection at linked sites . During this process , natural selection acting on adaptive or deleterious variants in the genome will also shape diversity at nearby neutral sites due to genetic linkage . However , less is known about the dynamic changes to diversity that occur in regions affected by selection at linked sites when a population undergoes a size change . We characterize these dynamic changes using thousands of human genomes and find that the population size changes experienced by humans have shaped the consequences of selection at linked sites across the genome . In particular , population contractions , such as those experienced by non-Africans , have disproportionately decreased neutral diversity in regions of the genome inferred to be under strong background selection ( i . e . , selection at linked sites that is caused by natural selection acting on deleterious variants ) , resulting in large differences between African and non-African populations . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"ecology",
"and",
"environmental",
"sciences",
"population",
"genetics",
"geographical",
"locations",
"human",
"genomics",
"invertebrate",
"genomics",
"simulation",
"and",
"modeling",
"dna",
"population",
"biology",
"research",
"and",
"analysis",
"methods",
"europe",
"ecological",
"metrics",
"genomics",
"species",
"diversity",
"animal",
"genomics",
"people",
"and",
"places",
"biochemistry",
"ecology",
"nucleic",
"acids",
"natural",
"selection",
"genetics",
"biology",
"and",
"life",
"sciences",
"dna",
"recombination",
"evolutionary",
"biology",
"evolutionary",
"processes"
] | 2018 | Human demographic history has amplified the effects of background selection across the genome |
T cell receptors ( TCRs ) are key to antigen-specific immunity and are increasingly being explored as therapeutics , most visibly in cancer immunotherapy . As TCRs typically possess only low-to-moderate affinity for their peptide/MHC ( pMHC ) ligands , there is a recognized need to develop affinity-enhanced TCR variants . Previous in vitro engineering efforts have yielded remarkable improvements in TCR affinity , yet concerns exist about the maintenance of peptide specificity and the biological impacts of ultra-high affinity . As opposed to in vitro engineering , computational design can directly address these issues , in theory permitting the rational control of peptide specificity together with relatively controlled increments in affinity . Here we explored the efficacy of computational design with the clinically relevant TCR DMF5 , which recognizes nonameric and decameric epitopes from the melanoma-associated Melan-A/MART-1 protein presented by the class I MHC HLA-A2 . We tested multiple mutations selected by flexible and rigid modeling protocols , assessed impacts on affinity and specificity , and utilized the data to examine and improve algorithmic performance . We identified multiple mutations that improved binding affinity , and characterized the structure , affinity , and binding kinetics of a previously reported double mutant that exhibits an impressive 400-fold affinity improvement for the decameric pMHC ligand without detectable binding to non-cognate ligands . The structure of this high affinity mutant indicated very little conformational consequences and emphasized the high fidelity of our modeling procedure . Overall , our work showcases the capability of computational design to generate TCRs with improved pMHC affinities while explicitly accounting for peptide specificity , as well as its potential for generating TCRs with customized antigen targeting capabilities .
T cell receptors ( TCRs ) are key elements of adaptive immunity , as they specifically recognize antigenic peptides bound to MHC proteins ( pMHCs ) on cell surfaces and are responsible for initiating immune responses against targeted cells . The TCR-pMHC interaction is of considerable importance in health and disease , notably in transplantation , autoimmunity , and is a target for development of vaccines and therapeutics for infectious disease and cancer [1]–[3] . For example , the adoptive transfer of genetically engineered T cells , whereby tumor-specific TCRs are transduced into T cells and then infused into the patient , is being explored as a means for cancer immunotherapy . Clinical trials of such genetically engineered T cells have shown promise in the treatment metastatic melanoma [4]–[6] and synovial cell carcinoma [7] , leading to durable tumor regression and long-term survival in patients . The observations that TCRs have relatively weak affinities towards pMHC ( typically 1–300 µM; ∼1000-fold lower than mature antibody/antigen interactions ) and that pMHC affinities are correlated to some extent with in vivo potency [8] have led to a number of efforts to engineer TCRs with enhanced binding affinity . These efforts include in vitro selection [9]–[13] as well as computational structure-based design [14]–[16] , resulting in up to 1 , 000 , 000-fold improvements in affinity . However , a major concern in enhancing TCR affinity is maintenance of peptide specificity . As TCRs recognize peptides presented by MHC proteins , yet invariably form contacts to both peptide and MHC [17] , enhancements to TCR affinity risk dangerous cross-reactivity if affinity-enhancing substitutions preferentially target the MHC protein . Such “off-target” interactions can be challenging to predict from peptide sequence and are a major concern for high affinity TCRs [18] . Indeed , the unanticipated cross-reactivity of a high affinity TCR resulted in serious consequences and deaths in a recent clinical trial [19] . Additionally , significant enhancements in antigen-specific affinity may be detrimental for T cell activity , as there is evidence of a TCR “threshold affinity” above which T cell responsiveness is attenuated [20] , [21] . Thus , careful control of affinity and specificity is crucial in the development of enhanced TCRs for therapeutic purposes . The αβ TCR DMF5 was originally isolated from tumor infiltrating lymphocytes present in a patient with metastatic melanoma [22] . DMF5 recognizes the 27–35 nonameric and 26–35 decameric peptide epitopes from the MART-1 melanoma antigen presented by the class I MHC protein HLA-A*0201 ( HLA-A2 ) , and was the second TCR to be used in clinical trials of genetically engineered T cells [5] . Without knowledge of structure or affinity , Robbins and colleagues previously examined a series of point mutations in DMF5 , generating variants that resulted in improved antigen-specific responses yet also showed evidence of reduced specificity , underscoring the need for incorporating structural information in the design process [23] . More recently , the DMF5 TCR has been crystallized by our laboratory in complex with both the MART-1 nonameric epitope ( AAGIGILTV; referred to as AAG ) as well as the anchor-modified decameric epitope ( ELAGIGILTV; referred to as ELA ) , both bound to HLA-A2 [24] . The structures show that despite the significant difference in peptide conformation between the ELA/HLA-A2 and AAG/HLA-A2 ligands [24] , DMF5 engages them with an identical binding mode . These structures along with associated affinity measurements provide an ideal opportunity to explore the applicability of computational structure-based design for rationally enhancing a clinically relevant TCR , while simultaneously exploring the impact on peptide specificity . Utilizing a refined algorithm initially developed for our redesign of the A6 TCR [14] , we applied structure-based design to the DMF5 TCR , generating variants and characterizing mutants with affinity enhancements of up to 400-fold toward ELA/HLA-A2 . Highlighting the ability of structure-based design to directly target regions of interest within protein interfaces , and in contrast with results seen with in vitro selection , the strongest affinity enhancement was achieved with only two previously identified amino acid substitutions [25] that directly interact with the peptide . Importantly , the highest affinity variant showed no detectable recognition of unrelated peptides presented by HLA-A2 . We determined the crystallographic structure of this variant bound to ELA/HLA-A2 , permitting a detailed analysis of the accuracy of the various structural modeling protocols , and together with the affinity measurements , a quantitative assessment of scoring functions and terms . Further , by purposely disrupting interactions with the ELA peptide , we were able to shift TCR specificity away from the ELA peptide toward the AAG peptide , albeit with more modest efficacy . Altogether , these results highlight the promise of structure-based design for TCR engineering , and provide a rich dataset for further improvements in design strategies , including the broadening of efforts to other TCR-pMHC systems . Lastly , given the ongoing use of the DMF5 TCR in efforts to develop immunological therapies for melanoma ( e . g . , [26] ) , the high affinity DMF5 variants identified here may have future clinical applicability .
We used the ZAFFI and Rosetta software tools [14] , [27] to predict the affinity changes of DMF5 mutants for ELA/HLA-A2 or AAG/HLA-A2 , simulating all point mutations for each DMF5 residue within 5 . 5 Å of the pMHC ligand in the tertiary structures . In total , we examined 589 substitutions of 31 DMF5 residues within each complex , which were then ranked based on predicted TCR-pMHC affinity . Twelve computationally designed mutations were chosen for experimental testing . To help maintain peptide specificity , with the exception of two αR27 mutants , we only chose mutants that were predicted to contact the peptides . The αR27 mutants were selected to compare with our previously designed substitutions at the corresponding position in the A6 TCR [14] , which shares the germline α chain gene ( TRAV 12-2 ) and some MHC contacts with DMF5 . We performed mutagenesis using soluble DMF5 gene constructs , expressed and purified the mutant proteins , and measured their binding affinities toward ELA/HLA-A2 and AAG/HLA-A2 via surface plasmon resonance ( Figure 1 ) . The mutations and their measured affinities for ELA/HLA-A2 and AAG/HLA-A2 are given in Table 1 , organized by the method through which they were selected: Affinity , Specificity , or Proline , as discussed in detail below . In addition , Table 1 includes four mutations , listed under “Test” , that we selected for measurement based on manual inspection of the TCR-pMHC structures . Mutations in the Affinity category were chosen on the basis of predicted enhancement in affinity towards both ELA/HLA-A2 and AAG/HLA-A2 . Three of the six mutations in this category had significantly improved affinities: αD26W , αD26Y , and βL98W . The two αD26 mutants had the highest measured binding affinities among all tested mutants ( up to 40-fold improvement for αD26W towards AAG/HLA-A2 ) , while the βL98W mutant had a 3-fold affinity improvement for both ELA/HLA-A2 and AAG/HLA-A2 . Mutations in the Specificity category were chosen on the basis of predicted differential affinity towards ELA/HLA-A2 and AAG/HLA-A2 . These were predicted to contact a portion of the interface that varies between the two peptides , where the alanine at the N-terminus of the nonamer is replaced by a larger glutamate residue in the decamer ( Figure S1 ) . Several mutations at TCR position αG28 were chosen that would potentially destabilize the interaction with ELA/HLA-A2 via steric hindrance while favoring AAG/HLA-A2 . Of the specificity-altering substitutions , all shifted specificity toward the AAG nonamer as predicted , albeit the shifts were relatively modest ( up to a 5-fold shift; Table 1 and Figure 2 ) . Based on work with the A6 TCR [28] , as well as the observation of proline CDR mutants in high affinity TCR selection experiments [9] , [11] , we tested three proline mutations that were predicted to stabilize CDR loops in the bound conformation while not negatively impacting contacts with the pMHC ( Proline category in Table 1 ) . None of these proline substitutions showed a significant improvement in affinity , indicating that while potentially reducing the entropic cost for binding , the magnitudes of any such improvements were not substantial enough to yield a net increase in binding free energy , possibly because these loops appear relatively rigid in the unbound DMF5 TCR [29] . Moreover , given the >1 kcal/mol loss in binding free energy with both ELA/HLA-A2 and AAG/HLA-A2 , the αG28P substitution may have directly or indirectly impacted pMHC contacts , consistent with its relatively buried position in the pMHC interface . Combining the affinity-enhancing αD26Y and βL98W mutations ( this double mutant is referred to as YW ) yielded a substantial improvement towards ELA/HLA-A2 . This high affinity double mutant was previously described in a brief report , with a preliminary affinity measurement yielding an approximate 200-fold enhancement [25] . Here , however , we measured a 400-fold improvement ( from 9 . 5 µM to 24 nM ) . The difference is attributable to our use of a kinetic titration binding assay in this case ( Figure 1b ) , which is more accurate at quantifying binding in the nanomolar range or higher , as it permits analyses of high affinity binders without requiring surface regeneration [30] . The on and off rates of the YW mutant towards ELA/HLA-A2 determined from the kinetic titration were 1 . 7×106 M−1 s−1 and 0 . 05 s−1 , respectively . The dissociation rate of wild type DMF5 from ELA/HLA-A2 was too fast to accurately measure [29] , indicating that the combined mutations result in a slower TCR off rate , as seen with the majority of affinity-enhanced TCRs [31] . The combined YW mutations were somewhat nonadditive ( −3 . 5 kcal/mol enhancement versus −2 . 5 kcal/mol assuming additivity ) , suggesting a modest degree of communication between the CDR1α and CDR3β loops; the same degree of cooperativity was also observed for the αD26W/βL98W ( WW ) mutant binding ELA/HLA-A2 ( Table 1 ) . Nonadditivity within TCR binding interfaces has been observed previously [32] , [33] , and could be attributable to structural or dynamic effects of mutations on neighboring loops . The YW variant also showed a smaller but still considerable 30-fold enhancement towards AAG/HLA-A2 . The reduced affinity enhancement is likely attributable to the lack of the N-terminal glutamate in the AAG peptide as discussed below . Given its dramatic affinity improvement toward both ELA/HLA-A2 and AAG/HLA-A2 , we next asked whether the high affinity YW variant could recognize targets other than the MART-1 nonamer and decamer . No binding was detectable towards HLA-A2 presenting the Tax or gp100 peptides , even at concentrations more than 25-fold higher than those used to characterize binding to wild type DMF5 ( Figure S2 ) . The Tax and gp100 peptides have markedly different sequences from ELA or AAG ( Tax: LLFGYPVYV; gp100: IMDQVPFSV ) , yet the conformations of HLA-A2 are identical in the four peptide/HLA-A2 crystal structures [24] , [34] , [35] . The lack of detectable binding of the high affinity DMF5 YW variant towards the other peptides thus suggests that we may have improved its specificity towards the MART-1 peptides , and at the minimum demonstrates that our design has avoided peptide-independent targeting of HLA-A2 . To quantify the performance of the design methods that we used to generate candidate mutations , ZAFFI and Rosetta , we compared predicted versus measured affinities towards ELA/HLA-A2 and AAG/HLA-A2 for each of the point mutations that were experimentally characterized ( excluding the αY50A and αG94T mutants , for which binding was too weak to measure ) . Mutants were scored with or without structural minimization ( referred to as Min and NoMin respectively ) , as shown in Figure 3 ( with scores in Table 2 ) . For both the Rosetta and ZAFFI scoring functions , the NoMin simulations yielded higher agreement with experimental data ( Figure 3a–b ) , with the Rosetta scoring function achieving an impressive 0 . 72 correlation with measured ΔΔGs ( excluding four outlier points correctly predicted to have poor affinities ) . Except for the proline mutant αG28P , the Rosetta NoMin protocol made no other false positive predictions , and its top four predictions ( αD26Y and αD26W for the two pMHCs ) had the highest measured affinities among all predicted point mutations ( βL98W was also correctly ranked highly , particularly for AAG ) . This predictive success is notable as the majority ( 8 out of 14 ) of these mutants involved glycine and proline , which are often overlooked during in silico studies due to difficulties predicting backbone-related effects [27] . The ZAFFI NoMin protocol gave a correlation of 0 . 59 with measured data ( again excluding several true negative outlier points due to predicted steric hindrance ) . Though it previously outperformed Rosetta in scoring A6 TCR mutants [14] , and correctly gave favorable scores for the DMF5 αD26 mutants , ZAFFI made several false positive DMF5 predictions for both AAG/HLA-A2 and ELA/HLA-A2 , possibly due to its parameterization on a more limited dataset than Rosetta and the distinct biophysical properties of the A6 and DMF5 interfaces . This led us to evaluate and reparameterize the terms in the ZAFFI function using a larger set of energy terms and mutants , as described further below . Both minimization-based protocols ( ZAFFI Min and Rosetta Min; Figure 3c–d ) , while displaying positive correlations with the experimental results , were lower in their predictive success than the NoMin protocols . However , ZAFFI Min scored the αD26 mutants favorably , and correctly identified βL98W ( for AAG ) as within the score cutoff for predicted binding improvement ( ≤−0 . 6; for ELA , βL98W was near this cutoff ) . Overall though , false positive predictions for ZAFFI and Rosetta led to relatively weak correlations , suggesting that minimization may have led to incorrect structures in some cases . We additionally tested other minimization protocols as well as more extensive side chain packing ( Table S1 ) , each of which gave lower correlations with measured energies than the relatively restrictive NoMin protocol . To examine the structural basis of the 400-fold binding affinity improvement and compare with the models generated during the design process , we crystallized and determined the structure of the DMF5 YW mutant bound to ELA/HLA-A2 at 2 . 56 Å resolution ( Figure 4 , with crystallographic data in Table S2 ) . Clear electron density was observed for the TCR-pMHC interface , and the positions of the mutated amino acids were unambiguous as indicated by an unbiased , iterative-build OMIT map [36] ( Figure S3 ) . As with other structurally characterized TCRs engineered for high pMHC affinity [12] , [13] , [37]–[39] , the docking orientation was conserved when compared to the wild-type complex , with a TCR-pMHC crossing angle of 32° , versus 33° for the wild-type . Essentially no perturbations of the interface CDR loops or peptide were observed ( 0 . 34 Å backbone atom RMSD for TCR and pMHC residues within 10 Å of the binding interface ) , indicating that our relatively conservative design strategy of selecting point substitutions against a fixed pMHC structure did not substantially alter the interface or proximal side chains ( Figure 4b–e ) . This tight structural conservation of the binding loops and target pMHC residues is in contrast to some high affinity TCRs generated by in vitro selection where moderate ( 1G4 designs c5c1 , c48c50 , c58c61 , c58c62 ) [12] , [37] or pronounced ( 2C designs m6 , m13 , m67 , and Mel5 design α24β17 ) [13] , [38] , [39] perturbations of CDR loops were exhibited , along with adjacent CDR loop remodeling [39] and addition of a synergistic ion adduct in the interface [12] . In the recently described structure of the c134 TCR [39] which is an in vitro selected variant of the A6 TCR with nearly 1000-fold improved affinity for Tax/HLA-A2 , the mutant CDR3β loop retained largely the same backbone structure as the wild-type loop , yet it led to a shifted footprint of the α chain over the pMHC . As anticipated from our modeling , both the tyrosine and tryptophan mutant side chains directly contact the MART-1 peptide in the αD26Y/βL98W structure , and make more extensive peptide contacts than their wild-type counterparts ( Table S3 ) . These mutations led to a 5% increase in buried solvent accessible surface area for the pMHC , from 1059 Å2 to 1113 Å2 . Unexpectedly , as explicit water molecules were not used in our structural modeling or scoring , a water-mediated hydrogen bond to the peptide was introduced between the mutant residue αY26 and the side chain of the N-terminal glutamate of the peptide , in addition to a direct hydrogen bond between side chains ( Figure 4c ) . This polar network may explain the superior affinity of αY26 versus αW26 for ELA/HLA-A2 , despite the fact that they were predicted to be similar ( ZAFFI ) or αW26 was preferred ( Rosetta; Table 2 ) . In contrast , αD26W binds more strongly than αD26Y to AAG/HLA-A2 , which lacks the N-terminal peptide glutamate and hydrogen bonding capability at that side chain . In light of the water-mediated contacts observed in the mutant crystal structure , we re-ran simulations using explicit water molecules from the wild-type and mutant structures , but no improvement in correlation was observed ( Table S1 ) . The crystal structure of the YW variant bound to ELA/HLA-A2 allowed us to evaluate the performance of several structural modeling protocols . After least squares fitting of the backbone of the TCR and pMHC interface residues to the crystal structure , we compared positions of the modeled side chains to those in the crystal structure ( Figure 5 , with RMSDs in Table 3 ) . In addition to the NoMin and Min methods , we evaluated models generated using two intermediate minimization methods: MinSC ( minimizing interface side chains only ) and MinBB ( minimizing interface backbone atoms ) . Finally , we re-modeled the engineered side chains in the context of the mutant crystal structure ( NoMinMut ) to determine whether accurately positioned backbone and neighboring side chain atoms could improve modeling results . For modeling the side chain of αY26 , all protocols performed well in predicting the general orientation of the Tyr side chain , with NoMin outperforming the other protocols ( RMSD = 1 . 06 Å ) . Though generally accurate , all models exhibited a rotation in the aromatic ring and a slight shift in the OH group with respect to the crystal structure . As these errors were possibly due to the absence of explicit waters in the modeling omitting the water mediated hydrogen bonding observed in the YW/ELA/HLA-A2 crystal structure , we re-ran the NoMinMut simulation with water molecules from that structure , but found little improvement in RMSD ( 0 . 98 Å , versus 1 . 13 Å without water molecules ) . The predicted side chain conformations for the mutant βL98W were more variable than αY26 , including a flip of the aromatic rings in the Min and MinBB models , leading to relatively high RMSDs ( >2 Å ) relative to the experimentally determined structure for this residue . The structure modeled without minimization had a sub-optimal positioning of the Trp side chain ( tilted away from the pMHC ) ( Figure 5 ) , which improved substantially ( from 1 . 52 Å to 0 . 89 Å RMSD ) when modeled in the context of the backbone and side chains from the mutant crystal structure . This indicates that Rosetta's packing protocol is sensitive to small structural perturbations and accurate modeling of backbone and neighboring side chains can lead to improved predictions . In light of the lower accuracy of the ZAFFI scoring function on the measured DMF5 point mutants ( Figure 3 ) than for the A6 TCR , we performed a systematic evaluation of scoring functions to better predict DMF5 affinities while still maintaining accuracy with the set of A6 mutants . We included several statistical potentials in addition to the energetic and knowledge-based terms from the original ZAFFI study [14] . Given that minimization yielded false positive results for both ZAFFI and Rosetta functions ( Figure 3 ) and that unminimized structures more closely matched the YW-ELA/HLA-A2 crystal structure , we used unminimized models for this analysis . In addition to correlation with measured ΔΔGs , we evaluated scoring functions using receiver operating characteristic area under the curve ( AUC ) in order to judge discrimination of binding improvement without penalizing true negative or true positive outliers . We identified a scoring function ( referred to as ZAFFI 1 . 1 ) with a higher correlation ( 0 . 74 ) than ZAFFI ( 0 . 59 ) and Rosetta ( 0 . 72 ) for the set of DMF5 point mutants ( excluding the four αG28 outlier mutants ) , and high AUC values for both DMF5 and A6 mutants ( Table 4 and Figure S4 ) . Correlation P-values are included in Table 4 for all functions , highlighting significant predictive performance of ZAFFI 1 . 1 ( p<0 . 001 ) for both sets of data . ZAFFI 1 . 1 includes six terms: van der Waals attractive and repulsive components , desolvation , intra-residue clash , hydrogen bonding and Coulombic electrostatics . While its correlation with A6 TCR data ( 0 . 65 ) was not as high as the original ZAFFI function ( 0 . 77 ) , both the correlation and AUC are considerably higher than Rosetta on that set of data . Although a few outlier points persisted , including αG28P in the AAG/HLA-A2 interface , the overall success of this function demonstrates that a relatively simple scoring function and packing scheme can be used to model a large proportion of energetic changes in three designed TCR-pMHC interfaces . To examine the performance of this function in the context of other protein-protein interactions , we applied it to two large sets of interface point mutants ( 285 mutants each ) of two proteins designed de novo to target influenza hemagglutinin ( Table S4 ) , recently used in a collaborative effort to evaluate protein design algorithms as part of the protein docking experiment CAPRI [40] . We found that ZAFFI 1 . 1 ( with NoMin packing ) performed similarly to the other tested functions for scoring the HB36 mutants ( r = 0 . 36; p = 2 . 1×10−10 ) , while for HB80 mutants it outperformed all other functions ( r = 0 . 5; p<2 . 2×10−16 ) , with a Kendall tau rank correlation ( 0 . 38 ) higher than we achieved in the CAPRI experiment using a ZAFFI-related function ( 0 . 31 ) , where our Kendall correlation surpassed all other groups [40] .
Structure-based design of TCRs provides a means to improve upon low wild-type affinities for pMHC while maintaining , improving , or altering specificities for desired targeting capabilities . While some studies have determined the fine specificities of designed TCRs using biophysical [39] , [41] and cell-based [23] methods , here we demonstrated that point substitutions selected using structure-based methods can be used to efficiently engineer pMHC specificity and affinity . We then utilized structural modeling and x-ray crystallography to gain atomic-level insights into these substitutions . We achieved higher affinity improvements than previously reported in structure-based TCR design , with just two point substitutions resulting in an approximately 400-fold affinity improvement , versus 150-fold for four combined point mutants of the BC1 TCR selected using molecular mechanics [16] , and 100-fold for four combined point mutants of the A6 TCR selected using ZAFFI [14] . Despite the structural plasticity commonly observed in TCR-pMHC interfaces [42]–[45] , our computational modeling and crystal structure indicate that carefully selected point substitutions can improve pMHC affinity and modulate peptide specificity without grossly perturbing the interface structure . We note though that a broad extension this approach to other TCRs of interest will likely entail further refinement of the energy function based on measured data , in addition to improvements in high-resolution modeling of TCR-pMHC complexes [46] . Large-scale datasets of mutant binding affinities , including the CAPRI data we utilized to assess our design functions [40] , can provide possible training sets for re-weighting terms and derivation of energy-based statistical potentials that would add discriminating power and predictive breadth to the ZAFFI function . Additionally , our analysis of the YW-ELA/HLA-A2 structure indicates that there is room for improving structural modeling of mutant residues , with modeling of fine structural effects and bound water molecules representing two avenues for further development . The modulation of nonamer versus decamer specificity by many point mutants of the DMF5 TCR highlights the sensitive nature of TCR-antigen recognition , as well as the potential to fine-tune TCR recognition properties via structure-based design . We achieved a shift in specificity toward the nonameric MART-1 peptide via mutation of αG28 residues that were predicted to clash with the decameric E1 residue but would be accommodated in the cleft near the nonameric A1 , similar in concept to the “knob-in-to-hole” designs utilized to alter binding specificity in other protein-protein interfaces [47] . The clash with the decamer was overestimated using the NoMin modeling methods ( which had the greatest overall predictive success ) , thus leading to lower than anticipated specificity shifts; better modeling of clashes through judicious use of minimization ( avoiding false positive predictions as we observed ) could potentially reduce such errors . In contrast , we found an increase in specificity ( >4-fold ) toward the decameric peptide with the DMF5 double mutants YW and WW , resulting from the cooperativity of these mutants in the presence of the decamer . This peptide-dependent cooperative effect is a previously undescribed mechanism for shifting TCR specificity . As the structure of the YW/ELA/HLA-A2 complex did not suggest any major alterations in the binding interface compared to the wild-type complex , this effect may be dynamic in nature . As recently reported , the Mel5 TCR mutant α24β17 , which targets ELA/HLA-A2 with a 30 , 000-fold affinity improvement over wild-type , was found to retain peptide specificity , albeit towards alanine substituted ELA variants rather than between the ELA and AAG decameric/nonameric peptides [13] . In this case specificity was mediated through subtle solvent interactions . By modeling solvent and dynamic effects , as well as exploring explicit specificity design methods , such as multi-state design [48] , greater control of TCR specificity could be achieved via rational engineering . Three of the α chain mutants we tested were previously examined in the A6 TCR ( αD26W , αG28I , and αG28L ) [14] , whose CDR1α and CDR2α loops are identical to DMF5 due to the common use of the TRAV12-2 gene . αD26W improved pMHC affinity significantly for both TCRs , though to varying extents . On the other hand , the αG28 mutants improved the affinity of A6 modestly ( ∼2-fold ) but resulted in no change or weakened affinity with DMF5 . This behavior likely follows from the positions of the mutations , as the αG28 mutants are predicted to make extensive contacts with the varying N-terminus of the peptide , while αD26W would primarily target the same HLA-A2 site to improve affinities for all three pMHCs . However , both αD26 mutants of DMF5 still exhibited a measurable peptide dependence with ΔΔG , compared with , for instance , βL98W which had identical effects in the context of both MART-1 peptides . Data from more mutants and positions , as well as other TCR-pMHC systems , such as the Mel5 TCR which shares the TRAV12-2 gene with DMF5 and A6 and also targets ELA/HLA-A2 with a similar docking mode [49] , would help to further delineate the extent of any conserved effects of affinity-enhancing or destabilizing mutants . Indeed , the structure of the high affinity α24β17 Mel5 TCR mutant in complex with ELA/HLA-A2 [13] features a large hydrophobic substitution at position αD26 ( Phe ) , which closely matches the αD26Y conformation and the pMHC binding site in the YW/ELA/HLA-A2 structure ( Figure S5 ) , although as Mel5 α24β17 contained 18 additional substitutions , the energetic effect of αD26F alone is unclear . A more detailed study of the impact of affinity-enhancing mutations in germline CDRs would help to further probe TCR germline binding permissiveness suggested by a recent double mutant cycle deconstruction of the interface with the A6 TCR [50] . In conclusion , we have shown that rational , computational-based design offers the potential to simultaneously alter the efficacy and antigen targeting of a therapeutic TCR , potentially enabling the development of improved TCRs for adoptive cell therapy [51] or biotherapeutics [52] customized to bind antigens presented by tumors or virally infected cells from individual patients . Given the ongoing use of the DMF5 TCR in clinical trials for cancer immunotherapy , the higher-affinity YW variant of DMF5 generated here may also be of potential clinical benefit .
As with our previous study designing the A6 TCR [14] , we used the “interface” mode of Rosetta 2 . 0 . 2 [27] to model point mutations of the DMF5 TCR . Command line options were specified to include extra chi1 , chi2 , and chi3 rotamers ( “-extrachi_cutoff 1 -ex1 -ex2 -ex3” ) . Only the mutant side chain was repacked ( the default behavior of this mode ) while the protein backbone from the wild-type structure was retained . Rosetta predicted mutant structures as well as ΔΔGs , and the structures were then re-scored by our energetic scoring function ZAFFI to generate its own set of predicted ΔΔG scores . The ZAFFI filter , parameterized using the A6 TCR data and designed to remove false positive predictions that destabilized native electrostatic contacts , was not used in this study , given that our focus was evaluation and development of binding energy prediction functions , and the new system and protocols being explored would require tuning of the parameters of this filter . However , the filter function was used to corroborate avoidance of mutations in some cases ( such as hydrophobic mutants of αQ30 ) where key hydrogen bonds would likely be disrupted . To generate predictions of point mutants using side chain and/or backbone minimization we used Rosetta 2 . 3 , a more recent version of this program that includes minimization functionality in its interface mutagenesis mode . Minimization was specified using the command line flags ( “-min_interface -int_bb -int_chi” ) to perform minimization of interface backbone and side chain atoms in the wild type and mutant structures ( “Min” protocol ) , while just “-int_chi” or “-int_bb” was used to perform only side chain or backbone minimization ( “MinChi” , “MinBB” ) . Point mutant simulations with explicit water molecules taken from the input structure were also performed using Rosetta 2 . 3 , using the command line flag: “-read_hetero_h2o” . We analyzed residue backbone conformations in the bound and unbound DMF5 TCR structures using a Ramachandran plot analysis server [53] ( http://zlab . bu . edu/rama/ ) . DMF5 CDR positions with favorable backbone conformations for proline ( as well as favorable pre-proline conformations for the preceding residue ) , in addition to either improved or maintained pMHC affinity predicted for the proline mutant by at least one prediction method , were selected for experimental mutation to proline . Expression and refolding of soluble constructs of DMF5 TCRs and HLA-A2 were performed as previously described [29] , [54] . In brief , the TCR α- and β-chains , the HLA-A2 heavy chain , and β2-microglobulin ( β2m ) were generated in Escherichia coli as inclusion bodies , which were isolated and denatured in 8 M urea . TCR α- and β-chains were diluted in TCR refolding buffer ( 50 mM Tris ( pH 8 ) , 2 mM EDTA , 2 . 5 M urea , 9 . 6 mM cysteamine , 5 . 5 mM cystamine , 0 . 2 mM PMSF ) at a 1∶1 ratio . HLA-A2 and β2m were diluted in MHC refolding buffer ( 100 mM Tris ( pH 8 ) , 2 mM EDTA , 400 mM L-arginine , 6 . 3 mM cysteamine , 3 . 7 mM cystamine , 0 . 2 mM PMSF ) at a 1∶1 ratio in the presence of excess peptide . TCR and pMHC complexes were incubated for 24 h at 4°C . Afterward , complexes were desalted by dialysis at 4°C and room temperature respectively , then purified by anion exchange followed by size-exclusion chromatography . Refolded protein absorptions at 280 nm were measured spectroscopically and concentrations determined with appropriate extinction coefficients . Mutations in the DMF5 α- and β-chains were generated by PCR mutagenesis and confirmed by sequencing . Peptides and plasmids were commercially synthesized and purified ( Genscript ) . Surface plasmon resonance experiments were performed with a Biacore 3000 instrument using CM5 sensor chips . In all experiments , TCR was immobilized to the sensor chip via standard amine coupling and pMHC complex was injected as analyte . All samples were thoroughly dialyzed in HBS-EP buffer ( 20 mM HEPES ( pH 7 . 4 ) , 150 mM NaCl , 0 . 005% Nonidet P-20 ) , then degassed for at least 15 minutes prior to use . Steady-state experiments were performed with TCRs coupled onto the sensor chip at 1000–1500 response units . Injections of pMHC spanned a concentration range of 0 . 5–150 µM at flow rates of 5 µl/min at 25°C . Multiple data sets were globally fit using a 1∶1 Langmuir binding model utilizing BIAevaluation 4 . 1 . Kinetic titration experiments were performed with TCRs coupled at approximately 500 response units . A series of five ELA titrations , spanning 10–160 nM and 20–320 nM at 2-fold increase per titration , were flowed over YW and WW respectively . Flow rates of 30 µl/min were used at 25°C . Data were fit with a 1∶1 association model with drift using BIAevaluation [30] . Crystals of the DMF5 YW-ELA/HLA-A2 complexes were grown from 12% PEG 3350 , 0 . 25 M MgCl2 buffered with 0 . 1 M HEPES ( pH 8 . 0 ) at 25°C . Crystallization was performed using sitting drop/vapor diffusion . For cryoprotection , crystals were transferred into 20% glycerol/80% mother liquor for 30 s and immediately frozen in liquid nitrogen . Diffraction data were collected at the 22ID ( SER-CAT ) beamlines at the Advanced Photon Source , Argonne National Laboratories . Data reduction was performed with HKL2000 . The ternary complexes were solved by molecular replacement using PHENIX and Protein Data Bank ( PDB ) entry 3QDG as the reference model [29] . Rigid body refinement , followed by translation/libration/screw ( TLS ) refinement and multiple steps of restrained refinement were performed . TLS groups were automatically chosen by phenix . refine . Once defined , TLS parameters were included in all subsequent steps of the refinement . Anisotropic and bulk solvent corrections were taken into account throughout refinement . After TLS refinement , it was possible to unambiguously trace the position of peptides and TCR CDR loops in all structures against σA-weighted 2Fo-Fc maps . Evaluation of models and fitting to maps were performed using COOT [55] . The template structure check in WHATIF [56] and MolProbity [57] was used to evaluate the structures during and after refinement . Atomic positioning was verified with an iterative-build OMIT map calculated in PHENIX [36] . Structures were visualized using PyMOL [58] . Analysis of hydrogen bonds was performed with HBPlus [59] , using hydrogen-acceptor maximum distance of 2 . 7 Å and a donor-acceptor maximum distance of 3 . 6 Å . Solvent accessible surface areas were measured in Discovery Studio ( Accelrys Inc . ) using a probe radius of 1 . 4 Å . The structure has been deposited with the Protein Data Bank ( PDB ID 4L3E ) . ROC AUC analysis was performed using the CROC package [60] . Multi-linear regression to determine weighting of terms was performed as described previously , using 760 measured point mutants from four enzyme-inhibitor complexes [14] . However , we used van der Waals attractive and repulsive terms from Rosetta [27] rather than the corresponding terms from ZRANK [61] , as the former led to some improvement in performance across the tested systems . As with the original ZAFFI training , we removed mutants with high clash during training ( van der Waals repulsive score >580 , corresponding to 48 mutants removed out of 760 ) . We included a number of statistical potential terms for evaluation that were recently tested for binding affinity prediction [62] , though none led to substantial improvements in predictive performance in this context . The terms and weights for the retrained energy function ( ZAFFI 1 . 1 ) are: van der Waals attractive: 0 . 57 van der Waals repulsive: 0 . 0045 solvation: 0 . 58 hydrogen bonding: 1 . 2 intra-residue repulsion: 0 . 026 electrostatics: 0 . 03 Solvation , hydrogen bonding , and intra-residue repulsion terms were obtained from Rosetta ( along with the van der Waals terms as noted above ) , while the electrostatics term is the long-range Coulombic electrostatics energy from ZRANK [61] . All correlations ( with the exception of the Kendall tau rank correlations reported in Table S4 ) are Pearson correlations . P-values for correlations were calculated using the program R ( www . r-project . org ) . | T cell receptors ( TCRs ) play a major role in immunity , recognizing peptide antigens presented by major histocompatibility complex proteins . Due to their capacity to target intracellularly produced proteins and initiate cell killing , there is significant interest developing TCR-based therapeutic strategies , particularly towards cancer . A concern with TCRs is their weak-to-moderate affinities , which limits therapeutic potential . While in vitro evolution has been used to enhance TCR affinity , with sometimes spectacular results , these techniques can reduce peptide specificity and offer little control over affinity enhancements . Here we explored the use of structure-based computational design to enhance TCR affinity , which in principle can permit control over both specificity and affinity gains . We examined a clinically relevant TCR recently used in melanoma immunotherapy , identifying and characterizing mutations which enhanced affinity with no detectable impacts on binding specificity . We solved a crystal structure of our highest affinity designed TCR in complex with antigen , which indicated high accuracy of the structural modeling during the design process , and we critically evaluated several design protocols and functions to further improve design success . These results provide valuable insights into the use of computational design for TCRs . Lastly , the enhanced affinity variants identified may be of potential clinical benefit . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"immune",
"system",
"proteins",
"biochemistry",
"protein",
"interactions",
"proteins",
"protein",
"structure",
"biology",
"computational",
"biology",
"proteomics",
"protein",
"engineering",
"macromolecular",
"structure",
"analysis"
] | 2014 | Computational Design of the Affinity and Specificity of a Therapeutic T Cell Receptor |
The Ewing sarcoma family of tumors ( EFT ) is a group of highly malignant small round blue cell tumors occurring in children and young adults . We report here the largest genomic survey to date of 101 EFT ( 65 tumors and 36 cell lines ) . Using a combination of whole genome sequencing and targeted sequencing approaches , we discover that EFT has a very low mutational burden ( 0 . 15 mutations/Mb ) but frequent deleterious mutations in the cohesin complex subunit STAG2 ( 21 . 5% tumors , 44 . 4% cell lines ) , homozygous deletion of CDKN2A ( 13 . 8% and 50% ) and mutations of TP53 ( 6 . 2% and 71 . 9% ) . We additionally note an increased prevalence of the BRCA2 K3326X polymorphism in EFT patient samples ( 7 . 3% ) compared to population data ( OR 7 . 1 , p = 0 . 006 ) . Using whole transcriptome sequencing , we find that 11% of tumors pathologically diagnosed as EFT lack a typical EWSR1 fusion oncogene and that these tumors do not have a characteristic Ewing sarcoma gene expression signature . We identify samples harboring novel fusion genes including FUS-NCATc2 and CIC-FOXO4 that may represent distinct small round blue cell tumor variants . In an independent EFT tissue microarray cohort , we show that STAG2 loss as detected by immunohistochemistry may be associated with more advanced disease ( p = 0 . 15 ) and a modest decrease in overall survival ( p = 0 . 10 ) . These results significantly advance our understanding of the genomic and molecular underpinnings of Ewing sarcoma and provide a foundation towards further efforts to improve diagnosis , prognosis , and precision therapeutics testing .
The Ewing sarcoma family of tumors ( EFT ) is a group of malignant small round blue cell tumors that arise in bone or soft tissue . Ewing sarcoma ( ES ) is the second most common type of primary bone tumor to affect children and adolescents and accounts for 2 . 9% of all childhood cancers [1] . Despite advances in multidisciplinary treatment leading to improved outcomes over time for localized disease , long term survival remains poor for patients with metastatic or relapsed disease [2] , [3] . The pathological diagnosis of Ewing sarcoma is based on the finding of a small round blue cell tumor ( SRBCT ) that stains for MIC2 ( CD99 ) but has absence of markers that characterize the other pathologically defined SRBCT variants . In larger centers an EWSR1 break-apart probe is used to detect a fusion event involving this gene , but in most cases this test is not required for a diagnosis of Ewing sarcoma . In previous case series , most EFT cases express one of several reciprocal translocations , most commonly t ( 11;22 ) ( q24;q12 ) between the amino terminus of the EWSR1 gene and the carboxy terminus of the FLI1 gene found in 85–90% of cases [4] , [5] . A number of variant translocations between an alternate member of the TET family of RNA-binding proteins and/or an alternate member of the ETS family of transcription factors have also been described [6] . Additional structural chromosomal changes are frequently found in EFT , including gain of chromosome 1q , 2 , 8 , and 12 , and losses of 9p and 16q [7]–[9] . Recurrent mutations in known tumor suppressor genes have also been described , though with lower frequency . Most notably , homozygous deletions of CDKN2A ( which encodes p16INK4a ) have been detected in 10 to 30 percent of cases and TP53 mutations in 3 to 14 percent of cases [10]–[19] . Unfortunately , increased understanding of these molecular alterations has yet to produce successful targeted therapies . We therefore performed next generation sequencing on a panel of Ewing sarcoma family tumors and cell lines to identify additional molecular alterations associated with this aggressive cancer .
In the whole genome sequenced samples , we detected an average of 361 somatic mutations per tumor in non-repetitive regions and an average of 6 somatic mutations per tumor in protein coding regions ( 0 . 15 mutations/Mb of coding sequence ) , placing EFT at the low end of the mutation rate spectrum compared to previously reported malignancies [20] . There were no recurrent somatic small variants at the gene level within the 6 samples ( Table S3 ) . Somatic structural variants in this cohort were assessed using analysis of paired-end clones with discordant ends plus sequence coverage data from the whole genome sequencing data . Structural variants that involved a copy number change were verified using high-density SNP arrays with high degree of concordance ( 33/35 = 94 . 3% ) . Significant findings include previously reported alterations such as the characteristic EWSR1-FLI1 gene fusion detected in all 6 samples , CDKN2A homozygous deletion in 2 samples , and frequent chromosomal gains and losses . Novel findings include multiple focal areas of loss of heterozygosity , several out-of-frame gene fusions , and a tandem-duplication within the STAG2 gene in one sample ( Table S4 ) . There was an average of 17 structural variations per sample and no areas containing a high-density of structural variations ( i . e . chromothripsis ) were discovered . This number of structural variants is comparatively low relative to other pediatric tumors that have been evaluated by similar methods [21] , [22] . In summary , WGS of 6 EFTs revealed the characteristic EWSR1 fusion genes , low mutational burden and structural variations , but two tumors had loss of STAG2 ( frameshift variant in EWS2017 and focal tandem duplication in EWS2020 ) , and two had deletion of CDKN2A ( EWS2009 and EWS2020 ) ( Figure 1 , Figure S1 ) . In all 31 cell lines in which RNA sequencing was performed , an EWSR1-ETS family fusion transcript was detected ( Table S5 ) . In 62 tumor samples analyzed by RNA sequencing , 55 contained an EWSR1-ETS family fusion including 28 EWSR1-FLI1 type I ( 51% ) , 11 EWSR1-FLI1 type II ( 20% ) , 11 other EWSR1-FLI1 variants ( 20% ) , 3 EWSR1-ERG ( 5% ) , and 2 EWSR1-FEV fusions ( 4% ) ( Table S5 ) . Of the 7 tumors remaining without a EWSR1-ETS fusion , one sample was found to have a novel FUS-NCATc2 fusion ( Figure S2A ) . Another EWSR1-ETS negative sample contained a novel CIC-FOXO4 fusion ( Figure S2B ) . Both of these novel fusions are in-frame and contain the functional domains of the associated genes important for oncogenic potential . A third sample contained an ETV6-NTRK3 fusion ( Figure S2C ) , an alteration previously reported in association with infantile fibrosarcoma , congenital mesoblastic nephromas , secretory carcinoma of breast , mammary analogue secretory carcinoma of salivary gland , and radiation-associated thyroid cancer [23]–[27] . Hierarchical clustering based on RNA expression show that the 7 tumor samples without a TET-ETS fusion , including the 3 with the above alternate fusions , cluster separately from the vast majority of EWS-ETS fusion-positive samples ( Figure 2A ) . Additionally , these 7 samples show low expression of a collection of EWSR1-FLI1 target genes as well as low expression of a Ewing sarcoma gene signature previously reported by our group [28] ( Figure 2B , Figure S3 ) . We therefore consider these samples to be molecularly distinct from EFT and omitted them for the purposes of mutational frequency calculation . The patients with alternate fusions were noted to be clinically aggressive and to have slightly atypical histologic features , also suggesting a difference from classic Ewing sarcoma ( Text S1 ) . Through WGS and targeted sequencing we identified recurrent inactivating mutations of STAG2 . To extend our genomic analysis we performed capillary sequencing of the 33 coding exons of STAG2 in our tumor panel and in an expanded cell line panel to confirm our sequencing findings and to evaluate for additional mutations . In total , we discovered STAG2 alterations in 30 of 101 ( 29 . 7% ) of EFT samples including 14 of 65 ( 21 . 5% ) clinical tumor samples and 16 of 36 ( 44 . 4% ) cell lines ( Table 1 , Table S5 ) . Four of these cell lines had previously been reported to harbor STAG2 mutations [29] . Mutations were confirmed to be somatic in all tumor samples in which germline DNA from the same patient was available for comparison ( 7 tumors ) . The vast majority of the STAG2 variants are loss of function mutations , including 10 nonsense , 8 frameshift , 3 splice-site , and 5 structural variants , as well as a 5′ deletion previously found to cause absent protein expression ( Figure 3A , Table S5 ) [29] . The remaining three mutations of unclear functional consequence include a tumor with point mutation in the 3′ untranslated ( UTR ) region , a cell line with a missense mutation , and a cell line with a complex in-frame insertion ( 1 bp deletion replaced by a 7 bp insertion ) . Interestingly , 5 samples ( 4 tumors and 1 cell line ) contained the same nonsense mutation , R216X . Mutations of STAG2 ( located on the X chromosome ) were always heterozygous in samples with female genotype . In all STAG2 mutated cell lines in which RNA sequencing data was available , the altered allele was exclusively expressed ( 15 cell lines ) , indicating in the case of female samples that the X chromosome harboring the wild type STAG2 allele was silenced . In tumor samples , all evaluable STAG2 mutated samples showed preferential RNA expression of the mutant allele with varying amounts of wild type allele ( median variant allele frequency 0 . 78 ) , likely due to varying amounts of normal tissue contamination . STAG2 mRNA expression was significantly lower in samples with truncating mutations , likely due to nonsense-mediated decay ( Figure S4 ) . Seven additional EFT samples ( five tumors and two cell lines ) in which STAG2 mutation was not identified by our methods also had very low STAG2 expression comparable to samples with a truncating mutation in STAG2 ( Figure S4 ) . Immunohistochemistry ( IHC ) analysis with an antibody that binds to an epitope at the C-terminus of the STAG2 protein confirmed that EFT tumors with truncating STAG2 mutations have absent STAG2 protein expression , while the admixed non-neoplastic stromal and endothelial cells had robust expression ( Figure 4 ) . Tumors with wild-type STAG2 had robust expression of STAG2 protein specifically in cell nuclei as expected ( Figure 4 ) . Tissue microarrays ( TMA ) from an independent cohort of genetically confirmed Ewing sarcoma cases [30] were evaluated for STAG2 expression by IHC . In 210 evaluable cases , loss of STAG2 expression was found in 30 tumors ( 14 . 3% ) , 28 of which demonstrated complete absence of STAG2 protein and 2 of which demonstrated mosaic loss of STAG2 ( Figure S5 ) . Western blots performed on the EFT cell line panel demonstrated complete absence of STAG2 protein in 13 cell lines and altered protein in 3 cell lines , concordant with all 14 samples with inactivating mutations and the 2 samples with low RNA expression but no identified mutation ( Figure 5 , Table S6 ) . We identified TP53 mutation in 4 of 65 ( 6 . 2% ) EFT tumor samples and in 23 of 32 ( 71 . 9% ) EFT cell lines tested . Almost all of the TP53 mutations we discovered are previously reported pathologic variants and/or are truncating mutations ( nonsense , splice site , or frameshift ) ( Figure 3B , Table S5 ) . RNA expression analysis showed that there were 4 additional EFT samples ( 2 tumors , 2 cell lines ) in which TP53 mutation was not identified but had low TP53 expression similar to those with a truncating mutation ( Figure S6 ) . CDKN2A deletion was detected in 9 of 65 ( 13 . 8% ) EFT tumors and in 16 of 32 ( 50% ) EFT cell lines tested based on DNA and/or RNA sequencing coverage ( Figure S7 , Figure S8A ) . The semi-quantitative nature of the PCR reactions combined with varying amounts of normal contamination potentially results in a decreased sensitivity to detect this finding in our tumor samples . Western blots performed on the EFT cell lines demonstrated complete absence of p16INK4A expression in all cell lines in which deletion was detected ( Figure 5 , Table S6 ) . In summary , we found that STAG2 , TP53 or CDKN2A was altered in 57 of 97 ( 58 . 7% ) of EFT samples ( excluding tumors lacking an EWSR1-ETS fusion ) in which these 3 genes were sequenced by at least one technology ( Figure 6 ) . This count includes 26 of 65 ( 40 . 0% ) clinical tumor samples and 31 of 32 ( 96 . 9% ) cell lines . In the clinical tumor samples , these alterations were typically mutually exclusive in 19 of 26 ( 65 . 5% ) although several samples had STAG2 mutations in association with TP53 mutations or CDKN2A deletions ( Figure 6 ) . In addition to the targeted gene panel that was sequenced , we examined variants in all genes from our RNA sequencing data to look for other potentially oncogenic mutations ( Table S7 ) . We discovered several well-established cancer mutations in single samples , including a BRAF V600E mutation ( cell line A673 ) , a PI3KCA mutation ( cell line ES4 ) that has been recurrently found in multiple cancer types , and a RAD51 alteration ( tumor EWS101 ) associated with familial breast cancer [31] . In addition we discovered the BRCA2 K3326X polymorphism in 5 samples , 4 from patient tumors and one in a cell line ( TC-106 ) . The patient frequency of 4 of 55 ( 7 . 3% ) EFT tumor samples tested with this finding is statistically higher than expected given the population frequency of 12 of 1094 controls having this polymorphism in the 1000 genomes database ( OR 7 . 1 , p = 0 . 006 ) . In our tumor samples , this polymorphism was mutually exclusive with STAG2 , CDKN2A , and TP53 mutations , though overlapped with STAG2 expression loss in one case ( Figure 6 , Table S5 ) . Germline material was not available to assess whether these findings represent germline or somatic changes in our patients . We discovered one additional BRCA2 missense mutation ( S2186T ) that is previously unreported in dbSNP and is of uncertain clinical significance ( Table S7 ) . We did not identify additional altered genes in our sequencing that were predicted driver mutations . A significant relationship was noted between STAG2 loss and TP53 mutational status in the EFT cell lines . In the 16 cell lines with deleterious STAG2 alterations ( inactivating mutation or loss of expression ) , there were 14 ( 87 . 5% ) with TP53 mutations , all missense . In the 20 remaining cell lines with intact STAG2 , only 9 of the 16 ( 56 . 3% ) evaluated by sequencing contained a TP53 mutation , 5 of which were truncating and 4 of which were missense . Western blots demonstrated that 11 of 16 cell lines with deleterious STAG2 alteration had overexpression of p53 protein ( Figure 5 , Table S6 ) . Conversely , only 2 of 20 STAG2-intact cell lines had detectable p53 . Congruent with these findings , TP53 transcript levels from RNA sequencing were approximately 4-fold higher in cell lines with deleterious STAG2 alteration ( log2 FPKM 5 . 26 vs . 3 . 47 , p = 0 . 0023 ) ( Figure S9A ) . To assess the functional consequence of TP53 overexpression related to STAG2 mutation , we assessed the RNA expression levels of CDKN1A ( which encodes p21WAF1/CIP1 ) as a marker of p53 activity . As expected , cell lines with TP53 mutation or expression loss showed lower levels of CDKN1A transcript than TP53 wild type cell lines ( log2 FPKM 0 . 09 vs 4 . 85 , p = 0 . 0005 ) . STAG2 mutation in the EFT cell lines similarly predicted for decreased CDKN1A transcript expression compared to samples without a detected mutation ( log2 FPKM −0 . 20 vs 2 . 31 , p = 0 . 018 ) ( Figure S9B ) . Despite a much lower rate of concordant TP53 mutation , there was also a significant association between STAG2 mutation and decreased CDKN1A transcript expression in the tumor cohort ( log2 FPKM 4 . 08 vs 5 . 18 , p = 0 . 039 ) ( Figure S9C ) and a trend towards increased TP53 expression ( log2 FPKM 5 . 51 vs . 5 . 38 , p = 0 . 19 ) . In summary , STAG2 mutation was associated with higher TP53 and lower CDKN1A expression in both tumors and cell lines and associated with more frequent missense TP53 mutations in cell lines . Clinical characteristics of the tissue microarray cohort were evaluated in relation to STAG2 IHC status . The 210 evaluable cases included 154 primary tumors , 46 recurrent/metastatic samples , and 10 tumors with limited clinical information ( Table S8 ) . STAG2 expression loss was more common in recurrent/metastatic samples than primary samples , though this difference did not reach statistical significance ( 21 . 7% vs . 12 . 3% , p = 0 . 15 ) . In 110 primary tumor samples in which clinical outcomes data was available , there was a trend towards a modest decrease in overall survival in patients whose tumors had STAG2 loss ( p = 0 . 10 ) ; this evaluation was limited , however , by small numbers of patients with STAG2 negative tumors in the analysis ( Figure S10 ) . Clinical information from the sequencing cohort was also analyzed , though survival information was not available . We found no significant differences in age , gender , stage , or primary tumor site ( extremity vs . non-extremity ) between STAG2 mutated and wild-type samples , though numbers were small in these comparisons ( Table S9 ) .
To our knowledge , this is the first and largest report to utilize next-generation sequencing technology to characterize the genomic landscape of Ewing sarcoma family of tumors and evaluate for recurrent mutations . We find a very low somatic mutational rate in EFT compared to most previously reported tumor types . We hypothesize this to be the case for multiple reasons . First , it appears that a number of pediatric tumor subtypes tend to have lower mutation rates than those reported in adult cancer [21] , [32]–[34] . This may be due in part to the shorter amount of time that the precursor cell has to accumulate passenger mutations during normal cell division but may also represent a fundamental difference common to pediatric cancers . For example , it is possible that pediatric cancers may be more epigenetically driven compared to adult cancers and therefore require a lesser genetic-level contribution to oncogenesis . Second , the low mutation rate of Ewing sarcoma even amongst several reported pediatric cancer types may reflect a fundamental characteristic of fusion-driven cancers . This is in keeping with differences noted between fusion-positive and fusion-negative rhabdomyosarcomas reported by our group and others [22] , [35] . Interestingly , we found by RNA sequencing that a significant number ( 11% ) of our tumor samples that were pathologically diagnosed as Ewing sarcoma family tumors lacked a characteristic TET-ETS fusion and appeared to be molecularly distinct from EFT by expression profile . Within this group , we report two novel fusions , FUS-NCATc2 and CIC-FOXO4 , that are in-frame and may be oncogenic based on available literature regarding the function of the genes involved . NCATc2 is a non-ETS family transcription factor that has recently been described as an alternate fusion partner to EWSR1 in a small series of “Ewing sarcoma-like” tumors [36] , but has not previously been reported to partner with the alternate TET family member FUS . CIC gene rearrangements , particularly CIC-DUX4 fusions , have been described in a group of aggressive undifferentiated small blue round cell sarcomas thought to be distinct from Ewing sarcoma [37] . FOXO4 , a forkhead family transcription factor , has been described as an uncommon fusion partner to MLL in acute leukemias [38] and as a rare PAX-gene fusion partner in rhabdomyosarcoma [39] . We discovered one additional fusion , ETV6-NTRK3 , which has been reported in other cancer types but not in EFT [23]–[27] . Whether these tumors should be considered as a variant of EFT or a distinct entity is debatable . Our RNA data suggests that these alternate fusion samples , as well as the other TET-ETS fusion negative samples , have a distinct expression pattern from the other EFT tumors and do not match well to a previously reported EFT expression signature [28] . Practically , the rarity of these variants amongst an already uncommon disease will make clinical distinction difficult . In our survey for genetic alterations , we discovered STAG2 mutations in 21 . 5% of Ewing sarcoma family tumor samples and 44 . 4% of EFT cell lines tested , the vast majority of which are inactivating mutations . STAG2 protein detection by IHC in an independent tumor cohort showed STAG2 loss in 14 . 3% of tumors . While immunohistochemistry will identify all tumors with homozygous deletions and truncating mutations , it will not detect tumors harboring missense mutations , in-frame insertions or deletions , or duplication events . This may help to explain the small discrepancy between our sequencing and immunohistochemical analyses . STAG2 mutation has previously been reported in one Ewing sarcoma tumor and in multiple EFT cell lines [29] and has additionally been reported as a recurrently mutated tumor suppressor gene in other tumor types including glioblastoma , urothelial carcinoma , and acute myeloid leukemia [29] , [40]–[44] . Mutations in TP53 and CDKN2A were found in frequencies similar to that previously reported [10]–[19] . In total we found that 40% of EFT clinical tumors and 97% of EFT cell lines have disruption of STAG2 , TP53 or CDKN2A . The striking difference in mutational frequencies between tumors and cell lines , particularly in TP53 , may be a result of culture conditions and the process of immortalization . Despite these frequency differences , the increased molecular characterization of a large selection of EFT cell lines evaluated in this study provides an invaluable resource for further study . In addition to the recurrent mutations in STAG2 , TP53 and CDKN2A , we found a high prevalence of the BRCA2 K3326X polymorphism , seen in 7 . 3% of our clinical tumor samples . Occurring in approximately 1% of the general population , this premature stop codon has not been shown to confer an increased risk of breast or ovarian cancer [45] and is classified as a benign variant by the International Agency for Research on Cancer Unclassified Genetic Variants Working Group [46] . In contrast , groups studying lung [47] , pancreatic [48] , and squamous esophageal cancers [49] have all reported a significantly increased rate of this polymorphism in the germline DNA of patients with these cancer types . In our cohort , as only tumor material was evaluated for this finding , we could not distinguish whether this was a germline or somatic change . Further study is warranted to clarify this aspect and to confirm the association . STAG2 encodes a subunit of cohesin , a structural protein complex involved in chromosomal organization and so named due to its function of creating “cohesion” between sister chromatids after DNA replication . In addition to STAG2 , other recurrent alterations in subunits of this complex have been reported across a number of cancer types [42] , [50] , [51] . Potentially , the oncologic mechanism for cohesin mutation is disrupted chromosomal segregation during mitosis leading to accumulation of structural mutations and aneuploidy [29] . Though we find EFT to have a low rate of aneuploidy overall in our comprehensively characterized WGS cohort , further work is indicated to clarify whether or not a STAG2 mutation is linked to increased aneuploidy in this tumor histology . Cohesin is also known to play a regulatory role in transcription [52] and is essential for recombinant-based DNA repair mechanisms [53] , though it remains to be seen if and how much each of these essential cellular processes are responsible for the oncologic transformation resulting from cohesin deficit . In our evaluation of the cellular impact of STAG2 in EFT , we note a significant intersection of STAG2 mutation with alteration of the TP53 pathway . We find a strong correlation between STAG2 loss and overexpression of p53 in EFT cell lines . We note that this overexpressed p53 protein very frequently contains a pathogenic missense mutation . STAG2 mutated samples also had low RNA expressional levels of CDKN1A ( encoding p21WAF1/CIP1 ) , a well-established mediator of p53 tumor suppressor activity [54] . Taken together these data suggest that transcriptional dysregulation of the p53-p21 axis may play a role in STAG2-mediated oncogenesis , at least in EFT cell lines . Though there was less overlap between STAG2 mutation and TP53 mutation in the sequenced tumor cohort , we noted the same pattern of decreased CDKN1A expression in STAG2 mutant samples . We found STAG2 loss to be more common in cell lines than tumors , more frequent in metastatic or recurrent disease than primary tumors , and to be associated with a trend towards modestly decreased survival . Given the significant percentage of tumors harboring a STAG2 mutation in this cancer type , further investigation into the oncogenic mechanism , clinical consequence , as well as strategies for directed therapy are warranted . For example , preclinical data suggest that cohesin deficiency may increase sensitivity to poly ( ADP–ribose ) polymerase ( PARP ) inhibition [55] , a drug class that has also been identified by systematic screening to be effective in Ewing sarcoma cell lines [56] , and that is undergoing clinical testing in this tumor type . Additionally , future sequencing efforts should be extended to evaluate for alternate routes to cohesin deficiency in EFT . This study demonstrates that at least a subset of Ewing's sarcoma is not a single hit disease driven solely by a EWS-ETS fusion gene , but rather is a genetically complex disease which harbors additional recurrent genetic alterations that likely contribute to the pathogenesis of EFT . Further studies will be needed to determine if the presence of these additional genetic aberrations will impact the sensitivity/resistance to small molecule inhibitors of EWS-FLI1 or PARP that are currently in development and early phase clinical trials .
All specimens for sequencing were obtained from patients with appropriate consent from the local institutional review board in accordance with the Children's Oncology Group and the National Cancer Institute . Clinical samples were obtained from collaborations with the Cooperative Human Tissue Network , the Children's Hospital of Westmead , Australia , the Children's Oncology Group , and the National Institutes of Health Clinical Center . Tumors were classified as a Ewing sarcoma family tumor by a sarcoma pathologist and the host institution using standing histological techniques . Fifty-two tumors were from the primary disease site and had not been exposed to previous treatment . Fifteen tumors were from recurrent/metastatic sites and for five tumors we lacked this clinical information . Clinical and pathological data for the sequencing cohort are summarized in Table S9 . Tumor samples were evaluated by a pathologist for the presence of more than 70% tumor content before DNA/RNA extraction and sequencing . DNA was extracted from qualifying tumor samples and matched blood using either AllPrep Mini ( Qiagen ) or Agencourt Genefind v2 ( Beckman Coulter ) DNA extraction kits . RNA was extracted using the RNeasy Micro Kits according to the manufacturer's protocol ( Quiagen ) . Genotyping confirmed independence of these samples . All EFT cell lines used in the study underwent short tandem repeat ( STR ) profiling for independence testing and all were confirmed to have a unique profile . This characterization is described in detail in Table S10 . Tissue microarrays were obtained from an independent cohort of genetically confirmed Ewing sarcoma cases [30] . The associated clinical information is summarized in Table S8 . Whole genome paired-end sequencing was performed using the Complete Genomics platform . Data analysis was accomplished using the CGA tools package v2 . 0 [57] , ANNOVAR v2012-05-25 [58] , and Circos v0 . 52 [59] in build hg19 . Somatic variants were determined first by comparison to the matched normal DNA . To remove artifacts specific to the sequencing platform , we eliminated any somatic variants also found in normal samples [50 in-house samples and 69 Complete Genomics samples ( http://www . completegenomics . com/public-data/69-Genomes/ ) ] . The Somatic Score ( http://media . completegenomics . com/documents/DataFileFormats+Cancer+Pipeline+2 . 0 . pdf ) is based on a Bayesian model and takes account of read depth , base call quality , mapping/alignment probabilities , and measured priors on sequencing error rate for both the germline and tumor variants . Verification by Sanger sequencing was performed on all high-confidence somatic variant calls ( by default Somatic Score> = −10 ) affecting protein coding or a splice site ( SNVs , substitutions , insertions , deletions ) , including 55 SNVs . We determined that more stringent somatic score cut-off was required in our cohort to achieve adequate positive predictive value of variants calls , likely due to the low mutation rate in our tumor type . Relative to all high-confidence variant calls we established a sensitivity of 86 . 7% and specificity of 90 . 7% for a Somatic Score cut-off of 3 . Somatic mutations at or above this score and all verified mutations with lower scores were used for further analysis . The Complete Genomics somatic copy number segmentation is based on 2-kb windows and utilizes coverage in the matched germline sample for normalization of the tumor sample coverage . Lesser allele fraction ( LAF ) calculations are based on allele read counts in the tumor at loci that are called heterozygous in the matched germline sample . In addition to default filtering done by the Complete Genomics segmentation algorithm , copy number variants were considered high-confidence if they were either large ( > = 10 kb AND containing > = 10 heterozygous SNPs for LAF calculation ) OR highly altered ( homozygous deletions and focal amplifications > = 5 copies ) OR supported by somatic junction ( s ) ( ex: junction detected spanning both ends of a region of LOH ) . Somatic junctions were called using CGAtools and junctions were filtered by footprints smaller than 70 bases , less than 10 discordant mate pairs , under-represented repeats , and presence in the baseline set of 69 Complete Genomics genomes . We additionally filtered junctions that were present in 50 in-house germline DNA samples that were sequenced on the same platform . Genomic sequencing was performed using a custom multiplex PCR designed to include the entire coding sequence of the majority of altered genes in the whole-genome sequencing discovery cohort as well as TP53 and in total encompassing 106 . 3 kB of target region ( Table S11 ) . Primers for the targeted sequencing were designed using the Ion Ampliseq designer ( Life Technologies ) and sequencing was performed on the IonTorrent PGM ( Life Technologies ) . PGM sequencing data was analyzed using Torrent Suite software v3 . 2 ( Life Technologies ) and ANNOVAR . Genomic sequencing was performed to an average mean coverage of 311× and variants were called using high-confidence thresholds and filtered to include only those that are protein altering and unreported or rare ( population allele frequency <0 . 005 ) in the dbSNP and 1000 genomes databases . Mutations of interest were verified by capillary sequencing with a <5% false positive rate . Sequence coverage data was calculated at a position , exon and gene levels to look for structural alterations of the recurrently mutated genes ( Figure S8 ) . Coverage data was visualized using the Integrated Genomics Viewer . PolyA selected RNA libraries were prepared for sequencing on the Illumina HiSeq2000 using TruSeq v3 chemistry according to the manufacturer's protocol ( Illumina ) . RNA sequencing was performed with an average yield of 18 . 6 Gb per sample . Raw reads were mapped using to ENSEMBL reference ( hg19 ) using TopHat2 . 0 [60] . Fusion analysis was done using TopHat 2 . 0 and DeFuse 0 . 6 [61] . The 3 alternated fusions described were confirmed using RT-PCR using flanking primers and Sanger sequencing of the resultant product . Expression FPKM results were obtained at both gene and transcript level using CuffLinks 2 . 1 [62] . The log2 FPKM expression results from TopHat mapping were median-normalized using in-house data from 63 normal tissue samples . Exon level expression was calculated using the formula RPKM = ( r * 109 ) / ( f * R ) , with r being the number of reads mapped to an exon , f being the exon length , and R being the total read count of the sample . Hierarchical clustering was performed on normalized log2 FPKM expression values at the gene level using Euclidean distance and Ward agglomeration method . For variant detection , samtools ( http://samtools . sourceforge . net/ ) is used to count the number of reads uniquely mapped to a position found as variant in DNA sequencing of the same sample or a position of interest based on a mutation being present in the TCGA ( http://cancergenome . nih . gov/ ) or compared to the reference genome hg19 in genes of interest . If there are reads supporting a variant base then the total reads supporting it are counted and variant allele frequency is calculated . SNP arrays were performed on the 6 tumor whole genome sequencing cohort to confirm copy number findings . SNP array analysis was conducted on HumanOmni2 . 5 or HumanOmni5 arrays ( Illumina ) and the data were analyzed with GenomeStudio ( Illumina ) and Nexus Copy Number v7 ( Biodiscovery Inc . ) . Copy number state and allelic ratio was manually assessed in all areas of copy number variation and structural variation predicted by WGS and was concordant with WGS prediction in 33/35 ( 94% ) . Individual exons of STAG2 were PCR amplified from genomic DNA using the conditions and primer pairs previously described [29] . PCR products were purified using the Exo/SAP method followed by a Sephadex spin column . Sequencing reactions were performed using BigDye v3 . 1 ( Applied Biosystems ) using an M13F primer and were analyzed on an Applied Biosystems 3730×l capillary sequencer . Sequences were analyzed using Mutation Surveyor ( SoftGenetics ) . Traces with putative mutations were reamplified and sequenced from both tumor and matched normal DNA from blood when available . A mouse monoclonal antibody to STAG2 from Santa Cruz Biotechnology ( clone J-12 , sc-81852 ) was used at a dilution of 1∶100 . Immunostaining was performed in an automated immunostainer ( Leica Bond-Max ) following heat-induced antigen retrieval for 30 min in high pH epitope retrieval buffer ( Bond-Max ) . Primary antibody was applied for 30 min , and Bond-Max polymer was applied for 15 min . Diaminobenzidine was used as the chromogen , and samples were counterstained with hematoxylin . Samples in which both the tumor and normal cells failed to stain for STAG2 were considered antigenically non-viable and were excluded from the analysis . Primary antibodies used were STAG2 clone J-12 ( Santa Cruz Biotechnology ) , p53 clone 7F5 ( Cell Signaling ) , p16 ( BD Pharmingen #554079 ) , p21 clone DCS60 ( Cell Signaling ) , and α-tubulin Ab-2 clone DM1A ( Neomarkers ) . Protein was isolated from EFT cell lines in RIPA buffer , resolved by SDS-PAGE , and immunoblotted following standard biochemical techniques . For the BRCA2 K3326X polymorphism , a two-tailed Fisher Exact Test was used to calculate p-value for the Odds Ratio significantly different from 1 . For RNA expressional analysis , a two-tailed student T Test assuming unequal variances was used to calculate a p-value for difference in population means . For tissue microarrays , the p-value for differences in frequency of STAG2 mutation in primary and recurrent/metastatic samples was calculated using two-tailed Fischer Exact Test . P-value for association of STAG2 expression with overall survival was calculated using univariate analysis . | The Ewing sarcoma family of tumors is a group of aggressive cancers that primarily affects the pediatric and young adult population . Increasingly , genomics are being used to better define the disease biology and to identify targets for therapy in many cancer types . Here , we report one of the first and largest genomic studies to date in the Ewing sarcoma family of tumors . Using a combination of modern sequencing techniques in >100 samples , we discover that Ewing sarcomas have a genome that is less complex compared to most cancer types previously surveyed . We find that this cancer is frequently affected by mutations in STAG2 , a gene that has recently gained attention due to its importance in the biology of several cancer types . We show that Ewing sarcoma patients whose tumors are affected by STAG2 loss may have a worse prognosis . Additionally , we identify a subset of tumors that were diagnosed as Ewing sarcoma that appear to be distinct from the majority based on genetic and molecular characteristics . Our findings help to define the genetic landscape of Ewing sarcoma and provide a starting point for improving individualization of diagnosis , prognosis and treatment in this cancer . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"oncology",
"medicine",
"and",
"health",
"sciences",
"cancer",
"genetics",
"ewing",
"sarcoma",
"genetics",
"biology",
"and",
"life",
"sciences",
"cancers",
"and",
"neoplasms",
"sarcomas"
] | 2014 | The Genomic Landscape of the Ewing Sarcoma Family of Tumors Reveals Recurrent STAG2 Mutation |
The tempo and mode of human knowledge expansion is an enduring yet poorly understood topic . Through a temporal network analysis of three decades of discoveries of protein interactions and genetic interactions in baker's yeast , we show that the growth of scientific knowledge is exponential over time and that important subjects tend to be studied earlier . However , expansions of different domains of knowledge are highly heterogeneous and episodic such that the temporal turnover of knowledge hubs is much greater than expected by chance . Familiar subjects are preferentially studied over new subjects , leading to a reduced pace of innovation . While research is increasingly done in teams , the number of discoveries per researcher is greater in smaller teams . These findings reveal collective human behaviors in scientific research and help design better strategies in future knowledge exploration .
Scientific knowledge refers to the body of facts and principles that are known in a given field . Modern civilization is built on the knowledge that humans have acquired about the world they live in , and the future of the human species and society critically depends on further accumulation of scientific knowledge . Patterns and mechanisms of human knowledge growth are jointly determined by the intrinsic structure of knowledge and human behaviors in knowledge exploration . Although such behaviors are of interest to many scientists including philosophers [1] , [2] , sociologists [3] , anthropologists [4] , economists [5] , physicists [6] , and psychologists [7] , they are poorly studied , due primarily to the lack of ideal cases in which ( i ) the structure of the knowledge is known , ( ii ) the knowledge is quantifiable , and ( iii ) the process of knowledge discovery is well understood and documented . As biologists , we notice that the above three requirements are all met for biological knowledge of the baker's yeast Saccharomyces cerevisiae . Knowledge can be described largely as relationships among a set of subjects . Over the past three decades , scientists have substantively deepened their understanding of yeast biology through the study of interactions among its ∼6000 genes [8] . By the end of 2007 , over 73 , 000 yeast gene-gene interactions had been discovered and documented in ∼5 , 400 publications authored by 11 , 238 researchers ( see Materials and Methods ) . Much of the structure of the knowledge about yeast biology can be described as a gene-gene interaction network , where the unit of knowledge is an interaction . Scientific publications record the approximate date of each relevant discovery , as well as the methodology used . As a case study , we here analyze the temporal growth of the known yeast gene-gene interactions to understand the tempo and mode of scientific knowledge expansion .
Gene-gene interactions are separated into two types: genetic interactions ( GIs ) and protein-protein interactions ( PPIs ) [9] . Two genes are said to interact genetically if the effect of one gene on a trait is masked or enhanced by the other . Two genes are said to have a PPI if their protein products physically bind to each other stably or transiently . The data we considered contain 37 , 809 PPIs among 4 , 913 genes and 35 , 231 GIs among 3 , 743 genes , respectively ( see Materials and Methods ) . Because of the difference in the nature of PPIs and GIs , we study the yeast PPI and GI networks separately . The PPI data were published from year-1982 to 2007 , spanning 26 years , while the GI data were published from year-1977 to 2007 , spanning 31 years ( see Materials and Methods ) . The number of new interactions discovered per year increased approximately exponentially over time ( Figure 1 ) , and there is no apparent sign of slowing of this exponential growth at present . The exponential growth can be attributed to the increased number of studies per year and/or the enhanced productivity per study over time ( Figure 2 ) . P ( k ) , the probability that a study discovers k novel interactions , is proportional to k−r , where r = 1 . 79 and 1 . 84 for PPIs and GIs , respectively , indicating that the per-study productivity roughly follows a power-law distribution ( Figure 3 and Figure S1 ) . We also observed that the number of co-authors per study increased over time ( Figure 4 ) , reflecting a general trend of increased collaboration in scientific research [10] , [11] . Increase of productivity per author over time is not significant for PPIs , but significant for GIs ( Figure S2 ) . However , within virtually every year , per-author productivity is strongly negatively correlated with the number of co-authors of the study ( Figure 5A and Table S1 ) , suggesting that small research teams are more efficient than large teams at all times . Considering the possibility that researchers of small teams may publish fewer papers than those of large teams , we calculated accumulated productivity per-author in a five-year window . Again , authors of small teams consistently outperform those of large teams ( Table S2 ) and this result remains qualitatively unchanged even when we consider the accumulated productivity of only those researchers who served at least once as the last author of a study in a five-year window ( Table S3 ) . However , the negative correlation between the productivity of a researcher and his/her mean team size appears to be weakening over the years ( Figure 5B and Tables S1 , S2 and S3 ) . The ∼6000 yeast genes have been individually deleted to examine their functional importance , which is defined by the amount of reduction in the fitness of yeast caused by each deletion [12] . We traced the first year of appearance ( birth year ) of each gene in the PPI and GI networks , and found that genes appearing earlier in the networks ( old genes ) are more important than those appearing later ( young genes ) ( Figure 6 ) . One possible explanation of this phenomenon is that a gene's importance arises from the sheer number of its interactions [13]–[15]; if each interaction has the same probability of discovery , highly interactive genes are incorporated into the knowledge network earlier simply because they have more interactions . However , we found that old genes are more important than young genes even when the number of now known interactions per gene is controlled for ( Spearman's partial correlation coefficient ρ = 0 . 13 , P = 1 . 8×10−17 for the PPI network; ρ = 0 . 10 , P = 5 . 3×10−9 for the GI network; Table 1 ) . This result remains unchanged when we further control for the level of gene expression ( Table 1 ) . Thus , important genes are studied earlier not simply because of their large numbers of interactions , but also because of their phenotypic importance that is beyond what is predicted from their numbers of interactions . During the growth of the yeast biological knowledge network , a new interaction can introduce zero , one , or two genes into the network . Generally speaking , follow-up studies tend to discover interactions involving “pre-existing” genes while novel studies tend to discover interactions between previously “uncatalogued” genes [16] . We separately simulated the growths of yeast PPI and GI networks by randomizing the birth years of all interactions while conserving the number of new interactions discovered each year . Interestingly , the growth of gene number in the real networks lags behind the random expectation for many years ( Figure 7 ) , suggesting that , compared with the random process , actual researchers tend to focus on finding properties of known genes rather than those of new genes . We conducted 1000 simulations of random growth and found that the number of genes is 655 . 1±10 at 1995 , the mid-point of PPI network growth , and this number is 676 . 1±14 . 6 for GI network at its mid-point of growth . Both numbers are significantly ( P<0 . 001 ) larger than the observed numbers ( 390 for PPI network and 454 for GI network ) in real growth . We also observed that the real growth pattern relative to the random pattern was reversed in recent years . However , this reserve is due to the fixation of total numbers of genes and interactions at year-2007 and does not suggest that the tendency of “novelty-aversion” has been reversed in research . The “novelty-aversion” phenomenon may arise from a high cost of novelty-seeking research and/or a high reward ( or desire ) for studying previously discovered genes [17] . As a consequence , the cohesiveness of the actual knowledge network is higher than that of a randomly growing network during the early years of yeast research ( Figure S3 ) . Many complex networks are naturally divided into communities or modules , such that interactions within modules are much denser than those between modules [18] . The temporal PPI and GI data allow us to study the relative growths of different modules in a knowledge network compared to random growths . We identified 12 and 16 modules from the present-day PPI and GI networks , respectively [15] ( see Materials and Methods ) . We transformed the network growth information into module growths by assigning one unit for every involved gene of a new interaction to the module that the gene belongs to . We then measured the deviation of the growth of each module from its expectation under homogenous growth , for each temporal PPI or GI network . Interestingly , although the network growth was contributed simultaneously by multiple modules in many years , the among-module heterogeneity in growth is striking , compared to random growths ( Figure 8 ) . For example , 4 . 7% of the PPI network growth was contributed by module #12 in year-2000 , but this number becomes 70 . 8% in year-2007 . The fluctuation index measured by mean Euclidean distance ( see Materials and Methods ) among these distributions is 0 . 40 and 0 . 42 for PPI and GI networks , respectively . Both are significantly larger than the expectations from simulated random growths of PPI ( 0 . 26±0 . 03 ) and GI ( 0 . 18±0 . 02 ) networks ( P<0 . 001; Figure 9 ) . This heterogeneous and episodic growth also leads to among-module variation in the maturation process of modules ( Figure 10 ) . One wonders whether the observed heterogeneous and episodic growth of PPI and GI modules is owing to some recent large-scale studies that focused on genes involved in specific cellular functions; PPIs and GIs discovered from such studies are expected to be localized to certain knowledge modules rather than evenly distributed among all modules . To examine the effect of large-scale studies , we separately examined the network growth before and after year-1999 . In the pre-1999 years , there was only 1 paper reporting >50 PPIs and 8 papers each reporting 20–50 PPIs , among the 919 papers on PPIs . Similarly , in this period , there were only 5 papers each reporting 20–50 GIs , among 1633 papers on GIs . In the post-1999 years , there were many large-scale studies . However , heterogeneous episodic growth of modules is found in both periods ( Table S4 ) . Thus , our observation is not simply a result of recent large-scale studies of specific cellular functions . The heterogeneous and episodic growth of knowledge modules has an important consequence . Like many complex networks [19] , connectivity is highly variable among nodes in the yeast PPI and GI networks . Most genes have one or a few interactions while a small fraction of genes have a very large number of interactions ( Figure S4 ) . Highly connected nodes ( hubs ) are known to be of both structural and functional importance to a network [13] , [14] , [19] ( see also Table 1 ) . Therefore , recognizing true hubs earlier would speed up the study of the network structure and function . However , hubs in today's network may not be hubs in the previous year's network and it is important to examine how stable hubs are during network growth . We arbitrarily define hubs in a given year as genes whose total connectivities in a network are among the top 10% of all available genes within the network at that time ( only temporal networks with at least 50 genes are considered ) . We examined hub turnover in each year by computing the proportion of temporal hubs that become non-hubs in the following year . For both the PPI and GI networks , hub turnover rates are usually high ( Figure 11 ) . Surprisingly , hub stability did not increase with the growth of the network . For example , 32 . 5% of year-2006 GI hubs became non-hubs in 2007 , and the corresponding number was 15 . 5% for year-2006 PPI hubs . This suggests that under the current mode of knowledge growth , it is difficult to predict true hubs before completion of network growth . By contrast , in the simulated random network growth , there is a trend of reduction in hub turnover over time . For example , in the GI network the turnover rate became <10% after year-1997 and <1% between year-2006 and 2007 . The birth of temporal hubs appears to be strongly associated with heterogeneous expansions of modules ( Figure 12 ) . The heterogeneous and episodic growth of network modules , and the related rapid hub turnover , are likely caused by a high reward ( e . g . , high-profile publications or large grants ) for or biased interest in studying certain topics at certain times . For example , when a human disease-associated gene is identified , its yeast ortholog could be subject to intense studies immediately . Human syntaxin 8 was cloned in 1999 [20] and characterized as a member of the t-SNARE ( target soluble N-ethylmaleimide sensitive factor attachment protein receptor ) superfamily involved in vesicular trafficking and docking , a critical cellular process implicated in many human diseases [21]–[23] . Soon after the discovery , its yeast ortholog YAL014C was investigated and its 5 PPIs were identified by two studies in 2000 [24] and 2002 [25] , respectively . In addition , different parts of a knowledge network are more likely to be discovered by different technologies that are invented at different times ( Figure 13 ) . For instance , in discovering PPIs , affinity approaches [26] tend to identify stable protein complexes while yeast two-hybrid assays [27] find dynamic interactions well . To further demonstrate this point , we directly compared two genome-wide studies that used either yeast two-hybrid assays [28] or affinity approaches [29] to discover PPIs . The across-module PPI distributions of the two studies are significantly different ( Table S5 ) . These results illustrate the importance of employing diverse approaches in knowledge exploration .
Although the PPI and GI networks analyzed here are still growing , they have been studied for ∼30 years and have encompassed most yeast genes . Thus , they serve as relatively good representations of the true and complete networks . For example , it is believed that we have already discovered ∼50% of all yeast PPIs [30] . Nevertheless , it is possible that we may have omitted some discoveries , although the BioGRID database , from which our data are acquired , is based on extensive literature searches [31] . To evaluate the potential effect of such omissions , we randomly excluded 10% of studies and repeated our analyses , and found that all major conclusions hold ( data not shown ) . It should also be pointed out that , although the unbiased random network growth was based on the year-2007 networks , all principles should be applicable to the final true and complete networks . The exponential growth shown in Figure 1 and the assumption that ∼50% of all PPIs in yeast have been identified predict that almost all yeast PPIs will have been discovered by year-2009 , if the fraction of false positive discoveries does not increase with the rate of discovery . However , it is fully expected that both the current and future PPI and GI networks contain false interactions . Because false understanding exists in any type of knowledge , it will be interesting to study how false interactions affect the discoveries of true interactions . Unfortunately , BioGRID contains no information about previously reported interactions that are later dismissed . In fact , it is extremely difficult to falsify a previously reported interaction , because ( i ) the falsification requires one to test an interaction with exactly the same technique and condition as used in the initial experiment that discovered the interaction , and ( ii ) such falsification is by definition negative evidence for the existence of the interaction and therefore could be subject to other interpretations . Thus , at present it is difficult to evaluate how false interactions affect the growth of yeast biology . In this work , we considered only the knowledge of the presence of an interaction and ignored detailed knowledge such as the strength of the interaction , the conditions under which the interaction occurs , and the biochemical or genetic basis of the interaction . It is difficult to analyze these types of knowledge at present because their structures are unclear . Paradigm shifts have been emphasized as an important mode of knowledge growth [2] . In the history of yeast research , the publication of the yeast genome sequence in 1996 [8] is widely thought to have triggered a paradigm shift from gene-based studies to genomic studies . However , such a shift in research scale and approach did not cause apparent changes in either the speed or pattern of discovery of new PPIs and GIs . Further analysis may reveal subtle signals of the paradigm shift that escaped our gross analysis . After all , our work represents just one step towards quantitative understanding of the tempo and mode of knowledge growth in the framework of network theories . Although the generality of our findings requires further evaluation , the lessons learned from this case study may help develop strategies for efficient knowledge exploration in the future .
Yeast protein-protein interaction data and genetic interaction data were downloaded from BioGRID ( http://www . thebiogrid . org ) . The publication year and author information for each interaction were extracted from NCBI ( http://www . ncbi . nlm . nih . gov ) using the PUBMED ID provided by BioGRID . Because we are interested in discoveries of new interactions , interactions that were reported in previous years were excluded . When a new interaction is reported by two or more publications of the same year , one of these publications was randomly chosen for further analyses . We measured the importance of a gene by the reduction in fitness of the yeast strain ( i . e . , growth rate ) in rich medium ( YPD ) when the gene is deleted . The fitness data were downloaded from http://www-deletion . stanford . edu/YDPM/YDPM_index . html . The expression levels of yeast genes are measured at mid-log phase of growth and obtained from a previous study [32] . Authors with identical names were not differentiated . Although this practice necessarily introduced errors , it should not affect our results , because authors with common names and rare names are not expected to behave differently in research ( e . g . , they should participate in large teams with equal probabilities ) . Random network growth was simulated by randomizing the birth year of each interaction while keeping the number of newly discovered interactions unchanged for each year . Network modules were identified using simulated annealing , which has been shown to perform better than other module-separating algorithms [15] . The parameters used were: iteration factor = 0 . 1 , cooling factor = 0 . 9 , and final temperature = 10−20 . For the PPI network , the giant component contains 99 . 72% of all genes and 99 . 98% of all interactions . The corresponding numbers are 98 . 18% and 99 . 89% , respectively , for the GI network . Relative growths of all modules in each year form a vector . The Euclidean distance between vectors of two consecutive years is then computed . The fluctuation index of a network is defined as the mean of Euclidean distances of all consecutive years . We transformed the network growth information into module growths by assigning one unit for every involved gene of a new interaction to the module that the gene belongs to . To measure the deviation of the actual growth of a module in a given year from the expected homogenous growth , we calculated a transformed chi-squares value , , where Oi is the observed growth of module i in a given year and Ei is the expected ( homogenous ) growth given the total growth of the network in the year and the relative size of module i in year-2007 . , where O is the total number of interactions discovered in a given year and Si is the relative size measured by the sum of node degrees of module i to the entire network in year-2007 . In short , for each year , the deviations from homogenous growth were calculated across modules . | It is of great interest to understand the patterns and mechanisms of scientific knowledge growth , but such studies have been hampered by the lack of ideal cases in which the structure of the knowledge is known , the knowledge is quantifiable , and the process of knowledge discovery is well understood and documented . The biological knowledge about a species is in part described by its protein interaction network and genetic interaction network . Here , we conduct a temporal meta-analysis of three decades of discoveries of protein interactions and genetic interactions in baker's yeast to reveal the tempo and mode of the growth of yeast biology . We show that the growth is exponential over time and that important subjects tend to be studied earlier . However , expansions of different domains of knowledge are highly heterogeneous and episodic such that the temporal turnover of knowledge hubs is much greater than that expected by chance . Familiar subjects are preferentially studied over new subjects , leading to a reduced pace of innovation . While research is increasingly done in teams , the number of discoveries per researcher is greater in smaller teams . These findings reveal collective human behaviors in scientific research and help design better strategies in future knowledge exploration . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"computational",
"biology/literature",
"analysis",
"computational",
"biology/genomics"
] | 2009 | On the Growth of Scientific Knowledge: Yeast Biology as a Case Study |
Analyses of stool from patients with acute watery diarrhea ( AWD ) using sensitive molecular diagnostics have challenged whether fecal microbiological cultures have acceptably high sensitivity for cholera diagnosis . If true , these findings imply that current estimates of the global burden of cholera , which rely largely on culture-confirmation , may be underestimates . We conducted a vaccine probe study to evaluate this possibility , assessing whether an effective killed oral cholera vaccine ( OCV ) tested in a field trial in a cholera-endemic population conferred protection against cholera culture-negative AWD , with the assumption that if cultures are indeed insensitive , OCV protection in such cases should be detectable . We re-analysed the data of a Phase III individually-randomized placebo-controlled efficacy trial of killed OCVs conducted in Matlab , Bangladesh in 1985 . We calculated the protective efficacy ( PE ) of a killed whole cell-only ( WC-only ) OCV against first-episodes of cholera culture-negative AWD during two years of post-dosing follow-up . In secondary analyses , we evaluated PE against cholera culture-negative AWD by age at vaccination , season of onset , and disease severity . In this trial 50 , 770 people received at least 2 complete doses of either WC-only OCV or placebo , and 791 first episodes of AWD were reported during the follow-up period , of which 365 were culture-positive for Vibrio cholerae O1 . Of the 426 culture-negative AWD episodes , 215 occurred in the WC group and 211 occurred in the placebo group ( adjusted PE = -1 . 7%; 95%CI -23 . 0 to 13 . 9% , p = 0 . 859 ) . No measurable PE of OCV was observed against all or severe cholera culture-negative AWD when measured overall or by age and season subgroups . In this OCV probe study we detected no vaccine protection against AWD episodes for which fecal cultures were negative for Vibrio cholera O1 . Results from this setting suggest that fecal cultures from patients with AWD were highly sensitive for cholera episodes that were etiologically attributable to this pathogen . Similar analyses of other OCV randomized controlled trials are recommended to corroborate these findings .
An estimated 2 . 9 million cases and 95 000 deaths occur each year due to cholera , caused primarily by Vibrio cholerae ( V . cholerae ) O1 , in endemic countries [1] . Until now , microbiological cultures of stools have provided an accepted gold standard for diagnosing cholera in patients with diarrhoea . Such cultures , particularly when done with alkaline peptone water overnight enrichment , have been regarded as having very high diagnostic sensitivity , as well as high diagnostic specificity . However , one influential paper has questioned the notion that conventional fecal cultures have high sensitivity in diagnosing cholera [2] . A study conducted in Dhaka , Bangladesh , where cholera is endemic , reported that conventional cultures identified V . cholerae 01 in stools in only 86 ( 66% ) of 131 patients with clinically suspected cholera identified during seasonal cholera outbreak who were positive by at least one of a panel of diagnostic tests consisting of culture , multiplex PCR , and direct florescent antibody tests [2] . The authors postulated that failure of culture methods to isolate V . cholerae may be caused by bacterial inactivation by in vivo vibriolytic action of the phages and/or prevention by host-induced mechanisms . In light of this conclusion , one implication is that the current estimates of cholera disease burden that are based on culture-confirmed cholera may be significant underestimates , and that a reassessment of past and recent cholera studies may be needed to guide public health policy on cholera control measures in countries affected by cholera . We reasoned that if conventional fecal cultures for cholera do indeed have only moderate diagnostic sensitivity , and if culture-negative cholera represented an appreciable fraction of cases of acute , watery diarrhoea ( AWD ) , the clinical syndrome of cholera , inactivated oral cholera vaccines ( OCVs ) , which are effective against culture-confirmed cholera [3–6] , should also exhibit detectable efficacy against cholera culture-negative AWD . In this sense , OCVs could be used as a “probe” to evaluate the hypothesis that conventional diarrhoeal cultures are an insensitive tool for the diagnosis of cholera . Herein , we report a re-analysis of a Phase 3 efficacy trial of inactivated OCVs in Matlab , Bangladesh to evaluate this possibility .
In this trial 25 416 individuals were vaccinated with at least two doses of WC-only OCV , and 25 354 received at least two doses of placebo . Of the 50 770 people vaccinated with either WC-only OCV or placebo , 791 first episodes of AWD from 786 patients during two years of follow-up among which 365 were culture-positive for V . cholerae O1 . Of the remaining 426 first episodes of culture-negative AWD , 215 ( 50 . 5% ) occurred in recipients vaccinated with WC and 211 ( 49 . 5% ) occurred in placebo recipients ( Fig 1 ) . A majority of the cases , 373 ( 87 . 6% ) occurred in individuals ≥5 years . The overall occurrence of cholera culture-negative AWD did not differ statistically between vaccinees and placebo recipients ( adjusted PE = -1 . 7%; 95%CI -23 . 0 to 15 . 9% , p = 0 . 859 ) ( Table 1 ) . In contrast the occurrence of cholera culture-positive AWD differed significantly between the two groups ( adjust PE = 51 . 7; 95% CI 39 . 9 to 61 . 2 , p<0 . 001 ) . When evaluating the incidence of cholera culture-negative AWD by age subgroups and for cholera season , we again failed to detect vaccine protection ( Table 1 ) . Finally , comparison of the incidence of severe cholera culture-negative AWD in vaccinees versus placebo recipients failed to detect protective efficacy of the OCV among all individuals ( adjusted PE = -5 . 9%; 95%CI -34 . 0 to 16 . 3% , p = 0 . 633 ) , in different age subgroups , and in the cholera season ( Table 1 ) . To examine the possibility that specificity of cholera culture is different during the cholera and non-cholera season , a secondary analysis of children ≥5 years , for whom OCV was protective , during the cholera season was conducted . Protection was also not detected under these conditions ( Table 1 ) .
Using OCV as a vaccine probe to detect culture-negative cholera during the first two years of follow-up in a placebo-controlled , randomized trial in Matlab , we failed to detect OCV protection against all episodes of cholera culture-negative AWD , by age groups , for cholera season , or by disease severity . In contrast , analyses showed 51 . 7% PE against culture confirmed cholera in patients with AWD during the same interval of follow-up . Before discussing the interpretation of these findings , it is important to address the limitations of our study . Only patients with diarrhoea severe enough to seek care at a health facility were captured in the surveillance and included in the analysis , thus our findings may not pertain to less severe cases of diarrhoea . Additionally , the study was conducted in a cholera endemic region where people had pre-existing natural immunity to cholera and where cholera culture was performed systematically , therefore the results of this study cannot be generalized for populations lacking such immunity or where cholera culture diagnostics are not routine . However , one would expect natural immunity to reduce the fecal shedding of ingested cholera vibrios , which would tend to increase rather than decrease the diagnostic sensitivity of conventional cultures . Further , this analysis was performed on the data collected over 3 decades ago which calls into question present day generalizability . At the time of the trial , the prevailing circulating strains were both El Tor and classical biotypes [7] . Since then , variants of the El Tor biotype have emerged and become predominant in Bangladesh and many other cholera-endemic areas [8] . Finally , the validity of our conclusions hinges on the assumption that the OCV under study was protective against culture-positive and culture-negative cholera . While there is no reason to doubt this assumption , there is no direct evidence to support it . On the other hand , our study had several strengths . The data were obtained from a prospective , placebo-controlled , individually randomized trial with comprehensive and systematic surveillance for all episodes of diarrhoea in the study population , including fecal cultures at a high-calibre diagnostic laboratory . Importantly , Matlab has a well-functioning demographic surveillance system , and patients in this study were accurately identified when they presented for care at the heath facilities . Note that non-differential misclassification of patient identities would have acted to diminish measured vaccine PE . Also arguing against such misclassification was the concurrent demonstration of vaccine PE against culture-proven cholera . Additionally , our analysis only included registered individuals in the Matlab demographic surveillance who had verifiably ingested vaccine or placebo . Finally , our study was adequately powered to detect vaccine protection in cholera culture-negative AWD . As shown in Fig 1 , the surveillance detected a total of 791 first episodes of AWD , of which 365 were culture-positive for cholera and 426 were culture-negative . If , as reported , the diagnostic sensitivity of conventional fecal cultures for cholera is 66% , we would expect 189 ( 44% ) of the 426 culture-negative AWD cases to be cholera . For an OCV that was 51 . 7% protective against cholera in the same setting and for the same duration of follow-up , a level of OCV protection against culture-negative AWD of 23% would have been detected . However , the upper boundary of the 95% confidence interval for measured PE in the primary analysis ( 16% ) excluded this value . OCV: oral cholera vaccine; WC: whole-cell only vaccine; BS+WC: B-subunit whole-cell vaccine; AWD: acute watery diarrhea It is important to emphasize that our vaccine probe study was designed to evaluate whether there was an appreciable fraction of cholera culture-negative cases of AWD in which patient symptoms could be etiologically attributed to infection by V . cholerae O1 . It is well documented that isolation of cholera vibrios from fecal specimens may not be sufficient per se to incriminate the isolated vibrios as the cause of the patient’s diarrhoea [9] . In the earlier Bangladesh study that reported low diagnostic sensitivity of conventional microbiological cultures , diagnoses of cholera were based on a panel of multiple diagnostic tests , including very sensitive molecular methods [2] . It is possible that some of cholera culture-negative cholera cases designated as cholera by the alternative tests in this study were due to false positive isolations . However , it is also possible that in many of the cases where fecal shedding of V . cholerae O1 was detected , vibrios may have been present but not the aetiology of symptoms . The contention that conventional culture methods do not capture all cholera cases has implications for cholera global burden estimates , which are already thought to be an underestimation due to the incomplete diagnostic testing and reporting of cholera cases in many settings . However , it is important to consider that identification of V . cholerae O1 in the stool does not always confirm the etiologic role of the isolated organisms in causing a patient’s diarrheal symptoms , and that sophisticated diagnostic technologies , in some cases , may overstate the fraction of diarrhoeal disease caused by cholera . While the findings of our study support the use of conventional fecal cultures to diagnose cholera , contemporary studies in endemic as well as non-endemic settings are needed to examine the validity of our findings .
The study was conducted in rural Bangladesh at Matlab , where icddr , b has been maintaining a field research centre since 1963 . Matlab is a low-lying riverine area that lies 55 km southeast of Dhaka , the capital of Bangladesh , and has remained endemic for cholera . Since 1966 a Health and Demographic Surveillance System ( HDSS ) , which consists of regular cross-sectional censuses and longitudinal registration of vital events , has been maintained in the study area [10] . The data were obtained from a Phase III efficacy study , an individually randomized , placebo-controlled trial design , conducted in 1985 in which persons aged 2 to 15 years , and non-pregnant females older than 15 years were assigned to receive three oral doses of one of the following agents: 1 ) cholera toxin B subunit killed whole-cell ( BS-WC ) cholera vaccine; 2 ) a vaccine identical to BS-WC , but lacking BS ( WC ) ; or 3 ) a placebo consisting of killed Escherichia coli K12 cells , as previously described [11] . Vaccination took place in 1985 , and of the 124 035 persons who were age-eligible for vaccination , 63 498 persons received all three doses of an assigned study agent . Surveillance for diarrheal illnesses was undertaken at all three diarrheal treatment centres serving the study population , where patients were assessed clinically and fecal specimens were collected for microbiological diagnosis of V . cholerae O1 with conventional culture techniques , including overnight enrichment in alkaline peptone water . To evaluate whether the use of conventional fecal cultures to define cholera underestimated the true incidence of cholera in the Matlab trial , we assessed whether recipients of at least two complete doses of the WC-only OCV protected against AWD that was culture-negative for cholera . We assumed that the protection by this vaccine against culture-negative cholera was equivalent to the vaccine’s protection against culture-confirmed cholera . We chose not to evaluate protection by BS-WC in this analysis , because this vaccine , in contrast to WC-only vaccine , cross protects against heat labile toxin ( LT ) - producing enterotoxigenic Escherichia coli diarrhoea , a cause of AWD for which conventional cholera cultures can be negative [12] . The post-vaccination follow-up selected for this analysis was two years , an interval in which the WC-only OCV was protective against culture-confirmed cholera . For this analysis , we defined a diarrheal treatment visit as a visit in which the patient reported three or more loose or liquid stools or one-to-two or an indeterminate number of loose stools with at least two objective signs of dehydration on initial physical examination ( feeble or absent pulse , tenting of skin , sunken eyes , or dry mucous membranes ) . Diarrheal visits were concatenated into diarrheal episodes when the date of onset of symptoms for one visit was 7 or fewer days after the date of discharge for the previous visit . The onset of an episode was the onset of first component visit of the episode . AWD episodes were diarrheal episodes for which no stool with visible blood was reported . Cholera culture-negative AWD episodes were those for which no fecal culture detected V . cholerae O1 . Cholera culture-positive AWD episodes required at least one culture during the episode that was positive for V . cholerae O1 . Such episodes were considered be severe if , at the time of any of the component visits for treatment , an absent or feeble pulse was noted and at least one additional objective sign of dehydration was described ( poor skin turgor , sunken eyes , or dry mucous membranes ) . This vaccine probe analysis was designed to measure the difference in the disease incidence of cholera culture-negative AWD between vaccine and placebo recipients . In the primary analysis we compared the overall occurrence of first episodes of cholera culture-negative AWD , with onsets from 14–730 days after receipt of the second dose , in subjects who had received at least two complete doses of killed WC-only vaccine or placebo , as earlier analyses had demonstrated PE to be equivalent for recipients of two and three doses and the vaccine was demonstrably protective against cholera during this follow-up interval [5] . In secondary analyses , we evaluated vaccine protection against cholera culture-positive AWD , as well as cholera culture-negative AWD by age at vaccination , season of onset , and disease severity . In these analyses , which were undertaken to address the possibility that vaccine protection might be unmasked when analysed for the older age group ( ≥5 years ) , during the cholera season , or against severe cholera , age was categorized as under five years versus five years and older; seasonality was classified as cholera season ( April-May and October-November ) versus other; and cholera was classified as severe or non-severe , as defined earlier . We measured vaccine PE against first episodes of cholera culture-positive and cholera culture-negative AWD in Cox proportional hazard regression models , in which time to event was measured in relation to receipt of the second dose , and deaths , out-migrations , and 730 days after the second dose were right-censoring events . In the analysis of cholera season , the events that occurred during the cholera season were counted in the numerator and the events that occurred outside of the cholera season were censored at the time of event . In these models , vaccination was expressed dichotomously as vaccine versus placebo . We controlled for potentially confounding variables , i . e . the variables which were independently associated with time to event at p value < . 10 ( two-tailed ) in a backward selection algorithm . To evaluate heterogeneity of vaccine protection among different subgroups ( age <5 and ≥5 years ) , interaction terms between vaccination and subgroup variables in these models were evaluated . Before including any variable as an independent variable in the model , we first determined whether the proportional hazard assumption was fulfilled for the variable . There was no violation of the assumption for variables included in this model . We estimated the hazard ratio ( HR ) for the outcome by exponentiating the coefficient for the vaccination variable in the model; the 95% confidence interval for the HR was estimated using the standard error of the coefficient . We considered P < . 05 ( two-tailed ) as the margin of statistical significance . The trial was approved by the Ethical Review Committees of the World Health Organization and the International Centre for Diarrhoeal Disease Research , Bangladesh ( now called icddr , b ) . All adult subjects provided oral consent prior to inclusion , and a parent or guardian of any child participant provided informed consent on their behalf . Inclusion in the vaccine registry was considered as documentation of consent . All data was anonymized during analysis . | Conventional microbiological culture has remained a relatively uncontested ‘gold standard’ for the diagnosis of cholera; however , emerging methods , including sensitive molecular tests , challenge the current paradigm . One pivotal article demonstrated that culture failed to detect cholera in one-third of the cholera-positive stool specimens confirmed by other methods . This finding underscored the absence of a reliable reference test , further complicated by newer tests outperforming the gold standard , leaving no suitable comparator . In this study , we used oral cholera vaccine as a probe to investigate the reliability of conventional culture as a diagnostic for cholera by measuring the effectiveness of the vaccine against cholera culture-negative acute watery diarrhea . We did not find any evidence of protection , implying that the culture diagnostics used were reliable . The dynamics of cholera transmission require a rapid response , and ascertaining the best rapid diagnostic test for early detection of outbreaks will maximize the effectiveness of chronically limited resources in high risk regions . As techniques advance , well-designed studies should be implemented to systematically evaluate their merit against established methods , and improved diagnostics , including rapid diagnostics and microbiological culture , should be implemented into cholera control programs to reduce cholera transmission by creating a better trigger for outbreak response . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"microbial",
"cultures",
"pathogens",
"vibrio",
"biological",
"cultures",
"tropical",
"diseases",
"microbiology",
"geographical",
"locations",
"vaccines",
"diarrhea",
"bacterial",
"diseases",
"vibrio",
"cholerae",
"signs",
"and",
"symptoms",
"gastroenterology",
"and",
"hepatology",
"neglected",
"tropical",
"diseases",
"infectious",
"disease",
"control",
"bacteria",
"bacterial",
"pathogens",
"bangladesh",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"cholera",
"cholera",
"vaccines",
"medical",
"microbiology",
"microbial",
"pathogens",
"people",
"and",
"places",
"diagnostic",
"medicine",
"asia",
"biology",
"and",
"life",
"sciences",
"organisms"
] | 2019 | Use of oral cholera vaccine as a vaccine probe to determine the burden of culture-negative cholera |
Weather factors are widely studied for their effects on indicating dengue incidence trends . However , these studies have been limited due to the complex epidemiology of dengue , which involves dynamic interplay of multiple factors such as herd immunity within a population , distinct serotypes of the virus , environmental factors and intervention programs . In this study , we investigate the impact of weather factors on dengue in Singapore , considering the disease epidemiology and profile of virus serotypes . A Poisson regression combined with Distributed Lag Non-linear Model ( DLNM ) was used to evaluate and compare the impact of weekly Absolute Humidity ( AH ) and other weather factors ( mean temperature , minimum temperature , maximum temperature , rainfall , relative humidity and wind speed ) on dengue incidence from 2001 to 2009 . The same analysis was also performed on three sub-periods , defined by predominant circulating serotypes . The performance of DLNM regression models were then evaluated through the Akaike's Information Criterion . From the correlation and DLNM regression modeling analyses of the studied period , AH was found to be a better predictor for modeling dengue incidence than the other unique weather variables . Whilst mean temperature ( MeanT ) also showed significant correlation with dengue incidence , the relationship between AH or MeanT and dengue incidence , however , varied in the three sub-periods . Our results showed that AH had a more stable impact on dengue incidence than temperature when virological factors were taken into consideration . AH appeared to be the most consistent factor in modeling dengue incidence in Singapore . Considering the changes in dominant serotypes , the improvements in vector control programs and the inconsistent weather patterns observed in the sub-periods , the impact of weather on dengue is modulated by these other factors . Future studies on the impact of climate change on dengue need to take all the other contributing factors into consideration in order to make meaningful public policy recommendations .
Dengue fever ( DF ) is the most common vector-borne viral disease in humans and is distributed worldwide , mainly in tropical and subtropical countries . In recent decades , dengue has been expanding globally possibly due to climate change [1] and highly intra and extra-country connectivity through traffic , commerce , and migration [2] . DF is caused by one of four distinct dengue virus serotypes ( DEN 1–4 ) . This viral infection has resulted in an estimated 50 million to 100 million annual cases of DF worldwide , with about 500 , 000 of these cases developing into life-threatening Dengue hemorrhagic fever ( DHF ) /Dengue shock syndrome ( DSS ) [2] , . In Singapore , which is a tropical island city state , DF is endemic , with year-round transmission observed . The integrated vector control program , implemented by the government , that started in the late 1960s resulted in a prolonged period of low dengue incidence [5] . The key strategy for dengue control in Singapore is to tackle the root of the problem , which is to deny Aedes mosquitoes the place to breed , i . e . , source reduction [6] , [7] . With a multi-pronged approach [6] , [7] , Singapore had adopted: 1 ) preventive surveillance and control , in which daily mosquito surveillance operations are conducted with the aid of the Geographical Information System; 2 ) public education and community involvement through working with construction sites , schools and community councils; 3 ) enforcement for carrying out intensive search and destroy operations at outdoor as well as indoor areas under legal laws upon notification of a dengue cluster; and 4 ) research for combating dengue disease including polymerase chain reaction , rapid antigen test kits , sequencing and bioinformatics , etc . In addition to the preventive surveillance approaches , general practitioners and hospitals in Singapore are obliged to report probable dengue cases to the Ministry of Health and all reported dengue cases of DF/DHF are then confirmed by one or more laboratory tests including anti-dengue IgM antibody , enzyme linked immunosorbent assay ( ELISA ) , and polymerase chain reactions ( PCR ) . To our knowledge , there was no change in the notification process during the period studied in this work . In Singapore , more than 80% of notified dengue cases were hospitalized [8] . Although under intensive dengue surveillance , we still experienced dengue hyperendemic in 2005 and in 2013 [9] , with the number of laboratory confirmed cases reaching 14 , 209 cases ( with 27 deaths ) and 22101 cases ( with 7 deaths ) respectively . The re-emergence of hyperendemic may be due to low herd immunity , shift of dominant serotypes , high subclinical dengue infection and weather conditions etc . In an earlier report based on Singapore dengue data [10] , it is estimated that only 1 out of 23 dengue cases are diagnosed and notified , which indicates a substantially high unreported dengue rate , i . e . , a majority of dengue cases is either asymptomatic or subclinical but they are able to transmit dengue viruses to uninfected mosquitoes to trigger further infections . Other than the high subclinical cases possibly causing the dengue transmission to worsen , the tropical weather condition favors the year-round presence of Aedes mosquitoes , which is key in the dengue-human transmission chain . Thus , a better understanding on the association between weather and dengue incidence is important for a more proactive surveillance strategy of dengue control . The impact of weather on dengue incidence has been widely studied [11] , [12] , [13] , [14] , [15] , [16] , [17] , [18] as it is relatively easy to obtain basic meteorological data in dengue affected countries . Earlier studies have found many specific relationships between weather factors and dengue incidence . For example , the seasonality of dengue is well established for Thailand [19] , [20] , [21] and Vietnam [22] , where dengue epidemic coincides with the rainy season . Malaysia also reported a strong seasonal pattern but its correlation to weather appears to be more complicated [23] . The number of dengue cases in Malaysia appears to be positively correlated with two to three month lag to the heavy rain in the first wet season of the year . For specific weather variables in Singapore , mean temperature and relative humidity were found to be the most important weather factors upon comparing models which considered long-term climate variability and linear lag effects of weather variables including temperature , humidity and rainfall [24] . In another study from Brazil [25] , maximum temperature and minimum temperature were found to be the best predictors for the increased number of dengue cases . In Singapore , a model consisting of lag effects of mean temperature and rainfall was built and applied to forecast the number of dengue cases over a 16 week period [15] , [26] , [27] . Mean temperature and relative humidity at a lag of 2 weeks and Niño Southern Oscillation Index at a lag of 5 weeks were found to have significant impact on dengue [24] . However , the effect of absolute humidity on dengue incidence , which reflects the combined impact of temperature and relative humidity , has not been well described . In addition to weather , the impact of the dynamics of circulation of dengue virus serotypes on dengue epidemiology has been well documented [28] . Infection with one serotype confers life-long immunity to that particular serotype [29] , [30] . Some studies have also reported a time-lagged correlation between dengue virus serotype dynamics and disease incidence rates [31] . The variation of dominant serotypes needs to be taken into account in studies of environmental factors on dengue incidence . In this study , we modeled and compared the effect of absolute humidity with the effect of temperatures ( maximum , minimum , mean ) , relative humidity , rainfall and wind speed on dengue in Singapore from 2001 to 2009 . The model used is a distributed lag non-linear model , i . e . , an over-dispersed Poisson model with regressions on autocorrelation , lagged effect of weather factors , population sizes and dengue trends . The model is further refined by comparing the impact of weather variables in sub-periods divided based on the dominant circulating dengue serotypes . The model selection criterion applied in this study is the Quasi Akaike's Information Criterion .
Singapore is a tropical island city state with approximately 710 . 2 km2 land area . The average size of the total population over the years , from 2001 to 2009 , is about 4 . 41 million ( Department of Statistics , 2013 ) . The mean temperature ranges from 25 . 2°C to 30 . 3°C , with the maximum daily temperature and maximum daily rainfall reaching up to 34 . 5°C and 479 . 7 mm respectively . A vector control program in the 1960s to 1980s had successfully prevented dengue outbreaks for two decades since 1973 , with less than 1 , 000 reported cases per year [5] . However , since 1989 , Singapore has observed increased notifications of dengue infection despite a low Aedes house index of less than 1% . The factors contributing to the re-emergence includes an increase in human population and density , increases in cross border and in country travel and low herd immunity , resulting from low transmission in the prior decade [5] . The most recent large outbreaks occurred in 2005 [32] and 2013 raise more concern on dengue spread in Singapore . Weekly notified DF/DHF cases in Singapore from 2001–2009 were retrieved from the Weekly Infectious Diseases Bulletin [9] of the Singapore Ministry of Health . The human population data used was based on the mid-year Singapore total population data obtained from the Singapore Department of Statistics [33] . Whilst all four dengue serotypes have mostly been detected in Singapore , typically there is one predominant circulating serotype , with switches in predominance associated with the outbreaks ( Table 1 ) . The dominant serotype was defined as one that causes more than 50% of cases sampled . The estimated proportion of each viral serotype was obtained from the Singapore Communicable Diseases Surveillance reports [34] of Singapore Ministry of Health . DEN-2 was the dominant circulating serotype in the years 2001–2003 , DEN-1 in 2004–2006 and DEN-2 in 2007 to 2009 . Weather data including Mean temperature ( MeanT , °C ) , Minimum temperature ( MinT , °C ) , Maximum temperature ( MaxT , °C ) , Rainfall ( Rain , mm ) , Relative humidity ( RH , % ) and Wind speed ( WindS , m/s ) were obtained from the National Environment Agency , Singapore . Absolute humidity ( AH , g/m3 ) , which is the mass of water in a unit volume of air , was estimated through dry bulb temperature and relative humidity using the approximated equation , assuming standard atmospheric pressure [35]: ( 1 ) where Tc is the dry bulb temperature ( in our studies , Tc is the daily mean temperature ) , andwhere Td is the dew point temperature . Td is approximated from the equation below , based on dry bulb temperature and relative humidity:where , and . Weekly weather data were calculated by averaging the daily weather values over each week . The relationship between AH , Tc and RH is presented in Figure 1 . Spearman rank correlation tests were then applied to assess the association between weekly dengue cases and weather factors for a range of time lags – from 0 to 20 weeks , over the whole study period ( from 2001 to 2009 ) ( see Figure 2 ) . As the number of dengue incidence is a Poisson count data , it is thus not feasible to check how it is linearly related to weather factors . As such , Spearman rank correlation is usually chosen as it is designed to assess how well two variables are monotonically related even if their relationship is not linear [36] . As autocorrelation was detected in each time series , it would not be appropriate to calculate p-values of the correlation coefficients by traditional methods . Therefore , the p-values were calculated through Adaptive Wavelet-Based Bootstrapping [37] with a sample size of 5000 . This was implemented in R software ( version 3 . 0 . 2; package ‘wmtsa’ ) . In this study , the p-value of the correlation coefficients between every two time series was calculated using this method . Furthermore , the associations between each weather predictor and the risk of dengue were modeled . The number of observed dengue cases , , at week , was assumed to follow an over dispersed Poisson distribution [38] with mean . The effect of weather variable on was described by a Distributed Lag Non-linear Model ( DLNM ) [39] , [40] given as follows: ( 2 ) where is the intercept , and are coefficients of the auto-regression terms , is a function to denote smoothed relationships between and a single weather factor ( i . e . , MinT , MeanT , MaxT , Wdsp , Rainf , RH or AH ) with a maximum lag number of , which enables to include the lag effect of predictors into the model . The nonlinear effect of weather factor was described by a natural cubic spline ( ns ) smoothing function with degrees of freedom ( df ) and knots at equally spaced quantiles , while the lag effect of was described by an ns smoothing function with df of . is the corresponding coefficients vector . is an ns smoothing function with df of 1 per year applied to fit the long-term trend of dengue incidence . Here , the df , = 9 and is the corresponding coefficients vector . is the mid-year population size of Singapore and is the offset term . Besides the DLNM , the single lag effect of each weather factor was also investigated . When considering the effect of weather factor at lag , was replaced by in Eq . 1 with being the lag number , and being the coefficients vector , i . e . , the effect of was modeled by an ns function with df of . In order to reflect the goodness-of-fit , Quasi Akaike's Information Criterion ( QAIC ) was used with a smaller QAIC implying a better fit [40] , [41] . QAIC is given by ( 3 ) where L is the log-likelihood of the fitted model with parameters ( in Eq . 2 , ) and ( i . e . , the estimated overdispersion parameter ) , whereas k is the number of parameters . In ( Eq . 2 ) , was selected from 0 to 20 weeks [15] . The df ( ) of each was selected from 1 to 5 , while the df ( ) of lag was selected from 1 to 3 . Higher df implies higher flexibility , but may introduce over-fitting . The selection criterion was QAIC and model flexibility . For the space of each weather variable , QAIC indicated = 4 or 5 for all weather variables; whilst for the lag dimension , QAIC indicated = 2 or 3 . In this article , we adopted = 4 and = 3 . The analyses were performed in R software ( version 2 . 13 . 2; package ‘dlnm’; R Development Core Team , 2011 ) [42] . We first investigated the maximum lag considering the overall effect of each weather variable on dengue incidence for the whole period . Once the best model was established based on the smallest QAIC , the model was further studied and evaluated for both the entire studied period and the three distinct sub-periods based on the predominant circulating serotypes .
We found that Absolute humidity ( AH ) was positively correlated with Relative humidity ( RH ) and Temperature ( see ( Eq . 1 and Figure 1 ) ) . The correlation coefficient between AH and RH is 0 . 21 , whilst the correlation between AH and mean temperature is 0 . 54 . A higher RH or a higher temperature was associated with a higher AH . However , the correlation between MeanT and RH was negative ( the correlation coefficient is −0 . 71 ) . Therefore , as a composite index of MeanT and RH , the impact of AH on dengue incidence was studied further . The Spearman rank correlation analysis , using time lagged weather data ( 0–20 weeks ) , showed that temperature ( MeanT , MaxT , MinT ) , absolute humidity and rainfall exhibited significant association with dengue incidence . On the other hand , no significant relationship was observed between dengue and wind speed , and relative humidity . The correlation between AH and dengue incidence was the highest ( its correlation coefficient was 0 . 234 with p-value<0 . 05 at a 7-week lag ) among all the studied weather variables ( see Figure 2 ) . The second highest correlation was between MeanT and dengue , with the lag period of 12 weeks and a corresponding correlation coefficient of 0 . 211 with p-value<0 . 05 . The correlation between rainfall and dengue incidence is , although significant , numerically quite small , about less than 0 . 15 . It was also observed that AH was associated with the smallest QAIC values , among all weather predictors in both single and distributed lag models ( see Table 2 ) . The best single lag effect of AH was 1 week , after adjustment for the impact of previous dengue incidence . When considering the cumulative lag effect of AH , a 0–16 weeks lag of AH showed the best fitting performance . Residual analysis is shown in Figure 3 . The smaller the fitted number of dengue cases was , the less the variability of the residual values would be seen ( Figure 3B ) . This supported our statement that overdispersion existed in the distribution of dengue . Autocorrelation function and partial autocorrelation function of residuals ( Figure 3C & Figure 3D ) demonstrated the independence of the residuals , implying that autocorrelation of the dengue cases has been explained by the DLNM-AH model . Summing up each single lag effect from 0 to16 weeks , the 17-week overall effect of AH on relative risk of dengue incidence for the full period is shown in Figure 4A . It can be seen that a higher AH was associated with a higher dengue incidence . It is important to note that that the relative risk here is the ratio of the probability of dengue incidence occurring at a certain value of a weather variable to the probability of the event occurring at a reference value of the same weather variable . The change of reference points may affect the width of confidence interval , but it will not affect the RR curve itself . In some research work , mean was chosen as reference [43] , while the point of overall minimum mortality was chosen as the reference in some other work [40] . Here , the reference value of AH is 22 . 4 g/m3 , which is both mean and median of AH during the studied period . The estimated weekly dengue incidence , using only the AH term ( i . e . , exp ( ) , see Eq . 2 ) is shown in Figure 5A . The correlation coefficient between the estimated dengue and observed dengue cases is 0 . 374 ( p-value<0 . 01 ) , which shows a moderate positive relationship . It can be clearly seen that the peaks of AH and dengue incidence are very well synchronized . As MeanT has been used as an indicator by National Environment Agency ( NEA ) of Singapore for dengue surveillance in recent years [44] , we also modeled MeanT's impact on dengue incidence and compared it with the impact of AH . Based on our model analysis , the longest lag that best reflects the effect of MeanT on dengue is 9 weeks . Residual analysis is shown in Figure 6 . Similar phenomena were detected in the residuals compared with the residuals of the DLNM-AH model . Nevertheless , slightly higher values were detected in autocorrelation function and partial autocorrelation function of residuals ( Figure 6C & Figure 6D ) . The effect of 0–9 weeks lag of MeanT for the full period is shown in Figure 4B . In general , it can be seen that a higher MeanT is associated with a higher risk of dengue incidence but this observed relationship does not hold true when the MeanT is higher than 27 . 8°C . The estimated number of weekly dengue cases using the MeanT term , described in Eq . 1 , is shown in Figure 5B , which showed that the correlation coefficient between the estimated dengue and the observed dengue cases is only 0 . 150 . In addition to studying the pattern for the entire period ( 2001–2009 ) , analyses were also carried out on the three distinct sub-periods , namely , 2001–2003 ( sub-period 1 , DENV2 ) , 2004–2006 ( sub-period 2 , DENV1 ) , and 2007–2009 ( sub-period 3 , DENV2 ) . The aim is to evaluate the coupling effect of weather factors as well as the impact of the dominant serotypes in each period . The overall effects of AH on dengue incidence in each sub-period are presented in Figure 7 ( A1 to A3 ) . In sub-period 1 and sub-period 2 , the impact of AH on dengue incidence was found to be similar to that observed in the whole period , i . e . increasing the AH generally increased the risk of dengue incidence . However , in sub-period 3 , it can be seen that the effect of AH on dengue was not significant . The effect of 0–9 weeks lag of MeanT for each sub-period is shown in Figure 7 ( B1 to B3 ) . It can be seen that the impact of MeanT on dengue incidence in the three sub-periods was not consistent across the three sub-periods or with the pattern observed during the whole period . In sub-period 1 , the impact of MeanT on dengue was not significant when MeanT was less than 27 . 8°C; whilst in sub-period 2 , this effect turned to be not significant when MeanT was higher than 27 . 8°C . Interestingly , the effect of MeanT in sub-period 3 was an inverse U curve , as shown in Figure 7 ( B3 ) .
Cross correlation analysis and DLNM modeling showed that AH was the best predictive weather factor among the weather factors studied . AH presented a more stable effect on indicating dengue incidence than MeanT did over the whole studied period as well as during sub-periods . A higher AH was associated with a higher dengue incidence . As such , AH could potentially be a better weather indicator for predicting dengue and assisting pro-active dengue prevention efforts in the future . The shift of dominant serotypes and pre-emptive measures taken against dengue vectors since 2005 in Singapore may possibly explain the inconsistent weather-dengue patterns observed . As such , further studies are recommended to identify , evaluate and possibly include more diverse virological , immunological , entomological and public health factors into the dengue models . | As dengue virus transmission is through a human-to-mosquito-to-human cycle , the influence of meteorological factors on dengue is likely to be associated with their impact on mosquito populations and behavior . Other than the influence of weather factors , the shift of dominant serotypes and pre-emptive measures taken against dengue vectors may possibly affect the dengue transmission trend . In this study , we investigate the impact of weather factors on dengue in tropical Singapore , taking into consideration the disease epidemiology and profile of virus serotypes . We found that absolute humidity , as a composite index of mean temperature and relative humidity , is a more stable and better predictor for modeling dengue incidence than the other unique weather variables when virological factors are taken into consideration . This research suggests that absolute humidity needs to be considered together with all the other contributing factors in order to make meaningful public policy recommendations for dengue control . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"infectious",
"disease",
"epidemiology",
"epidemiology",
"biology",
"and",
"life",
"sciences",
"population",
"biology"
] | 2014 | Statistical Modeling Reveals the Effect of Absolute Humidity on Dengue in Singapore |
The genome of the fission yeast Schizosaccharomyces pombe encodes 17 kinases that are essential for cell growth . These include the cell-cycle regulator Cdc2 , as well as several kinases that coordinate cell growth , polarity , and morphogenesis during the cell cycle . In this study , we further characterized another of these essential kinases , Prp4 , and showed that the splicing of many introns is dependent on Prp4 kinase activity . For detailed characterization , we chose the genes res1 and ppk8 , each of which contains one intron of typical size and position . Splicing of the res1 intron was dependent on Prp4 kinase activity , whereas splicing of the ppk8 intron was not . Extensive mutational analyses of the 5’ splice site of both genes revealed that proper transient interaction with the 5’ end of snRNA U1 governs the dependence of splicing on Prp4 kinase activity . Proper transient interaction between the branch sequence and snRNA U2 was also important . Therefore , the Prp4 kinase is required for recognition and efficient splicing of introns displaying weak exon1/5’ splice sites and weak branch sequences .
Introns are removed from pre-mRNAs by spliceosomes , which are highly dynamic macromolecular complexes consisting of five small nuclear RNAs ( snRNAs; U1 , U2 , U5 , and U4/U6 ) associated with specific proteins in subcomplexes called snRNPs . In vitro , spliceosomal subcomplexes assemble on pre-mRNAs in a time-dependent manner . The intron to be removed is defined during the formation of spliceosomal complex B , which contains the base-paired U4/U6 snRNP as well as the other three snRNPs . ATP-dependent helicases control the specific rearrangement of the B complex so that catalysis of the transesterification reactions can occur in spliceosomal complex C , formed by the departure of snRNPs U1 and U4 and rearrangement of snRNPs U2 , U6 , and U5 , culminating in the splicing reaction [1–3] . Although pre-mRNA splicing is an important part of regulated gene expression , little is known about the assembly and activation of spliceosomes in vivo . Introns are presumably recognized and removed by the spliceosome during or shortly after transcription , suggesting that parts of the spliceosomal complex must be recruited to transcribed chromatin areas , installed at introns , and then activated for catalysis . Several lines of evidence suggest that the 5’ splice site ( SS ) is defined by direct interactions with snRNA U1 , and that definition of the 3’ SS also involves the interaction of the branch sequence ( bs ) with snRNA U2 [4 , 5] . In the fission yeast Schizosaccharomyces pombe , approximately 45% of genes contain at least one intron , with some genes containing as many as 15 . Compared with the introns in budding yeast Saccharomyces cerevisiae , fission yeast introns are relatively small: the average intron sizes in S . cerevisiae and S . pombe are 256 nt and 83 nt , respectively . In S . cerevisiae , only 5% of genes contain introns , and the 5’ SS and bs at the 3’ region of the intron are highly conserved; in contrast , in S . pombe , the 5’ SS and bs are as variable as the corresponding sequences in mammals [6–8] . Regulated alternative splicing has not been observed in mitotically active fission yeast cells [9]; however , certain genes appear to be regulated by splicing during sexual differentiation . The 80 nt intron in the pre-mRNA encoding the meiotic cyclin Rem1 , for example , is spliced after the initiation of meiosis when the gene is transcribed by the Forkhead family transcription factor Mei4 [10 , 11] . We identified and characterized the Prp4 kinase , which phosphorylates the spliceosomal protein Prp1 ( scPrp6/hsU5-102K ) in vitro and in vivo . Prp4 also phosphorylates Srp2 , one of the two typical SR ( serine/arginine-rich ) protein family members present in fungi [12 , 13] . Prp1 contains 16 direct C-terminal tetratricopeptide repeats ( TPRs ) , preceded by an N-terminal domain of approximately 30 kDa that contains no known motifs . The C-terminal region containing the TPRs is highly conserved across a wide range of organisms , whereas the N-terminal region is not [13] . Prp4 phosphorylates Prp1 at sites in the N-terminal region [14 , 15] . In fission yeast , the structural integrity of the N-terminal domain is essential for pre-mRNA splicing . Short deletions throughout the N-terminus of Prp1 do not prevent spliceosome assembly but lead to the formation of stalled precatalytic spliceosomes that contain unspliced pre-mRNA and the U1 , U2 , U5 , and U4/U6 snRNAs , indicating that the N-terminus of Prp1 is involved in early spliceosome activation [14] . In mammals , the Prp4 ortholog Prp4K phosphorylates equivalent sites in the N-terminal region of Prp1 , and also phosphorylates SR proteins [13 , 16 , 17] . SR proteins contain one or two N-terminal RNA recognition motifs ( RRMs ) and a C-terminal RS domain enriched in arginine–serine dipeptides [18] . In general , SR proteins control a network of RNA processing events , including the regulation of SS selection [19] . In mammals , they can act as splicing enhancer or silencer depending on their position of binding [20] . It is known that they play an important role not only in alternative splicing but also in constitutive splicing [21 , 22] . In case of constitutive splicing they take part in intron recognition at the exon1/5´ SS by interacting with hsU1-70K ( spUsp101 ) as well as at the 3´ SS by interacting with hsU2AF1 ( spUaf2 ) [23–26] . In S . pombe there are two SR proteins known , Srp1 ( hsSRSF2 ) and Srp2 ( hsSRSF4/5/6 ) [27 , 28] . They were found as single proteins but also as a complex depending on their phosphorylation states [29] . Srp2 is known to be phosphorylated by Prp4 kinase while Srp1 is not [13] . However , it was shown that overexpression of Srp1 can suppress the splicing phenotype of the mutant allele prp4-73ts [30] . The temperature-sensitive allele prp4ts caused at the restrictive temperature of 36°C a cell-cycle arrest in the G1 and G2 phases . This phenotype was also observed at the permissive temperature of 25°C when the mutant allele prp4-73ts was expressed from a multicopy plasmid [15] . Cell-cycle arrest in the G1 and G2 phase has been observed primarily in mutants of genes involved in cell-cycle regulation . Therefore , in our attempt to elucidate the mechanism underlying this phenotype , we first examined the splicing of cell-cycle regulatory genes that contain introns , e . g . , cdc2 , res1 , and res2 . We performed these experiments in cells expressing the analogue-sensitive ( as ) mutant prp4-as2 , which encodes a kinase that can be chemically inhibited . This analysis revealed two classes of introns in fission yeast: those whose splicing is dependent on Prp4 kinase activity ( Prp4-dependent ) and those whose splicing is independent of Prp4 kinase activity ( Prp4-independent ) . This finding was confirmed and extended by a genome-wide search for Prp4-dependent and -independent introns , which demonstrated that both intron classes are sometimes present within the same gene . For detailed characterization , we focused on two genes , res1 and ppk8; the single res1 intron is Prp4-dependent , whereas the single ppk8 intron is Prp4-independent . These two intron classes are affected by mutations in the exon1/5’ SS region or the bs of the same intron . The potential interactions between the exon1/5’ SS region and snRNP U1 , and between the bs and snRNP U2 , determine the Prp4 dependency of the intron . Taking into consideration the results of this and previous studies , we propose that phosphorylation of different substrates by Prp4 kinase helps the spliceosome to recognize and splice efficiently introns with weak SSs that differ from the consensus sequence .
For these experiments , we constructed a conditional analogue-sensitive allele , prp4-as2 , which allows reversible inhibition of Prp4 kinase using an ATP analogue . To inhibit Prp4 , 10 μM of 1NM-PP1 was added to growing cultures of cells expressing the prp4-as2 allele . Addition of inhibitor caused growth arrest and a concomitant decrease in the number of septated cells ( Fig 1A and 1B ) . Cells arrested in the G1 and G2 phases of the cell cycle , as observed for the prp4ts strain . Indeed , fluorescence-activated cell sorting ( FACS ) analysis revealed cells with both 1C and 2C DNA content after 60 min inhibition of Prp4 ( Fig 1C ) . This growth arrest was transient , and cells resumed growing after approximately 180 min , as indicated by the disappearance of the 1C peak and the reappearance of septated cells after 240 min ( Fig 1 ) . The transient arrest in the G1 and G2 phases of the cell cycle led us to hypothesize that the genes whose splicing was affected by Prp4 inhibition were directly or indirectly involved in mitotic cell-cycle progression . For example , the gene encoding the cell-cycle regulator Cdc2 contains four introns that interrupt its open reading frame . Oscillating Cdc2 kinase activity levels regulate cell-cycle progression in G1 and G2 [31] . The decision point at which cells either make the transition from G1 phase to DNA synthesis or exit the cell cycle for conjugation is known as START . Cdc2 kinase and components of the Mlu1-binding factor ( MBF ) are involved in control of START by regulating the expression of genes required for DNA replication . The multimeric MBF complex consists of Cdc10 , Res1 , and Res2 [32 , 33] . The cdc10 gene is intronless , whereas the res1 and res2 genes each contain a single intron ( of 127 nt and 164 nt , respectively ) located in the 5’ region of the open reading frame [34 , 35] . Thus , we focused on the splicing of intron-containing genes that regulate the transition from G1 phase to DNA synthesis . Semiquantitative reverse transcription polymerase chain reaction ( RT-PCR ) analyses were performed to measure the mRNA and pre-mRNA levels of res1 , res2 , cdc2 , and two control genes , rpl29 and tbp1 ( Fig 1D ) . rpl29 contains one intron and encodes large ribosomal subunit protein 29; tbp1 contains three introns and encodes the TATA-binding protein [36] . The highly efficient splicing of res2 and rpl29 were barely affected by inhibition of Prp4 kinase . By contrast , for res1 and tbp1 , unspliced pre-mRNAs accumulated and almost no mRNA was detected , as little as 10 min after the addition of inhibitor . This strong inhibition of splicing was transient , and mature spliced mRNA transcripts of both genes were observed again after 60 min . After 180 min , spliced mRNA levels were similar to those observed in the absence of inhibition ( Fig 1D ) . Splicing of all four introns of the cdc2 transcript was only slightly affected by inhibition of Prp4 kinase , and mature cdc2 mRNA was detected at all time points . Remarkably , the splicing pattern of res1 and tbp1 remained basically the same throughout the time course ( Fig 1D ) . Collectively , these results indicated that the introns of the five genes we investigated could be categorized into two classes: Prp4-dependent and Prp4-independent . Subsequent experiments showed that the res1 intron was primarily responsible for the cell-cycle arrest in G1 following Prp4 inhibition: replacing the wild-type res1 gene with an intronless copy ( res1Δintron ) led to a similar growth delay , but the cells now primarily arrested in G2 phase ( S1 Fig ) . To further examine these two classes of introns , we performed a genome-wide search for additional Prp4-dependent and–independent introns . RNA prepared from the prp4-as2 strain grown in the presence ( 30 min and 60 min exposure , +Inh ) or absence ( -Inh ) of the 1NM-PP1 inhibitor were subjected to RNA sequencing ( RNA-seq ) , and the resultant sequence reads were aligned to the spliced or unspliced fission yeast genome reference sequence . To examine global changes in splicing efficiency , the Relative Splicing Efficiency Index ( RSEI ) of annotated fission yeast introns was calculated for the untreated and treated datasets . Fission yeast introns were divided into two classes . The first class contained the 72% of all introns with a positive RSEI in the presence and absence of 1NM-PP1 , indicating that splicing of these introns was Prp4-independent . This class included res2 , rpl29 , and cdc2 ( Fig 2A ) . The second class contained the 28% of introns for which RSEI was positive in untreated cells but negative in cells treated with 1NM-PP1 , indicating that splicing of these introns was Prp4-dependent . This class included res1 and tbp1 ( Fig 2A ) . Notably , we were unable to identify any gross sequence features that differentiated these two classes of introns . For example , there were no significant differences in intron size or obvious additional sequence motifs that were specific to either class ( Fig 2B ) . Comparison of the SS sequences of Prp4-independent and -dependent introns revealed only slight differences between the two classes especially at position -1 in the exon 1 and positions +4 to +6 in the 5´ SS . The Prp4-dependent introns differed more frequently at these positions from the consensus sequence compared to Prp4-independent ones ( Fig 2C ) . Moreover , in genes containing several introns , not all introns necessarily behaved in the same way upon inhibition of Prp4 . This different behaviour of introns within one gene was also observed for temperature-sensitive alleles of other splicing factors [37 , 38] . As shown in Fig 2D , the mrp17 gene , encoding mitochondrial ribosomal subunit Mrp17 , has one Prp4-dependent and one Prp4-independent intron; by contrast , rpb5 , encoding a DNA-directed RNA polymerase subunit , contains two Prp4-independent introns , and tbp1 contains three Prp4-dependent introns . We wished to characterize the changes in splicing efficiency when the exon1/5’ SS region and bs of these two intron types were mutated . For these experiments , we selected two genes , each containing one intron of similar size and structure , but with RSEI of opposite sign following inhibition of Prp4 kinase . The Prp4-dependent gene was res1 , which contains a 127 nt intron , has an RSEI of -1 . 36 , and is essential for mitotic growth ( Fig 3A ) . The Prp4-independent gene was ppk8 , which contains a 117 nt intron and has an RSEI of +1 . 89 ( Fig 4A ) . This gene is non-essential for mitotic growth; it encodes a putative serine/threonine kinase potentially involved in signal transduction [39] . Because we wanted to compare the pre-mRNA splicing of these functionally very different genes following mutation , we constructed two reporter genes , called res1’ and ppk8’ , respectively . Both reporter genes are driven by the nmt1-8 promoter and contain a frameshift mutation early in exon 1 , a HindIII restriction site upstream of the 5’ SS , and the nmt1 termination region for 3’ end processing . These manipulated genes were introduced into the leu1 locus by homologous recombination ( Figs 3A and 4A ) , and semiquantitative RT-PCR analyses were performed using the appropriate primers . As shown in Fig 3B–3C , the res1’ intron at the leu1 locus was spliced in a Prp4-dependent manner , like endogenous res1+ , whereas the intron of ppk8’ was Prp4-independent like endogenous ppk8+ ( Fig 4B–4C ) . This is consistent with the notion that the Prp4 dependency of a gene is not governed by its genomic context: neither chromosomal location nor the identity of the promoter and 3’ termination region determined whether an intron was spliced in a Prp4-dependent manner . Therefore , the differences in Prp4 dependency must be due to subtle differences in or around the intronic sequences . Base-pairing interactions between pre-mRNA and snRNA U1 play a role in establishing and determining the 5’ SS [40–42] . This recognition process is more complicated in mammals than in yeast , because alternative pre-mRNA splicing requires selection of one out of several possible 5’ SSs [43] . In fungi , however , and particularly in fission yeast , there is little or no alternative splicing [9] . In fission yeast , it has been suggested that nine nucleotides from the 5’ end of the U1 snRNA interact with the pre-mRNA , base-pairing with six nucleotides of the 5’ SS and three nucleotides of exon 1 , to determine the 5’ SS [44] . In addition , the 5’ end of snRNA U1 in fission yeast becomes pseudouridinylated ( Ψ ) . This mechanism is also conserved in eucaryotes . However , in mammalian U1 snRNA two adjacent nucleotides in this region are pseudouridinylated , whereas in fission yeast , only nucleotide number 3 from the 5’ end of snRNA U1 is pseudouridinylated . This nucleotide in the U1 snRNA interacts with the 5’ SS nucleotide +4 of the pre-mRNA ( Figs 3B and 4B ) . In general , pseudouridinylated nucleotides base-pair with A , C , G and U in an A-form RNA duplex , but the highest thermal stability can be found between A-Ψ and G-Ψ [45–48] . To analyse if an increased base-pairing potential leads to a Prp4-independent intron and vice versa , several mutations were introduced into the exon1/5´ SS of the reporter genes res1´ and ppk8´ ( Figs 3 and 4 ) . We mutated positions -3 , +3 and +4 ( Fig 3D , res1´-1 ) or only positions +3 and +4 ( Fig 3D , res1´-2 ) of the res1’ intron . These changes , which increased the potential interactions between U1 snRNA and the pre-mRNA by at least four hydrogen bonds , converted the res1’ intron into a Prp4-independent intron . A time-course RT-PCR experiment revealed that these mutations allowed efficient splicing in the presence ( +Inh ) or absence ( -Inh ) of inhibitor ( Fig 3D ) . As a control , we also mutagenized position +1 or +2 of the res1’ intron to a C or A , respectively; the resultant mutants were not recognized efficiently as introns independent of Prp4 activity ( Fig 3E and S2 Fig , res1´-14 ) . At the time that introns were discovered , it was shown that the GU at the 5’ end is necessary for recognition of an intron [49] . Therefore , this control experiment confirms our interpretation that the transient interaction of the exon1/5’ SS region with snRNA U1 , including nine nucleotides from the 5’ end of U1 , allows efficient splicing in the absence of Prp4 kinase activity ( compare Fig 3D , res1’-1 and res1’-2 with Fig 3E ) . To test whether only one additional interaction at position +3 or +4 is already sufficient for Prp4 independency , both mutants were constructed and resulted in Prp4-independent introns . While an interaction at position +3 lead to an efficiently spliced intron at all time points , an interaction at position +4 caused a slightly decreased splicing efficiency after inhibition ( Fig 3F ) . For all further experiments the Prp4-independent res1´-2 exon1/5´ SS was used to analyse which additional mutations lead to Prp4 dependency again . When these continuous interactions were interrupted by mutating position +5 in the intron from a G to an A , the intron became Prp4-dependent once again ( Fig 3G ) . All mutations at position +5 that did not allow Watson–Crick hydrogen bonding with a C at position 2 of snRNA U1 caused the res1’ intron to be Prp4-dependent ( Fig 3G and S2 Fig , res1’-15 and res1’-16 ) ; therefore , these experiments also prove that proper interaction between U1 and the exon1/5’ SS region is established by formation of hydrogen bonds between the two opposing bases . By contrast , the Prp4 independence of this intron was unaffected by all mutations at position +6 ( the last 5’ SS nucleotide ) that did not allow hydrogen bonding with the A at position 1 of the U1 snRNA ( Fig 3H , res1´-7 and S2 Fig , res1’-17 and res1’-18 ) . But if there is no interaction at positions +3 , +4 and +6 the former Prp4-dependent intron ( Fig 3C , res1´ ) is not recognized anymore even in presence of Prp4 kinase activity ( Fig 3H , res1´-8 ) . Thus , increasing the base-pairing potential between U1 snRNA and pre-mRNA within the 5´ SS results in Prp4-independently spliced introns and vice versa . Similar , rules seemed to apply for mutations at positions -1 , -2 and -3 of exon 1 of the res1´ gene ( Fig 3I ) . If only position -1 or position -2 can form a Watson–Crick hydrogen bond , therefore creating a weaker interaction in the exon 1 compared to res1´-2 , splicing efficiency decreased after inhibition of Prp4 , but the intron was still spliced independently ( Fig 3I , res1’-10 and res1’-11 ) . However , completely absent or very weak hydrogen bonding with the nucleotide at position -3 of exon 1 caused the intron to be Prp4-dependent ( Fig 3I , res1’-12 and res1’-13 ) . Like ppk8 , the intron of the ppk8’ gene integrated into the leu1 locus was also Prp4-independent ( Fig 4A–4C ) . First , the 5’ SS nucleotide +1 G was mutated to a C , the intron was no longer recognized regardless of Prp4 kinase activity , as demonstrated by the exclusive presence of pre-mRNA and absence of mature mRNA ( Fig 4D ) . Then position +3 was mutated from an A to a U , preventing Watson–Crick hydrogen bonding with the snRNA U1 , and position +4 from a U to an A , creating an A-Ψ interaction ( Fig 4E , ppk8’-2 ) . This shows that changing the position of two hydrogen bonds of the 5´ SS converted a Prp4-independent intron into a Prp4-dependent one ( compare Fig 4C , ppk8´ with 4E , ppk8´-2 ) . When the interactions at both positions , +3 and +4 , were absent , the resultant intron was no longer recognized ( Fig 4E , ppk8’-3 ) . However , changing the interactions within the exon 1 creating a ppk8’ gene containing the exon1/5’ SS sequence of the res1+ gene was Prp4-dependent ( Fig 4F , ppk8’-4 ) . Next , we introduced mutations at the end of exon 1 to determine the effect of the pairing of these nucleotides on Prp4 dependency . If the nucleotides at positions -1 , -2 , and -3 of exon 1 in ppk8’ were unable to form hydrogen bonds with the appropriate positions in snRNA U1 , the intron was spliced in a Prp4-dependent manner ( Fig 4F , ppk8’-5 ) . The same rule applied if hydrogen bonding only occurred at position -3 ( Fig 4F , ppk8’-6 ) . However , if a potential hydrogen bond could be formed with the nucleotide at position -1 or -2 , stabilizing the interactions between exon 1 and snRNA U1 , the intron was spliced efficiently in a Prp4-independent manner ( Fig 4F , ppk8’-7 and ppk8´-8 ) . Taken together , these results clearly demonstrate that the number and position of the potential hydrogen bonds between the nucleotides of the exon1/5´ SS region of an intron and the 5’ end of the U1 snRNA is one reason whether an intron is spliced in a Prp4-independent or -dependent manner . As discussed above , we did not find any obvious differences in the bs consensus between +RSEI and–RSEI introns ( Fig 2C ) . The consensus branch sequence of S . pombe is 1 . C/U 2 . U 3 . A/G/U/C 4 . A 5 . C/U; the most frequent bs sequences are CUAAC ( 42% ) and CUAAU ( 23% ) [39 , 50] . The branch point A is the fourth nucleotide in this sequence [51–53] . The 5 nt degenerate bs of fission yeast is similar to that of mammals [6 , 8] . Recognition of the bs via base-pairing of snRNA U2 is also conserved , and the nucleotide at position 3 in the bs of S . pombe is opposite a pseudouridine in snRNA U2 [45 , 47 , 54] . The bs of the res1 intron has the most common sequence ( CUAAC ) , and the pseudouridine in snRNA U2 is at position 39 from the 5’ end ( Fig 5A ) . To study the influence of mutations in the bs on Prp4 dependency , we used res1´ transcripts with exon1/5´ SS regions AAG/GUAAGU ( Prp4-independent ) and AAG/GUUUGU ( Prp4-dependent ) , respectively . We combined mutations in the 5´ SS with mutations in the bs . As expected , mutation of the branch point in position 4 ( from A to U ) combined with the Prp4-dependent 5’ SS prevented recognition of the intron , regardless of the presence or absence of inhibitor ( Fig 5B , res1’-A ) . The reporter gene carrying the Prp4-independent 5’ SS sequence was spliced extremely inefficient , and it was further inhibited by inactivation of Prp4 ( +Inh ) ; consequently , only a small amount of mRNA was detected at the end of the time course ( Fig 5B , res1’-2A ) . The third nucleotide in the bs , which is degenerate for A , G and U , is supposed to interact with the pseudouridine at position 39 in snRNA U2 . When we mutated this nucleotide from A to U , we converted the Prp4-independent intron into a Prp4-dependent one ( compare Fig 5C , res1´-2B with Fig 3D , res1’-2 ) . In contrast , the res1’-B gene , which already contains the Prp4-dependent exon1/5’ SS region , was inefficiently spliced even in the absence of Prp4 inhibitor . After addition of kinase inhibitor pre-mRNA accumulated completely ( Fig 5C , res1’-B ) . These findings are in accordance with the notion that mutations in the third position of the bs lead to a Prp4-dependent intron when there is no Ψ-A interaction . In addition , if the intron was already Prp4 kinase-dependent because of a weak interaction between the exon1/5’ SS region and snRNA U1 , the mutation in the bs intensifies the effect ( Fig 5C , res1´-B ) . The second nucleotide ( U ) in the bs is 100% conserved in all S . pombe introns . When we changed this U to a G , the intron was no longer recognized , neither without nor with Prp4 kinase inhibitor ( Fig 5D ) . Furthermore , mutations in the first or the last nucleotide of the bs combined with the Prp4-dependent exon1/5´ SS lead to an extremely inefficient splicing event which was further intensified after inhibition of the kinase ( Fig 5E , res1’-D and Fig 5F , res1’-E ) . However , when combined with the Prp4-independent exon1/5’ SS these mutations were spliced independently , but after inhibition of the kinase the splicing efficiency decrease slightly ( Fig 5E , res1’-2D and Fig 5F , res1’-2E ) . To summarize the influence of mutations within the bs it is obvious that in case of a weak exon1/5´ SS the negative effect on intron recognition is intensified which leads mostly to intron retention . In contrast mutations in the bs combined with a strong exon1/5´ SS show different results depending on the position of the mutations . Except position 2 all others show an improvement of intron recognition as long as Prp4 kinase is active .
The 5’ SS and the bs of the introns in S . cerevisiae are highly conserved ( GUAUGU and UACUAAC , respectively , in almost all introns ) [53 , 55 , 56] , whereas the corresponding sequences in S . pombe are much more degenerate; GUAAGU and GUAUGU are the most frequent 5’ SSs ( 29% and 21% of all introns , respectively ) , and CUAAC is the most frequent bs ( 42% ) [39] . In this context we have shown that Prp4 kinase is one of the major components to facilitate proper recognition of introns with weak SSs . This kinase is involved in the process that helps to sense and influence proper base-pairing between the exon 1/5’ SS region and snRNA U1 and between the bs and snRNA U2; in this background , proper base-pairing refers to an interaction that accurately determines the 5’ and 3’ SS , leading to an efficient splicing event . In this study we have shown by mutating the exon1/5’ SS and the bs of reporter genes that a Prp4-dependent intron can be changed into an–independent one and vice versa ( Fig 3C and 3D and Fig 4C and 4E ) . The experimental set up also demonstrates that the information for Prp4 dependency resides in the region of the SSs of an intron and has been confirmed by inserting introns with weak and strong SSs into an intronless gene ( S3 Fig ) . Regarding Prp4 dependency the positions for proper base-pairing between the exon1/5’ SS region and snRNA U1 play different roles . For example , positions +1 and +2 are invariable as it was known before [49] . Mutations of these positions lead to accumulation of unspliced pre-mRNA in the presence and absence of Prp4 kinase activity ( Figs 3E , 4D and S2 , res1´-14 ) . This is different , if one considers position +3 and +4 . Position +4 in the 5’ SS is the most degenerate , and can be occupied by any of the four nucleotides ( Fig 2C ) . This position has been suggested to appear opposite a pseudouridinylated nucleotide of snRNA U1 ( Fig 3B ) . Pseudouridine can base-pair with all nucleotides , although the thermodynamic parameters of these wobble base-pair interactions are different from those of Watson–Crick interactions [46 , 48 , 54] . If there is only one Watson-Crick interaction present at position -1 in exon 1 and the base-pairing at position +4 in the 5´ SS is interrupted , the intron is spliced Prp4-independently ( Fig 4C ) . In contrast , an interruption at position +3 leads to a Prp4-dependent intron ( Fig 4E , ppk8´-2 ) . However , if there are two Watson-Crick interactions in exon 1 , at positions -1 and -2 , it does not matter which position , +3 or +4 , is mutated ( Fig 3F ) . The intron is spliced Prp4-independently . These results show that a stable interaction between the exon1/5´ SS and snRNA U1 within this region is needed for Prp4 independence . It also indicates the influence of the interaction with the pseudouridine which changes the helical structure at this position leading to a different binding affinity at position +4 [57 , 58] . Furthermore , interruption of the Watson-Crick interaction at position +5 leads to Prp4 dependency ( Fig 3G and S2 Fig , res1´-15 and res1´-16 ) . However , interruption of base-pairing at position +6 only becomes relevant , if there are no Watson-Crick interactions at positions +3 and +4 of the 5’ SS . In this case the intron is not recognized and therefore retained even in presence of Prp4 kinase activity . This shows that only three and intermittent base-pair interactions in the 5´ SS with snRNA U1 are insufficient for intron recognition ( Fig 3H ) . The snRNA U1 not only interacts with the 5´ SS of the intron , but also with the last three nucleotides of the exon 1 [44] . In general , the 5’ SS consensus sequence differs from the exon 1 sequences in that the 5’ intron sequences are much more highly conserved , whereas the three nucleotides of the exon 1 sequences are much more variable ( Fig 2C ) . For example , for three introns in the same gene , it would be uncommon for the three nucleotides upstream of the 5’ SSs to be identical . Therefore , this region was also examined in this study regarding Prp4 dependency . As we have shown , if there is no interaction within exon 1 these introns are spliced Prp4-dependently ( Fig 3I , res1’-13 and Fig 4F , ppk8’-5 ) . The same results were obtained when an interaction only at position -3 was present ( Fig 3I , res1’-12 and Fig 4F , ppk8’-6 ) . Interestingly , formation of hydrogen bonds at positions -1 or -2 could stabilise the interaction between snRNA U1 and exon1/5´ SS , leading to Prp4 independence ( Fig 3I , res1’-10 and res1’-11 and Fig 4F , ppk8’-7 and ppk8’-8 ) . This rule could also be confirmed by the two introns of the wildtype gene mrp17 ( Fig 2D ) . In this case , intron I has only one possible Watson-Crick base-pairing at position -3 ( CCA/GUAAGU ) and is Prp4-dependent ( compare with Fig 3I , res1´-13 ) . On the contrary , intron II displays one Watson-Crick interaction at position -2 and one wobble base-pairing at position -3 ( UAA/GUAUGU ) which leads to Prp4 independence ( compare with Fig 4F , ppk8´-7 ) . Probably , stabilising this interaction within the exon 1 helps to determine the proper site where the first transesterification reaction occurs . Additionally , mutations in the bs were combined with a strong or weak exon1/5´ SS which had different effects on intron recognition and splicing efficiency ( Fig 5 ) . When combined with a weak exon1/5´ SS , mutations within the bs lead to intron retention in nearly all cases even without inhibition of Prp4 kinase . Therefore , the accumulation time course of pre-mRNA after kinase inhibition reflects an additive effect ( Fig 5B–5F , res1´-A-E ) . In combination with a strong exon1/5´ SS the mutations in the bs showed different effects on splicing . The change of the branch point is almost invariable and therefore leads to intron retention even if no kinase inhibitor was added ( Fig 5B ) . The mutation at position 2 also resulted in complete intron retention even if no kinase inhibitor was added ( Fig 5D , res1´-2C ) . Most interestingly , this position is 100% conserved in all S . pombe introns ( Fig 2C ) . The third position in the bs is the most degenerate , and can be occupied by any nucleotide ( Fig 2C ) ; this position interacts with a pseudouridinylated nucleotide at the end of snRNA U2 which is responsible for bulging out the branch point [54 , 59] ( Fig 5A ) . The mutation of this position leads to a clear Prp4-dependent intron ( Fig 5C , res1´-2B ) . On the contrary , the interruption of possible base-pairing at positions 1 and 5 seems to play a minor role since splicing is still independent on Prp4 kinase activity , although after inhibition of the kinase splicing efficiency decreased slightly in both cases ( Fig 5E , res1’-2D and 5F , res1’-2E ) . Although the underlying rules seem to be very complex , it is obvious that Prp4-dependent introns are distinguished from Prp4-independent introns by their reduced potential for hydrogen bonding between the exon1/5’ SS region and snRNA U1 or between the bs and snRNA U2 . A complementary interaction between the exon1/5´ SS region and snRNA U1 serves as a default state , marking the structure for the first transesterification reaction in the pre-spliceosome . A similar structural marking for proper hydrogen bonding is also associated with the bs-U2 interaction thereby determining the nucleotides where the second transesterification reaction will occur . So far , we can only speculate here about the consequences of the phosphorylation of Prp1 and Srp2 by Prp4 kinase and propose that the phosphorylation by Prp4 might play a role in stabilising the interaction between the SSs and the snRNAs allowing time in concert with the other proteins to display the proper intron borders for the transesterification reactions . Prp1 is a spliceosomal protein operating at the level of precatalytic spliceosomes; therefore , it is reasonable to speculate that phosphorylation of Prp1 could be involved in adjusting a precatalytic spliceosome on introns displaying weak SSs until proper hydrogen bonds between the pre-mRNA and snRNA U1 and U2 are formed . Phosphorylation of Srp2 by Prp4 might play a similar role , helping precatalytic spliceosomes with the recognition of introns and thereby stabilising their interaction with weak exon1/5’ SSs and weak branch sequences . Indeed , it is known for mammalia that the SR protein hsSRSF1 binds to the pre-mRNA and subsequent phosphorylation affects its interaction with a protein of the U1 particle [23] . In S . pombe it has been shown that Srp2 interacts with spUaf2 which binds to the 3´ SS [60] . It seems likely that it also takes part in exon1/5´ SS recognition . However , there are still open several questions regarding this mechanism . Particularly , we hope to identify further components involved and thereby advance our knowledge about the function of Prp4 kinase in the splicing of introns displaying weak SSs .
The standard genetic and molecular techniques used in this study were described previously [61 , 62] . All strains used in this study are listed in S1 Table . The analogue-sensitive prp4-as2 kinase allele was generated by introducing a point mutation into the kinase domain at position 238; this mutation changes the gatekeeper phenylalanine residue to alanine [63 , 64] . The prp4-as2 allele was fused to the kanMxR gene using the pRS426 vector in yeast recombinational cloning [65 , 66] . The resultant construct was used to produce a PCR fragment , containing the prp4-as2 kinase allele and the resistance marker , which was transformed into the wild-type strain L972 . Growing colonies were selected on plates containing geneticin . Proper replacement of the prp4 locus on chromosome III was confirmed in geneticin-resistant transformants by PCR using the appropriate primers [67 , 68] . To construct reporter genes , res1´ and ppk8´ were fused to the thiamine-repressible nmt1-8 promoter by cloning the open reading frames into vector pML81HA [69 , 70] . Both open reading frames contain a frameshift between the HA-tag and the ATG , creating a stop codon at the beginning of the gene; this is intended to exclude the influence of additional Res1 molecules on cell-cycle regulation . To distinguish between the original res1 and ppk8 transcripts and the res1´ and ppk8´ transcripts at the leu1 locus by RT-PCR , a HindIII restriction site was introduced into exon 1 of both genes ( Figs 3A and 4A ) . The forward primers in each pair ( res1_Mut_F , ppk8_Mut_F ) detect the HindIII site . To determine the pre-mRNA splicing patterns of intron-containing genes , RNA was extracted from whole-cell extracts . For RT-PCR , RNA was treated with RQ1 RNase-free DNase ( Promega ) to eliminate possible DNA contaminants . Five micrograms of RNA was treated with 2 . 5 U RQ1 DNase in reaction buffer containing 40 mM Tris–HCl ( pH 8 . 0 ) , 10 mM MgSO4 , and 10 mM CaCl2 . After incubation for 10 min at 37°C , RQ1 DNase was inactivated by the addition of 2 mM EGTA ( pH 8 . 0 ) and heating to 80°C for 10 min . RNA was reverse transcribed and cDNA was amplified using Tth reverse transcriptase ( Roboklon , EURX ) . RNA was incubated for 3 min at the calculated annealing temperature in the presence of 1× Tth RT Buffer , 0 . 25 mM dNTP mix , 20 pmol reverse primer , 2 mM MnCl2 , and 0 . 25 U/ml Tth RT , followed by incubation at 70°C for 25 min . PCR mix containing 1× PCR-Buffer Pol A , 80 pmol reverse primer , 100 pmol forward primer , and 2 mM MgCl2 was then added . cDNA was amplified with 28−45 cycles of 94°C for 30 s , 53−60°C for 30 s , and at 72°C for 60 s . The primer sequences are provided in S2 Table . PCR products were resolved on 2% agarose gels . Directional mRNA sequencing libraries were prepared by combining the Illumina TruSeq mRNA-Seq and Illumina TruSeq small RNA protocols . Briefly , mRNA selection was performed using 4 μg of total RNA and oligo-dT beads , as described in the low throughput protocol for Illumina TruSeq RNA sample preparation . The mRNA was subjected to fragmentation at 94°C , treated with Antarctic Phosphatase ( NEB ) and T4 polynucleotide kinase ( NEB ) , and then purified using RNeasy MinElute spin columns ( Qiagen ) . TruSeq indexed RNA adapters were ligated to the RNA and further processing , including 11 cycles of PCR for library amplification , was performed as described in the Illumina v1 . 5 small RNA protocol . Finally , fragments corresponding to an insert size of 250–500 nt were selected on a 6% Novex TBE gel ( Invitrogen ) . After elution from the gel slice , library quality was confirmed using a DNA 1000 Bioanalyzer chip on an Agilent 2100 Bioanalyzer and . Sensitive quantitation was performed using a KAPA Library Quantification Kit ( Kapa Biosystems ) . Five indexed libraries were pooled and run in each HiSeq lane using Illumina HiSeq v3 sequencing chemistry . Base calling was performed using the Illumina pipeline software version 1 . 8 . 1 ( within HCS 1 . 4 . 8 ) . Adapters used during library preparation were removed from reads using the TagDust tool [71]; approximately 1% of the initial reads were removed in this way . Reads were mapped to the S . pombe genome using the TopHat software ( v2 . 0 . 5 ) [72] . The TopHat alignment was performed using the annotation defined in the Ensembl database ( version ASM294 v1 . 15 ) , taking into account the orientation of the reads . The percentage of mapped reads was approximately 95% for all samples . Finally , the unequivocally mapped reads ( 85–89% of the initial reads ) were selected for further analysis . The average coverages for the exonic and intronic spaces were determined considering only genomic elements longer than 30 bp . This information was then plotted as three different graphs showing the coverage of 5’ exons , introns , and 3’ exons . Exons that were defined simultaneously as the 5’ exon for one intron and the 3’ exon for another intron were considered twice . Finally , the coverage was calculated by normalization to the total number of mapped reads using BEDTools [73] . Plots were generated with custom R scripts . The RNA-seq data have been deposited in NCBI's Gene Expression Omnibus and are accessible through GEO Series accession number GSE75517 ( https://www . ncbi . nlm . nih . gov/geo/query/acc . cgi ? acc=GSE75517 ) For each intron , the number of split and unsplit reads that mapped to the splice junction ( defined as the three bases around the 5’ exon junction ) were counted , and the base-2 logarithm of the ratio between the two values was calculated . This value is called the Relative Splicing Efficiency Index ( RSEI ) . If the RSEI was positive , then more reads indicated spliced mRNA than unspliced pre-mRNA . On the other hand , a negative RSEI indicates more unspliced than spliced RNA . Only intron sequences with more than ten reads for each sample were used for further analysis . This approach identified 2557 Prp4-independent introns ( i . e . , those with a positive RSEI irrespective of Prp4 inhibition ) and 1008 Prp4-dependent introns ( i . e . , those with a negative RSEI in the presence of kinase inhibitor ) . | Prp4 is an essential protein kinase that is involved in the splicing of some introns . Using a conditional mutant of Prp4 , we showed that a subset of genes , including several cell cycle–regulatory genes , are dependent on Prp4 for splicing . Furthermore , we could convert genes between Prp4-dependent and -independent states by introducing single-nucleotide mutations in the exon1/5’ splice sites and branch sequence of introns . This work shows that Prp4 activity is required for splicing surveillance in a subset of mRNAs . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2016 | Prp4 Kinase Grants the License to Splice: Control of Weak Splice Sites during Spliceosome Activation |
Low-frequency sound localization depends on the neural computation of interaural time differences ( ITD ) and relies on neurons in the auditory brain stem that integrate synaptic inputs delivered by the ipsi- and contralateral auditory pathways that start at the two ears . The first auditory neurons that respond selectively to ITD are found in the medial superior olivary nucleus ( MSO ) . We identified a new mechanism for ITD coding using a brain slice preparation that preserves the binaural inputs to the MSO . There was an internal latency difference for the two excitatory pathways that would , if left uncompensated , position the ITD response function too far outside the physiological range to be useful for estimating ITD . We demonstrate , and support using a biophysically based computational model , that a bilateral asymmetry in excitatory post-synaptic potential ( EPSP ) slopes provides a robust compensatory delay mechanism due to differential activation of low threshold potassium conductance on these inputs and permits MSO neurons to encode physiological ITDs . We suggest , more generally , that the dependence of spike probability on rate of depolarization , as in these auditory neurons , provides a mechanism for temporal order discrimination between EPSPs .
In order to localize acoustic objects along the horizontal plane , the nervous system is able to distinguish microsecond differences in the arrival time of sound at the two ears , referred to as interaural time differences ( ITDs ) . Low sound frequencies are the most useful signals for detecting ITDs , and animals with good sensitivity below 1 , 500 Hz tend to perform best at this perception [1] . In mammals this computation is first performed by medial superior olivary neurons ( MSO ) in the auditory brain stem . Each MSO neuron receives two sets of excitatory inputs on its bipolar dendrites , with each set activated by one ear . When both excitatory pathways are activated within a narrow time window , the MSO neuron detects the coincident excitatory synaptic inputs and fires action potentials . When the pathways are activated asynchronously , the MSO neurons do not respond . Thus , an ITD response function is the representation of the variation of MSO discharge rate with the relative delay of the two inputs and , therefore , the position of a sound along the horizontal plane [2] . One influential theory holds that ITD encoding is based on an arrangement of axonal delay lines [3] . In this model , the differences in the sound's time of arrival at the two ears is transformed into a spatial map of ITD detecting neurons , sometimes referred to as a “place” code . Thus , an MSO neuron would discharge maximally when a specific ITD is exactly compensated by an internal delay that arises as a consequence of differences in the length of axons that are driven by the two ears . In fact , evidence for this mechanism has been found in birds and mammals [4]–[8] . However , since the discharge rate of many MSO neurons increases over the physiological range of ITDs [9]–[12] , this information could also be used to encode the azimuthal position , sometimes referred to as the “slope” code [13] . Additionally , there is evidence to suggest that inhibitory inputs to MSO play a role in tuning the response function within the physiological range of ITDs [11] , [14] . In previous models of ITD processing , the propagation time between the ipsi- and contralateral ears to the MSO neurons is implicitly assumed to be equal ( excluding Jeffress's internal delay lines ) . However , MSO neurons are positioned to one side of the brainstem , and the ipsilateral pathway is expected to be shorter than the contralateral . For example , one study has shown in vivo that many superior olivary neurons display longer latencies for the contralateral pathway [15] . Thus , any mechanism that relies on temporal precision must take this into account . We have tested this premise using a novel in vitro preparation that preserves each pathway . Our results support a new mechanistic explanation for the compensation of a longer contralateral response latency , and the positioning of the ITD response function in the physiological relevant range . The mechanism takes advantage of a difference in the dynamics of ipsi- and contralateral excitatory synaptic inputs . Using a computational model , we demonstrate that these asymmetric excitatory synaptic dynamics can significantly alter the ITD responses of MSO neurons .
Asymmetries in circuit architecture can have a significant effect on ITD processing . Specifically , the contralateral projections from ventral cochlear nucleus ( VCN ) to MSO are longer than those from the ipsilateral side ( Figure 1A , difference in afferent lengths between ipsilateral VCN to MSO and contralateral VCN to MSO ≈2 . 45 mm; Paul Nakamura and Karina Cramer , personal communication ) . To measure this difference functionally we used a thick brain slice preparation from gerbils that preserves the afferent pathways to the superior olivary complex ( Figure 1A; see Methods ) . Whole cell recordings were obtained from MSO neurons while activating each pathway at the same anatomical position on each side; the pathway between the stimulation point and the cochlea , which is eliminated in this preparation , is assumed to be identical for each side ( Figure 1A ) . We first found that the response latency did , in fact , differ between the two pathways . An analysis of evoked postsynaptic potentials ( PSPs ) and currents ( PSCs ) showed that the latencies to peak of contralateral responses were on average about 500 µs longer than those of ipsilateral responses on the same recorded neuron ( Figure 1B , C; average differences in latency to peak for PSPs: 573±62 µs , n = 54; for PSCs: 589±81 µs , n = 37 , see Methods section ) . This difference was apparent on a cell-by-cell basis because the difference of latencies ( contralateral - ipsilateral ) was significantly different than zero ( see gray bars in Figure 1B ) . In presenting the following experiments , we refer to the in vitro inter-stimulus time difference as ITD . Thus , if threshold were to depend solely on PSP amplitude , then the measured disparity in PSP latencies would predict that the peak ITD response would occur when the contralateral PSP leads by approximately 500 µs ( Figure 1D; predicted , thin curve ) . This ITD value is sufficiently large that the response function would fall largely outside of the physiological range for gerbils , which is ±130 µs [16] . In contrast , we found ITD response functions in which MSO firing rate was maximal when bilateral stimuli were delivered with smaller delays of ≈100 µs ( Figure 1D; measured , thick curve ) . This finding suggests that an intrinsic integration mechanism must compensate for the longer contralateral path . MSO neurons are exquisitely sensitive to the rate of depolarization . Therefore , in order to understand the integration of subthreshold bilateral inputs that lead to a spike , we examined the dynamics of synaptic inputs . Our starting assumption had been that synaptic properties are identical for each of the two excitatory inputs to MSO . We examined this assumption by measuring the rising PSP slopes because their time scale is within the same range as the coincidence detection window as manifested by the width of the ITD response function ( i . e . , 0–250 µs ) . Ipsilaterally evoked PSCs had significantly steeper rising slopes than contralateral PSCs ( Figure 1E , F ) ( ipsilateral: 1 . 04±0 . 15 nA/ms , contralateral: 0 . 62±0 . 06 nA/ms; p = 0 . 01 , n = 35 ) . This difference was apparent on a cell-by-cell basis because the difference of PSC slopes ( contralateral - ipsilateral ) was significantly different than zero ( see gray bar in Figure 1F ) . This result was independent of stimulus amplitude in all tested neurons ( see Figure S1 ) . The differences in the slopes of the PSCs could compensate , in part , for the disparity in delay between the two pathways . Our computational model ( below ) showed that even a modest asymmetry in rising slopes could shift the ITD response function from its hypothetical position ( based on latencies to peak ) to the observed location in the in vitro experiment ( Figure 1D ) . To determine how this asymmetry in excitatory synapse kinetics might compensate for the differences in path length , it was first necessary to determine the contribution of synaptic inhibition . To address this issue , we obtained ITD response functions under current clamp ( CC ) , before and after application of a glycine receptor antagonist , strychnine ( SN ) . As shown in Figure 2A and 2B , when synaptic inhibition was present ( control ) , the maximal firing occurred for contralateral leading stimulation , consistent with in vivo recordings [9]–[12] . When synaptic inhibition was blocked ( Figure 2A and 2B , SN ) the maximal firing rate was close to zero ITD , also consistent with an in vivo study [11] . We calculated the ITD at which peak firing probability occurred ( “best ITD” ) for the population of recorded neurons ( Figure 2C ) and found that under control conditions the peak was at 105±35 µs ( contra-leading ) , while under SN conditions it was at −62±38 µs ( ipsi-leading ) . Therefore , the effect of synaptic inhibition was to shift ITD tuning towards contralateral leading stimuli . Since this shift is in the wrong direction to compensate for the longer contralateral path , we next considered the role of asymmetric excitatory responses . In the presence of inhibition ( control ) , the ipsilaterally evoked normalized PSP slope was 2 . 71±0 . 12 ms−1 and the contralateral slope was 2 . 49±0 . 10 ms−1 ( Figure 2D ) . When inhibition was blocked ( SN ) , evoked EPSP slopes were significantly different between ipsi- and contralateral responses ( ipsilateral: 2 . 61±0 . 11 ms−1; contralateral: 2 . 21±0 . 14 ms−1 , see Figure 2D and also Figure S2 ) . Blockade of glycinergic inhibition increases the differences in the PSP slopes . More specifically , inhibition always increases the slope ( Figure 2D , from squares to triangles ) , but more so for the contralateral responses ( Figure 2D , right column ) . Such steepening occurs for either fast or slow inhibitory synaptic conductance transients ( see Figure S3 for theoretical support ) . In the fast case ( Figure S3 , left ) , the decaying brief IPSC coincides with rising EPSC and the summed current therefore rises faster than the EPSC alone . The effect is stronger on contralateral inputs because the IPSC will more fully decay during the EPSC rise . In the slow case , the IPSC transiently reduces the effective time constant , accelerating the rise although less dramatically than does a fast IPSC ( Figure S3 , right ) . The effect is stronger for contralateral inputs partly because integration of slower inputs is affected more by time constant changes ( leakage matters in addition to capacitive integration ) . Another major contributing factor related to active currents is explained below with our model . Thus , we confirmed that synaptic inhibition reduced the effect of shifting the ITD response function towards zero ITD , and leads us to suggest that the compensation arises from the excitatory asymmetry described above ( Figure 1 ) . How can such a small asymmetry in EPSP slope influence ITD sensitivity in MSO neurons ? We addressed this question by using a computational MSO neuron model that was driven by bilateral trains of excitatory and inhibitory inputs temporally modulated with a periodic function representing VCN responses to pure tone stimuli . Each cycle's composite input was generated from many small excitatory postsynaptic conductances ( EPSGs ) with statistics that depended on VCN afferent activity that varied with sound frequency and amplitude ( see Methods; [17] , [18] ) . Figure 3 shows a simplified version of the simulated MSO inputs to illustrate the variability of the composite EPSGs and integrated EPSPs due only to the jitter on the mini-EPSGs time release . Here , we exclude firing rate modulation throughout the sinusoidal input's cycles , although it is employed in the detailed model used for the simulated ITD functions . Using only differences in vector strength of the simulated inputs from the VCN arriving to each dendrite of the MSO neuron model we modeled differences in rising slope of the bilateral EPSPs ( Notice: without delaying the composite EPSP peak , see triangles in Figure 3 for EPSG peaks ) . These differences led to shifts in the ITD response function that are large enough to compensate for the longer contralateral input pathway . For a given EPSG input , the evoked EPSPs and spike threshold will be determined by the active currents . In MSO and other auditory processing centers , a low threshold potassium current ( IKLT ) exerts control on spike threshold [19]–[21] . This fast IKLT imposes a filtering effect on the synaptic inputs allowing only steep EPSG slopes to evoke an action potential [22] , [23] . Therefore , steeper EPSGs are more likely to trigger spikes , even when shallower EPSGs may have greater amplitude , as is shown in our simulations . When bilateral subthreshold inputs arrive at an MSO neuron , there is a higher probability of eliciting a spike when the steeper EPSG arrives first . Figure 4A shows how a pair of EPSGs , one fast and one slow , can produce a very different outcome , depending on their order of arrival . When a faster input arrives first this will enable spike generation ( Figure 4A and 4B , left side ) . When a slower input arrives earlier it leads to a slower rising EPSP that recruits more IKLT conductance , which hinders spike generation even though a faster EPSG arrives subsequently ( Figure 4A and 4B , right side ) . To show the essence of the ITD response function shift due to the asymmetry in the kinetics of the excitatory inputs we delivered inputs to the model with different vector strength ( Figure 3 ) and calculated their probability to evoke spikes for different input delays ( ITD response function , Figure 4C , D ) . If the contralateral composite EPSP was slower-rising , the bilateral combined EPSP had different rising dynamics when the ipsilateral inputs led than when the contralateral inputs led ( Figure 4C , EPSPs schematics ) . Consistent with previous findings [21] , [24]–[26] , the shallower-leading combined EPSP was associated with a lower probability of firing . Therefore , the ITD function shifted towards the ipsilateral leading side ( Figure 4C ) . The asymmetry in firing rate probability caused by an asymmetry in inputs' rising slopes is due to the voltage-dependence of IKLT conductance . We explain this ( Figure 4D ) by showing that with the same set of bilateral asymmetric EPSPs that generate a shift of ∼400 µs ( Figure 4D , thick black curve ) , the shift of the ITD's response function disappears ( Figure 4D , brown curve ) if we fix the IKLT conductance at its resting value , in order to maintain the neuron model's time constant and input resistance intact . We next asked whether the asymmetry in the excitatory inputs could compensate for an intrinsic input delay of ≈500 µs as measured in our in vitro preparation . The simulations showed that the integration of hypothetical symmetric EPSPs led to an ITD response function that was shifted to the contralateral leading side due to the intrinsic contralateral axonal delay ( Figure 5 , thin black curve ) . When asymmetric EPSGs were introduced in the model to generate EPSP slopes similar to those found in our experiments , the ITD function shifted towards the ipsilateral-leading direction due to the favorable response when a steep EPSP occurs first ( Figure 5 , thick black curve ) . Our experimental data were consistent with this theoretical explanation: most of the neurons displayed this asymmetry in excitatory inputs . Thus , when we subtracted contralateral slope from ipsilateral slope for each individual neuron , the average difference was 0 . 69±0 . 18 nA/ms for EPSCs and 0 . 40±0 . 12 ms−1 for normalized EPSPs . Inclusion of synaptic inhibition made the simulated EPSPs less asymmetric . The hyperpolarization from inhibition transiently reduced IKLT . The reduction of this conductance would no longer favor spike generation when fast EPSPs are followed by slow EPSPs . The ITD response function was reduced on the ipsilateral-leading side , giving the appearance of a shift towards the contralateral-leading side ( Figure 5 , orange and violet curves ) , as observed experimentally in vitro ( Figure 2B ) and as reported previously in vivo [11] , [14] .
Our experimental and computational findings identified key biophysical factors that , together , position the ITD response function in the biologically relevant range . We first confirmed the presence of an internal delay of the longer contralateral pathway ( Figure 1B ) . In itself , this would cause MSO neurons to fire mostly to ITDs with stimuli having large contralateral leading stimuli that are outside the physiological range . Our experimental and computational results suggest a novel excitatory synaptic mechanism that could compensate for the disparity in path length . An asymmetry in the slopes of EPSPs ( Figure 2D ) can bias the ITD coding in favor of the ipsilateral-leading inputs ( Figures 4 and 5 ) , and this repositions the ITD function within the physiological range , as found in vivo [9]–[12] . The presence of a fixed internal latency difference will affect all models of ITD processing . Jeffress [3] assumed tacitly that the two paths were equal in length except for the small differences along one spatial axis of the encoding nucleus . Others have suggested that the shorter path length from the ipsilateral ear is compensated by an additional span of axon ( e . g . , see schematic in [2] ) , or a difference in myelination between the two pathways [27] . If the difference in path length to MSO for the gerbil is ≈2 . 45 mm ( Paul Nakamura and Karina Cramer , personal communication ) , then our electrophysiological measurements of response latency difference of 500 µs would yield a propagation speed of 4 . 9 m/s . Thus , it appears that there is an internal latency difference to gerbil MSO that is not compensated for by an axonal property . It is this functional characteristic that must be addressed if MSO neurons are to encode ITDs in the physiological range ( ±130 µs; [16] ) . Our electrophysiological measurements indicate that the rising PSP slope is larger for the ipsilateral input to MSO neurons on a cell-by-cell basis ( Figures 1 and 2 ) . The functional implications for this finding are illustrated in a computational model which demonstrates that this property can compensate for the aforementioned difference in path length ( Figure 5 ) . The general principle , which is that the rising slope of an EPSP determines the probability of firing , is consistent with findings from other systems [21] , [25] , [26] , [28] . Here , we have adapted this principle to resolve the general problem of compensating for different input latencies due to path length . How might the EPSP asymmetry arise ? In the model we allowed for more jitter in the arrival times of identically shaped unitary ( minimal ) EPSPs on the contralateral side , which slowed the rise of the composite EPSPs . This idealization , for demonstrating plausibility in the context of our point neuron model , could be elaborated and explored in a neuron model that has bilateral dendrites with cable properties [29] . Many alternative mechanisms are also possible . Bilateral differences in dendritic morphology or the dendritic positioning of excitatory terminals could also lead to an asymmetry in the rising slope of composite EPSPs [30] , [31] . Although longer electrical distances would promote broadening of composite EPSPs in a passive dendrite , IKLT in the dendrites can reduce the effect by shortening the tail of EPSPs as they propagate toward the soma in MSO neurons and cable models [32] . Alternatively , the distribution of active currents could modulate the dendritic integration of synaptic inputs . For example , dendritic sodium channels are able to selectively boost EPSPs on one dendrite , and this would modify their rising slope ( cortex: [33] ) . It is important to consider the in vivo time scale of inhibition and excitation because it will determine the temporal integration window and the extent to which ITD curves will be affected by the mechanisms described above . It is possible that the time scales in vivo are faster than in the brain slice because a cell is in a high conductance state ( e . g . , many more active inputs as compared to brain slice ) . In addition , the degree of afferent synchrony could have been unnaturally high in our preparation because the stimulus simultaneously recruits all VCN afferents to MSO . However , the model demonstrated that the effect of slope is robust when implemented with vector strength values that have been reported in vivo ( Figure 3; using model from [18] ) . Since we also showed that synaptic inhibition somewhat counteracts the shifting effect of the asymmetric excitation , it is important to consider its kinetics . The time scale for inhibition has only been studied in vitro , and even the fastest IPSPs have either been recorded from animals between 12 to 25 postnatal days [34] , or at room temperature [35] . Interestingly , we found that while the magnitude of the inhibitory effect depends on IPSP time scale , it is likely to play an important role in ITD coding no matter what the actual time scale value turns out to be ( Figure 5; Figure S3 ) . The faster rising EPSPs that were elicited by ipsilateral afferents could overcome the penalizing effect of a rapidly activating outward current like IKLT ( Figure 4B ) . Many previous reports have demonstrated a robust effect of IKLT on the integration time of EPSPs [19] , [20] , [28] . In this study , we applied this property to anatomically independent bilateral inputs and demonstrated computationally that IKLT influenced the ITD function . Together , our findings lead us to propose a general principle . Passive neuronal integration to a threshold would not distinguish the temporal ordering in inputs that may have different rising slopes . Subthreshold dynamic negative feedback such as IKLT ( comparably as fast as integration ) will bias the integration . Firing will be favored when the steeper-rising input occurs first . Inhibition , by deactivating the negative feedback , can reduce the bias . The competition between these two effects in the MSO , leads to a positioning of the ITD response function with its slope in the physiological range , as seen in vivo [11] . Thus , the synaptic property compensates for the intrinsic latency disparity . Time-difference encoding could exploit these mechanisms in this extremely short window of integration time ( 130 µs ) or , more generally , in other windows where the biophysical components and time scales are appropriately matched . Generalizing , we propose a novel neuronal mechanism for temporal order selectivity . Subthreshold dynamic negative feedback can increase a neuron's firing probability to segregated subthreshold inputs when faster ones precede slower ones , even if the slower one is of similar or larger amplitude .
All protocols were reviewed and approved by New York University Institutional Animal Care and Use Committee . Postnatal day ( P ) 17–25 gerbils ( Charles River ) were used to generate thick ( 450–500 µm ) horizontal slices ( N = 91 ) from the ventral auditory brainstem . Each slice contained the MSO nucleus , the medial nucleus of the trapezoid body ( MNTB ) , and the lateral nucleus of the trapezoid body ( LNTB ) . Animals were deeply anesthetized ( chloral hydrate , 400 mg/kg ) , perfused intracardially with artificial cerebrospinal fluid ( ACSF: 123 mM NaCl , 4 mM KCl , 1 . 2 mM KH2PO4 , 1 . 3 mM MgSO4 , 24 mM NaHCO3 , 15 mM glucose , 2 . 4 mM CaCl2 , 0 . 2 mM ascorbic acid; pH = 7 . 35 after bubbling with 95% 02/5% CO2 ) at 32°C . The brain was then dissected free in 32°C oxygenated ACSF , and one horizontal slice was obtained with a Leica vibratome . The slice was incubated at 36°C for 20 min , and at 22°C for 1 h before being transferred to the recording chamber where oxygenated ACSF was perfused at a rate of 2 ml/min at 32°C; temperature was regulated by L&N temperature controller . The afferents arising from both VCNs were visualized as compact bundles . Thus , ipsilateral and contralateral bundles were stimulated at the site of their origins with bipolar tungsten electrode and stimulation was delivered by two stimulus isolation units ( Dagan ) . The distance between the MSO and the two stimulation sites was approximately 0 . 5 mm for the ipsilateral pathway and 1 . 5 mm for contralateral pathway . Whole cell current-clamp recordings were obtained mostly from medial and dorsal MSO neurons ( Axoclamp2A ) . The recordings and stimulation were computer driven ( Windows XP ) through Labview software ( National Instruments ) . The neurons were visually identified using infra-red differential interference contrast ( IR-DIC ) microscopy ( Olympus ) . The internal patch solution contained ( in mM ) 127 . 5 potassium gluconate , 0 . 6 EGTA , 10 HEPES , 2 MgCl2 , 5 KCl , 2 ATP , 10 phosphocreatinine ( Tris salt ) , and 0 . 3 GTP ( pH 7 . 2 ) in the case of CC protocol and ( in mM ) 127 . 5 cesium gluconate , 0 . 6 EGTA , 10 HEPES , 2 MgCl2 , 5 KCl , 2 ATP , 5 QX-314 , 10 phosphocreatinine ( Tris salt ) , and 0 . 3 GTP ( pH 7 . 2 ) in the case of voltage clamp ( VC ) protocol . In order to block synaptic inhibitory inputs , we used SN in CC experiments and SN and bicuculine to block glycinergic/gabaergic inputs in VC experiments . EPSPs in CC and EPSCs in VC were recorded when single square pulses repeatedly ( 20 Hz ) of 25–50 µs were delivered via the stimulating electrodes to initially evoke minimum amplitude responses , maximum amplitude subthreshold responses , and subthreshold-unilateral/suprathreshold-bilateral responses . High stimulus currents ( 0 . 5 to 10 . 0 mA ) and short pulse durations ( 25–50 µs ) were used to avoid the overlap of stimulus artifact with evoked responses . The data were analyzed following these basic criteria: slopes of the rising phase ( 20% to 80% ) of the responses , for unilateral stimulations . For all the parameters that were measured for bilateral stimulations responses ( i . e . , peak-delay , slope ) , the intervals of confidence ( p values ) were computed using t test over the difference between ipsi- and contralateral responses on the same neuron . All data variability is expressed in standard deviation . In addition , 100 to 500 Hz stimulus trains of 10 stimuli were applied ( total number of spikes per train delay were counted ) to generate ITD tuning response function . A minimum of four trials were run to get a smooth ITD response function . In the case of CC data the slopes of PSPs and EPSPs were computed when bilateral responses were similar in amplitude , to avoid differential effect of active currents , and were normalized to decrease population variability due to biophysical heterogeneity among neurons . We used a computational model of MSO neurons based on the parameters described by Rothman and Manis ( 2003 ) [36] for a point VCN neuron [36] . We chose a membrane time constant of 0 . 3 ms , similar to the one reported for MSO neurons after P20 [20] . Bilateral input trains with different delays were created by injecting currents ( conductance based synaptic-like currents ) such that the trains of EPSPs consisted of composite minimal EPSPs ( 32 or 64 minimal EPSPs were used to create a ∼8 mV composite EPSP; more EPSGs were used for higher input frequencies ( 1 . 1 KHz ) to generate a smooth voltage time course ) . Minimal EPSGs had fixed form: alpha functions with time constant τsyn of 0 . 1 ms for excitation and 0 . 4 for inhibition , scaled to have specified area and peak proportional to 1/τsyn . Different minimal EPSG statistics led to different slopes and half-widths , which are summed in order to create the composite suprathreshold EPSGs ( see Figure S4 ) . These EPSPs have envelopes resembling alpha functions with time constants that ranged from 0 . 1 to 0 . 8 ms [17] . This range of ( in vivo based ) EPSP time constants was slightly faster than those obtained from our experiments because our recordings were made at 32°C and the simulations were performed at 37°C . The same results were obtained using values of rising EPSP slopes from our experiments at 22°C as well as the kinetics of our computational model , to eliminate any temperature effect . The asymmetry in simulated EPSP kinetics was modeled by varying the jitter of unitary events . The amount of jitter was based on the observed variability in EPSC amplitudes , slopes , and half-widths obtained in our brain slice recordings . ITD functions were created from bilateral EPSP or PSP trains ( 40 cycles ) at frequencies ranging from 250 to 1 , 100 Hz . A minimum of 10 trials ( per ITD ) were run to get a smooth ITD response function . The differential equations of the model were integrated numerically using fourth-order-Runge-Kutta scheme with a time step between 1 and 0 . 25 µs; refining the time step did not lead to noticeable differences in the computed solutions . In all the simulations the contralateral inhibitory input leads the contralateral excitation by 0 . 2 ms . This time difference was imposed between the peak of the composite IPSPs and the composite EPSPs from the contralateral input side . The result in Figure 4 showing that inhibition shifts the ITD response function towards contralateral leading side holds even for bigger delays between contralateral inhibition and excitation ( unpublished data ) . | Animals can locate the source of a sound by detecting microsecond differences in the arrival time of sound at the two ears . Neurons encoding these interaural time differences ( ITDs ) receive an excitatory synaptic input from each ear . They can perform a microsecond computation with excitatory synapses that have millisecond time scale because they are extremely sensitive to the input's “rise time , ” the time taken to reach the peak of the synaptic input . Current theories assume that the biophysical properties of the two inputs are identical . We challenge this assumption by showing that the rise times of excitatory synaptic potentials driven by the ipsilateral ear are faster than those driven by the contralateral ear . Further , we present a computational model demonstrating that this disparity in rise times , together with the neurons' sensitivity to excitation's rise time , can endow ITD-encoding with microsecond resolution in the biologically relevant range . Our analysis also resolves a timing mismatch . The difference between contralateral and ipsilateral latencies is substantially larger than the relevant ITD range . We show how the rise time disparity compensates for this mismatch . Generalizing , we suggest that phasic-firing neurons—those that respond to rapidly , but not to slowly , changing stimuli—are selective to the temporal ordering of brief inputs . In a coincidence-detection computation the neuron will respond more robustly when a faster input leads a slower one , even if the inputs are brief and have similar amplitudes . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"neuroscience/sensory",
"systems",
"computational",
"biology/computational",
"neuroscience",
"computational",
"biology/systems",
"biology"
] | 2010 | Asymmetric Excitatory Synaptic Dynamics Underlie Interaural Time Difference Processing in the Auditory System |
Accumulated research has demonstrated the beneficial effects of dietary restriction on extending lifespan and increasing cellular stress resistance . However , reducing nutrient intake has also been shown to direct animal behaviors toward food acquisition . Under food-limiting conditions , behavioral changes suggest that neuronal and muscle activities in circuits that are not involved in nutrient acquisition are down-regulated . These dietary-regulated mechanisms , if understood better , might provide an approach to compensate for defects in molecules that regulate cell excitability . We previously reported that a neuromuscular circuit used in Caenorhabditis elegans male mating behavior is attenuated under food-limiting conditions . During periods between matings , sex-specific muscles that control movements of the male's copulatory spicules are kept inactive by UNC-103 ether-a-go-go–related gene ( ERG ) –like K+ channels . Deletion of unc-103 causes ∼30%–40% of virgin males to display sex-muscle seizures; however , when food is deprived from males , the incidence of spontaneous muscle contractions drops to 9%–11% . In this work , we used genetics and pharmacology to address the mechanisms that act parallel with UNC-103 to suppress muscle seizures in males that lack ERG-like K+ channel function . We identify calcium/calmodulin-dependent protein kinase II as a regulator that uses different mechanisms in food and nonfood conditions to compensate for reduced ERG-like K+ channel activity . We found that in food-deprived conditions , calcium/calmodulin-dependent protein kinase II acts cell-autonomously with ether-a-go-go K+ channels to inhibit spontaneous muscle contractions . Our work suggests that upregulating mechanisms used by food deprivation can suppress muscle seizures .
The excitability of neuromuscular circuits must be regulated to ensure appropriate behavioral responses under different conditions . When cell excitability is irregular due to channelopathy defects , inappropriate motor output can lead to different medical conditions . One member of the ether-a-go-go ( EAG ) family of K+ channels , the human ether-a-go-go–related gene ( hERG ) –encoded , delayed inward , rectifying voltage-gated K+ channel , has received attention due to its association with the cardiac condition long QT syndrome [1] . Most mutations in hERG reduce channel conductance , which causes prolonged depolarizations that result in cardiac arrhythmias [2–6] . Due to the physiological significance of this protein , there is ongoing research into the biophysical properties of hERG K+ channel function . However , little is known about other signaling pathways that act with these channels to control excitable output within specific physiological contexts . We reported previously that the Caenorhabditis elegans homolog of hERG , UNC-103 , regulates the movements of the male's copulatory spicules before and during male mating behavior [7] . Prior to mating , the two spicules are held within the male tail via their attachments to dorsal and ventral protractor and retractor muscles . During mating , males rhythmically contract the spicule muscles to protract their spicules through the hermaphrodite vulva . Similar to the irregular cardiac muscle depolarizations caused by defective hERG , deletion of unc-103 causes spontaneous contractions of the spicule muscles , inducing spicule protraction in the absence of mating cues . Interestingly , the unc-103 ( 0 ) deletion phenotype is incompletely penetrant , suggesting that redundant pathways regulate this motor output [7 , 8] . The incidence of spontaneous spicule muscle seizures in unc-103 ( 0 ) males is reduced by food deprivation [9] . This suggests that signal transduction pathways , activated under nutrient-poor conditions , specifically modulate the expression of the unc-103 ( 0 ) phenotype . Food deprivation globally modulates many behaviors to direct the animal toward food acquisition [10–13] . In C . elegans and higher organisms , dietary deprivation not only represses non–food-foraging behaviors , but also has positive physiological effects on stress resistance and aging . This suggests mechanisms , activated under food-deprived conditions , are possible targets to treat illnesses caused by channelopathies and stress . Food deprivation likely modulates diverse behavioral outputs by acting on different molecular regulators within specific neuromuscular circuits . We previously reported that feeding behavior influences the spicule circuit via signaling from the muscular feeding organ , the pharynx , and pharyngeal-associated neurons [9] . In this study , we used the feeding state/spicule protraction relationship in C . elegans to identify a key molecular regulator involved in suppressing defective ERG-like K+ channel/UNC-103 muscle contractions . We found that calcium/calmodulin-dependent kinase II ( CaMKII ) is a central regulator within the spicule circuit that modulates sex-muscle excitability under both well-fed and food-deprived conditions . Biochemical and genetic studies in C . elegans and Drosophila have independently linked CaMKII activity with the K+ channel function of ERG and its close relative EAG [14–16] . In this study , we describe two physiological contexts in which UNC-43 , the worm homolog of CaMKII , works with ERG-like and EAG K+ channels to regulate a specific muscle output . Specifically , we propose that under well-fed conditions , CaMKII/UNC-43 works redundantly with ERG-like K+ channel/UNC-103 to suppress spicule protraction in periods between mating , while under food-deprived conditions , CaMKII/UNC-43 and EAG K+ channels suppress spicule protraction via a parallel mechanism .
Previously , we found ERG-like K+ channel/UNC-103 suppresses spontaneous sex-muscle contraction prior to mating [7] . C . elegans are normally grown on nematode growth medium ( NGM ) plates containing a lawn of OP50 bacteria . Under these conditions , 42% of unc-103 ( 0 ) males display constitutive spicule protraction ( Table 1 ) . Since deletion of unc-103 causes less than half of the males in a population to protract their spicules permanently , we hypothesized that additional mechanisms regulating sex-muscle contraction are functioning in the non–spicule-protracted males . To determine these additional mechanisms , we analyzed a mutant allele , sy574 , that was isolated in the same screen that identified unc-103 as a regulator of spicule protraction [7] . sy574 induces spicule protraction in 56% of males ( Table 1 ) ; hermaphrodites move , lay eggs , defecate , and respond to mechanosensation normally ( unpublished data ) , demonstrating that sy574 disrupts one behavior . sy574 and ERG-like K+ channel/unc-103 ( 0 ) superficially induce the same phenotype and might affect the same regulatory pathway . In sy574 and unc-103 ( 0 ) males , we measured the timing of constitutive spicule protraction to determine the extent of the similarities between the mutant phenotypes . Males were allowed to develop into adults , and then checked every hour for the mutant phenotype . We found that after 5 h , 30% of unc-103 ( 0 ) males had protruding spicules . In contrast , only 25% of sy574 males ( out of 56% that eventually display the mutant phenotype ) displayed spontaneous spicule protraction . Thus , the majority of sy574 males have sex-muscle seizures later than unc-103 ( 0 ) males , suggesting these two mutations affect spicule protraction differently . To determine if sy574 worked in a different pathway from unc-103 , we constructed a sy574; unc-103 ( 0 ) double mutant . A total of 97% of unc-103 ( 0 ) ; sy574 males displayed spontaneous spicule protraction , which is higher than the 75% expected if the two mutations acted additively ( Table 1 ) . This suggests that sy574 disrupts a separate signaling pathway from unc-103 ( 0 ) . We used single nucleotide polymorphism mapping to position the sy574 lesion to a 570-kb region on Chromosome IV between cosmids R102 and K08F4 [17] . We then injected PCR products of candidate genes into sy574 animals to test for rescue , and we also performed complementation tests with alleles of genes located in the region . Our complementation tests suggested that the recessive sy574 allele affects the gene unc-43 , since four loss-of-function ( lf ) unc-43 alleles did not complement the sy574 phenotype ( Table 2 ) . The noncomplementing unc-43 ( lf ) alleles , including the nonsense unc-43 ( n1186 ) allele , induce general locomotor defects , muscle seizures , and constitutive spicule protraction . Since sy574 animals only displayed one of many unc-43 ( lf ) phenotypes , we reason that sy574 is a weak lf allele that encodes a protein with sufficient function to regulate muscles and neurons involved in general C . elegans behaviors . However , the spicule protraction circuit must be more sensitive than other cells to perturbations in UNC-43 function . The one allele that complemented sy574 is unc-43 ( e408 ) ( Table 2 ) . unc-43 ( e408 ) males display locomotor and muscle seizure defects similar to animals containing the nonsense unc-43 ( n1186 ) allele; however , unc-43 ( e408 ) males do not display spontaneous spicule protraction ( Table 2 ) . Although unc-43 ( e408 ) animals do not display this abnormality , unc-43 ( e408 ) does affect spicule protraction , and its analysis is described in later sections . unc-43 encodes the one C . elegans copy of CaMKII , a serine/threonine kinase responsible for phosphorylating multiple substrates that control many cellular functions [16] . CaMKII contains three functional domains: a N-terminal catalytic region responsible for substrate recognition , binding , and phosphorylization; an autoinhibitory domain responsible for blocking the catalytic region and keeping the protein inactive in the absence of calcium; and a C-terminal self-association domain by which the enzyme forms complexes of six- and 12-member rings [18] . Sequencing unc-43 from sy574 animals revealed two point mutations located in different functional domains . One mutation changes a glycine to glutamate at amino acid 170 , near the substrate recognition domain of the catalytic region ( Figure 1A ) [19 , 20] . The second sy574 mutation is located in the self-association domain and changes an alanine to valine at amino acid 465 ( Figure 1A ) . Genetic analyses of unc-43 ( sy574 ) and unc-103 ( 0 ) suggest that CaMKII/UNC-43 and ERG-like K+ channel/UNC-103 act redundantly to suppress spontaneous sex-muscle contractions when food is available . Therefore , we asked which cells require their function under these standard conditions . The C . elegans male contains two spicules , each associated with two protractor and two retractor muscles and a nonessential accessory muscle ( the anal depressor ) . Contraction of the protractor muscles forces the spicules through the cloacal opening , whereas contraction of the retractor muscles draws the spicules back into the tail . The SPC , hook , and the postcloacal sensilla neurons are functionally associated with the spicule muscles and trigger contractions upon mating stimulation [8 , 21 , 22] . Work by others using antibodies to rat CaMKII showed that in hermaphrodites , CaMKII/UNC-43 is broadly expressed in neurons , muscles , and intestines [16] . To determine if unc-43 is expressed in the male spicule circuit , we PCR-amplified an 11-kb region upstream of the unc-43 start codon , ligated the PCR fragment to the cyan fluorescent protein ( CFP ) gene , and injected the ligation product into C . elegans . In males , we found that CaMKII is broadly expressed in excitable cells , including the spicule protractor and retractor muscles , SPC , and postcloacal sensilla neurons ( Figure 1B and 1C ) . For unc-103 , we previously reported the unc-103 loci are expressed from at least six promoters ( promoters Punc-103A through Punc-103F ) . These promoters express mRNA with distinct first exons that give rise to the isoforms unc-103A through unc-103F . Many of the promoters express unc-103 broadly in neurons and muscles in both sexes . However , promoter Punc-103E expresses unc-103 in approximately seven neuron pairs in the head , two pharyngeal neurons , and the spicule protractor muscles , but not spicule retractors in the tail , whereas Punc-103F expresses unc-103 in multiple neurons , including the SPC and postcloacal sensilla neurons that control spicule muscle contractions [23] . To determine where CaMKII/unc-43 is functioning in the male to suppress premature sex-muscle contraction , we created an unc-43 cDNA construct using the full-length isoform unc-43g [24 , 25] . To determine if our construct was functional , we drove unc-43 cDNA expression using the ubiquitously expressed hsp-16 heat-shock promoter . We injected this construct into unc-43 ( sy574 ) animals and found that heat-shocked males showed significant reduction in sy574-induced spicule protraction compared to control siblings ( Table 3 ) . Since CaMKII/unc-43 is expressed in many tissues , we wanted to identify where unc-43 function is required to control spicule protraction . Using the aex-3 promoter , which is broadly expressed in neurons , and the tnt-4 promoter , which is expressed in the pharynx , we determined that unc-43 expressed in these cell types has no effect on unc-43 ( sy574 ) –induced spicule protraction ( Table 3 ) [9 , 26] . However , expressing unc-43 via the lev-11 pan–body-wall and sex-muscle promoter reduced the mutant phenotype ( Table 3 ) [27] . Thus , CaMKII/unc-43 is required in the muscles to prevent premature muscle contraction . Once we had identified that muscle CaMKII/unc-43 regulates spicule protraction , we asked which of the two point mutations in unc-43 ( sy574 ) affects normal kinase function . To address this , we generated an unc-43 isoform lacking the self-association domain , but containing the kinase and inhibitory domains . We expressed this construct in muscles using the lev-11 promoter , and found that it also rescued unc-43 ( sy574 ) –induced protraction ( Table 3 ) . This suggests that the sy574 mutation in the catalytic domain , and not in the self-association domain of unc-43 , is responsible for constitutive spicule protraction . Since CaMKII/unc-43 appears to regulate sex-muscle output via a separate mechanism than ERG-like K+ channel/unc-103 , we asked if unc-103 also acts in muscles . The unc-103 gene in C . elegans contains six first exons that produce tissue-specific isoforms . In hermaphrodites , the unc-103E isoform expressed from Punc-103E in sex muscles is required to rescue an unc-103 ( 0 ) –induced egg-laying defect [23] . Therefore , in addition to determining which tissues require functional unc-103 to control spicule protraction , we also asked if the rescue was isoform specific . We used isoform-specific unc-103 genomic constructs , and found that the neuron-specific isoform unc-103F , expressed via the aex-3 panneuronal promoter or the neuronal-specific Punc-103F promoter , did not rescue unc-103 ( 0 ) –induced protraction ( Table 3 ) . However , the unc-103F and unc-103E isoforms driven in muscles via the lev-11 panmuscle promoter or by the sex-muscle promoter Punc-103E restored control of spicule protraction in unc-103 ( 0 ) males ( Table 3 ) . Thus , similar to CaMKII/unc-43 , ERG-like K+ channel/unc-103 acts in muscles to regulate spicule protraction , and unlike hermaphrodite egg-laying behavior , specific unc-103 isoforms are not essential for regulation . Interestingly , although both CaMKII/UNC-43 and ERG-like K+ channel/UNC-103 act in muscles , the functional expression of unc-43 in regulating spicule muscle contraction is not as restricted as unc-103 . unc-43 , when expressed broadly in body-wall muscles and all sex muscles from the lev-11 promoter , rescued the unc-43 ( sy574 ) –induced spicule protraction . However , unc-43 expressed in body-wall muscles from the acr-8 promoter ( see Text S1 ) or expressed in the sex muscles from the Punc-103E promoter was not sufficient to suppress constitutive spicule protraction ( Table 3 ) . This suggests that under standard conditions , UNC-43 might act in both body-wall and spicule protractor muscles to regulate spicule protraction behavior . Alternatively , since lev-ll but not the Punc-103E promoter drives transcription in the retractor muscles , UNC-43 might be required in the protractor and retractor muscles , whereas UNC-103 is required only in the protractor muscles . Since both CaMKII/unc-43 and ERG-like K+ channel/unc-103 are functioning in muscles , we asked what other proteins might work with these molecules . From previous work , lf alleles of L-type voltage-gated Ca++ channel ( L-VGCC ) /egl-19 and ryanodine receptor Ca++ channel ( RyR ) /unc-68 suppress mutant unc-103–induced spontaneous protraction [7 , 28 , 29] . Similar to the previously reported interactions between mutant unc-103 and egl-19 , unc-43 ( sy574 ) required wild-type egl-19 to induce spicule protraction ( Table 1 ) . Interestingly , unc-43 ( sy574 ) ; unc-68 ( 0 ) males still spontaneously protract their spicules , suggesting that , in contrast to ERG-like K+ channel/unc-103 , CaMKII/unc-43 is not responding to the influx of calcium via RyR/unc-68 ( Table 1 ) . L-VGCC/EGL-19 and RyR/UNC-68 have different roles in controlling spicule muscle contraction during male mating . Rhythmic contractions mediated by UNC-68 result in the spicules prodding the hermaphrodite's vulva , whereas tonic contraction mediated by EGL-19 forces the spicules through the vulval slit . The acetylcholine agonist levamisole activates muscle contraction though UNC-68 , while EGL-19 is activated by the acetylcholine agonist arecoline ( ARE ) [8] . Since unc-43 ( sy574 ) was suppressed by egl-19 ( lf ) , but not unc-68 ( 0 ) , the ARE , not the levamisole , stimulatory pathway is perturbed by the unc-43 ( sy574 ) lesion . To dissect the CaMKII/unc-43 pathway , we looked at unc-43 mutant responses to ARE . Virgin adult males at 1 d old that had not protracted their spicules were placed in various concentrations of ARE and observed for spicule protraction . The concentration at which 90% of wild-type males protract their spicules ( EC90 ) was 579 μM , whereas the EC90 of unc-43 ( sy574 ) males was 268 μM ( Figure 2A and 2D ) . In contrast to wild-type , the dominant gain-of-function allele unc-43 ( n498gf ) and the lf allele unc-43 ( e408 ) were greater than ten times more resistant to the drug ( Figure 2A and 2D ) . Thus , the unc-43 ( e408 ) –encoded kinase displays some gain-of-function properties in the spicule protraction circuit , although in all other behaviors it displays loss-of-function properties . To identify how unc-43 ( e408 ) might function in the spicule protraction circuit , we sequenced the unc-43 gene from e408 animals and found a point mutation that causes a serine to leucine change at amino acid 179 ( Figure 1A ) . This amino acid change is near the substrate recognition site of the catalytic domain , suggesting the unc-43 ( e408 ) lesion is altering the kinase's interactions with its substrates . The n498gf and e408 alleles allowed us to determine if CaMKII-induced ARE resistance is mediated by ERG-like K+ channel/unc-103 . Previously , unc-103 was shown to be downstream of unc-43 in respect to defecation behavior [16] . However , our analyses of unc-43 ( sy574 ) and unc-103 ( 0 ) suggest that unc-43 and unc-103 can act in separate pathways . One explanation for this discrepancy is that CaMKII/UNC-43 can function in multiple pathways to control behaviors , including spicule protraction behaviors ( Figure 3 ) . It is likely the unc-43 alleles used in this report up- or downregulate different facets of spicule protraction regulation . To test if unc-43 also activates unc-103 in the spicule circuit , we constructed double mutants of unc-43 ( n498gf ) and unc-43 ( e408 ) with unc-103 ( 0 ) and found they are less resistant to ARE ( Figure 2B ) . Thus , the genetics of the sy574 allele and the pharmacology of the n498gf and e408 alleles demonstrate that UNC-43 can act concurrently upstream and parallel to UNC-103 ( Figure 3 ) . However , the unc-103 ( 0 ) –induced reduction of unc-43 ( gf ) and unc-43 ( e408 ) pharmacology is not complete , indicating there are other factors CaMKII activates to suppress muscle excitability . To identify other factors that could be activated by CaMKII/unc-43 , we tested unc-43′s interaction with the ether-a-go-go ( EAG ) K+ channel/egl-2 [30] . We considered EAG K+ channels because work in Drosophila showed that direct phosphorylation by CaMKII upregulates channel activity [15 , 30 , 31] . First , we asked if egl-2 expresses in similar tissues to unc-43 and unc-103 . Previous reports showed egl-2 expression in the sensory neurons and sex muscles of hermaphrodites [30] . We found similar expression in males , including expression in the sex muscles but not neurons in the spicule protraction circuit ( Figure 1D and 1E ) . Next , we asked if egl-2 acts downstream of unc-43 by combining unc-43 ( lf ) mutations with a gain-of-function egl-2 allele . egl-2 ( n693gf ) was able to reduce the nonsense allele unc-43 ( n1186 ) and unc-43 ( sy574 ) –induced spicule protraction , but , interestingly , had no effect on unc-103 ( 0 ) –induced protraction ( Table 1 ) . The genetic interaction between egl-2 ( n693gf ) and unc-43 ( sy574 ) , but not unc-103 ( 0 ) , is consistent with unc-43 ( sy574 ) disrupting a pathway parallel to UNC-103–mediated regulation ( Figure 3 ) . Since egl-2 ( n693gf ) suppressed unc-43 mutant alleles , we asked if the effects of activated CaMKII/unc-43 require functional EAG K+ channel/egl-2 . We isolated the rg4 deletion ( 0 ) allele to address this question and found that egl-2 ( 0 ) animals are superficially wild-type ( Table 1 ) . We then combined egl-2 ( 0 ) with unc-43 ( n498gf ) and unc-43 ( e408 ) and found that egl-2 ( 0 ) was able to reduce the ARE sensitivity of both mutations ( Figure 2C ) . Though deletions in egl-2 and unc-103 increased drug sensitivity of both unc-43 mutant backgrounds , neither alone restored it to wild-type levels . We generated triple mutants containing egl-2 ( 0 ) and unc-103 ( 0 ) with unc-43 ( gf ) or unc-43 ( e408 ) to see if removing both K+ channels increase ARE sensitivity . We found that the EC90 of unc-103 ( 0 ) ; unc-43 ( gf ) ; egl-2 ( 0 ) and unc-103 ( 0 ) ; unc-43 ( e408 ) ; egl-2 ( 0 ) were 1 . 2 mM and 483 μM , respectively ( Figure 2B–2D ) . Thus , both ERG-like K+ channel/unc-103 and EAG K+ channel/egl-2 are required to moderate some of the effects of the activated unc-43 ( gf ) allele , and all of the effects of the unc-43 ( e408 ) allele ( Figure 3 ) . The pharmacological analyses of genetically activated CaMKII/UNC-43 alleles identified EAG K+ channel/EGL-2 as a molecule that mediates UNC-43 signaling and acts parallel to ERG-like K+ channel/UNC-103 ( Figure 3 ) . However , the results did not reveal when wild-type sex muscles require regulation by UNC-43 and EGL-2 . Indications of when these molecules are used came from our previously reported observations [9] . Under food-deprived conditions , the percentage of ERG-like K+ channel/unc-103 ( 0 ) males displaying constitutive spicule protraction dropped to 9% from the 33% in food . In contrast , under the same dietary-deprived conditions , the percentage of CaMKII/unc-43 ( sy574 ) and unc-43 ( n1186 ) males displaying constitutive spicule protraction showed no statistically significant difference from standard food conditions ( Table 4 ) . We had also previously demonstrated that perturbing pharyngeal pumping with a missense allele of tropomyosin/lev-ll ( rg1 ) can suppress unc-103 ( 0 ) –induced spicule protraction via activity from the pharyngeal neurosecretory motor ( NSM ) neuron . Consistent with dietary deprivation , the lev-ll ( rg1 ) allele also did not reduce the penetrance of the unc-43 ( sy574 ) phenotype [9] . This suggests CaMKII might be required to mediate the effects of dietary deprivation . To test if food deprivation requires CaMKII/UNC-43 to suppress ERG-like K+ channel/unc-103 ( 0 ) –induced spicule protraction , we generated double-mutant combinations between unc-103 ( 0 ) and the unc-43 alleles sy574 and e408 . We tested both alleles because while on food they affected spicule protraction in opposite directions . In food conditions , we found that , like unc-43 ( sy574 ) , the unc-43 ( e408 ) allele increased the penetrance of unc-103–induced protraction . This was not surprising given that the genetic and pharmacology data suggested the unc-43 ( e408 ) –encoded kinase has reduced functions , but , in the spicule protraction circuit , the mutant kinase upregulates UNC-103 and EAG K+ channel/EGL-2 activity ( Figure 3 ) . Since activated EGL-2 has no effect on the unc-103 ( 0 ) phenotype ( Table 1 ) , we hypothesize that the lf unc-43 ( e408 ) allele acts similar to unc-43 ( sy574 ) and synthetically interacts with unc-103 ( 0 ) to increase the incidence of constitutive spicule protraction ( Figures 3 and 4 ) . In contrast to food conditions , food-depriving unc-103 ( 0 ) ; unc-43 ( sy574 ) males reduced constitutive spicule protraction from 88% to 62% ( a percentage similarly displayed by unc-43 ( sy574 ) single mutants ) , whereas depriving unc-103 ( 0 ) ; unc-43 ( e408 ) males of food did not change the phenotype . This suggests CaMKII/UNC-43 is required to suppress sex-muscle excitability under food-deprived conditions . In regards to unc-103 ( 0 ) –induced muscle excitability , the sy574-encoded kinase can suppress the unc-103 ( 0 ) phenotype . However , under food-deprived conditions , it is unable to suppress its own induced constitutive protraction defect . In contrast , the e408 mutation can disrupt UNC-43′s ability to transduce food deprivation signals ( Figure 3 ) . To determine if CaMKII/UNC-43 suppresses sex-muscle contraction during food deprivation in ERG-like K+ channel/unc-103–expressing cells , we expressed CaMKII/unc-43 from the Punc-103E promoter in the sex muscles and a few head neurons of unc-103 ( 0 ) ; unc-43 ( e408 ) males . We found that rescuing unc-43 in these cells restored food deprivation–induced suppression of the unc-103 ( 0 ) phenotype ( Table 4 ) . Since our pharmacological and genetic studies suggested that CaMKII/UNC-43 can act through EAG K+ channel/EGL-2 , we asked if these channels are also required for food deprivation–mediated suppression of the ERG-like K+ channel/unc-103 ( 0 ) phenotype . On food , unc-103 ( 0 ) ; egl-2 ( 0 ) males behaved similarly to unc-103 ( 0 ) single mutants ( Table 4 ) . However , food deprivation was unable to suppress constitutive spicule protraction in unc-103 ( 0 ) ; egl-2 ( 0 ) males . Thus , like UNC-43 , EGL-2 is used during food deprivation to reduce sex-muscle excitability ( Figure 3 ) . To determine if EGL-2 was functioning in UNC-43– and UNC-103–containing cells , we expressed wild-type egl-2 cDNA from the lev-11 and unc-103E promoters in unc-103 ( 0 ) ; egl-2 ( 0 ) males . We found that expressing egl-2 from either promoter restored the ability of food deprivation to suppress unc-103 ( 0 ) –induced spicule protraction . This demonstrates that EAG K+ channel can function in the same cells as CaMKII to compensate for lack of ERG-like K+ channel function under dietary deprivation .
Intense research in organisms ranging from fungi to vertebrates has uncovered beneficial properties of dietary restriction in delaying aging and increasing cellular stress resistance . However , despite the benefits on physiology , reduction in food accessibility tends to promote food foraging , which if successful , will reduce the effects of dietary restriction . In laboratory vertebrates , multiple studies have shown that reduction in diet can lead to short-term increases in locomotor activity [32–36] and in specialized behaviors such as food hoarding [37–39] . In parallel to vertebrate studies , dietary restriction–induced physiological and behavioral changes have been investigated in the nematode C . elegans . The nematode's compact nervous system and musculature provides a simpler and complimentary model to determine how molecules in excitable cells transmit the effects of dietary changes to motor outputs . When food is reduced or even deprived from C . elegans , the nematode displays not only increased longevity and resistance to cellular stresses , but also enhancement in locomotor activity and discrimination between odorants [13 , 40–43] . These behavioral changes presumably facilitate movement toward a food-related odorant source; but , if that source does not result in food satiation , the animal will move towards an alternative source . In addition to upregulating food-acquiring behaviors , food deprivation also depresses behaviors that are not essential and potentially distracting from feeding . In the absence of food , behaviors such as pharyngeal pumping , defecation , egg-laying , and mate searching are reduced [12 , 44–46] . When considering the utility of chronic caloric restriction as a mechanism to promote physical wellness in humans , one must consider how brain circuitries normally used in nonfeeding behaviors might be affected . To address this , we used C . elegans male mating to identify molecules used to attenuate a behavioral state during dietary deprivation ( Figure 4 ) . Under standard conditions , food-satiated C . elegans males display mating behavior upon contact with a hermaphrodite . The male uses sex-specific neurons and muscles located in his tail to recognize a hermaphrodite , scan her body for her vulva , locate and protract his copulatory spicules into the vulva , transfer his genetic material , and withdraw from the hermaphrodite to find additional mates [21] . Deletion of the unc-103–encoded ERG-like K+ channel causes males to display most motor aspects of mating behavior spontaneously; however , since spicule movements provide a facile motor read-out , we have focused on how UNC-103 defects affect the spicule protraction circuit . Prior to mating stimulation , UNC-103 acts in the spicule protractor muscles to keep spontaneous acetylcholine secretion from presynaptic cholinergic neurons ( such as the SPC motor neurons ) from inducing premature contractions . We found that under well-fed conditions , defective channels cause ∼30%–40% of males to protract their spicules constitutively [7] . Under food-deprived conditions , the percentage of males displaying precocious spicule protraction drops to ∼9%–11% [9] . This suggests that additional mechanisms act in parallel with UNC-103 to attenuate cell excitability , and in food-limiting conditions , certain aspects of those parallel mechanisms are upregulated . We isolated the sy574 allele of CaMKII/unc-43 gene in the same genetic screen that identified ERG-like K+ channel/UNC-103 as a regulator of mating behavior [7] . This unc-43 allele causes ∼50% of males to display constitutive spicule protraction; other behaviors in both sexes appear normal . In conjunction with a deletion of unc-103 [23] , 88%–97% of double-mutant males display mating-independent spicule protraction . Similar to the unc-103 mutant phenotype , defective unc-43–induced spontaneous spicule muscle contractions require neurotransmitter secretion from upstream cholinergic neurons such as the SPC neurons ( unpublished data ) . This strongly suggests that CaMKII participates in a regulatory mechanism parallel to ERG-like K+ channel activity . CaMKII has a broad expression pattern in neurons and muscles , and has been demonstrated to regulate multiple general behaviors in both sexes of C . elegans [16] . Unlike the nonsense allele unc-43 ( n1186 ) that causes seizures in both sexes and induces 100% spicule protraction in males , the sy574 mutant kinase must be grossly functional in many behavioral circuits , but has reduced activity for regulating sex-muscle contraction . The sy574 lesion induces two amino acid changes that map to the kinase domain and the self-association domain of UNC-43 . However , the change in the self-association domain might not be relevant; a truncated unc-43 transgene lacking the self-association domain is sufficient to rescue the sy574 spicule defect . The G170E change in the kinase domain of the sy574 allele is located in an amino acid residue that is conserved in CaMKII of diverse species . The structure of the UNC-43 kinase domain has been solved [47] , and the G170E change lies in a structural region of the kinase domain that has not been intensely dissected and characterized , but could be involved in substrate recognition . The differences in phenotypic severity between the nonsense unc-43 ( n1186 ) and missense unc-43 ( sy574 ) alleles suggest that the nonpolar-to-acidic amino acid substitution at position 170 might disrupt the efficiency of how the UNC-43 ( sy574 ) kinase phosphorylates certain effectors in the spicule protractor muscles . CaMKII/UNC-43 has been previously shown to upregulate the activity of ERG-like K+ channel/UNC-103 in the defecation circuit [16 , 23] . We show through pharmacology and genetics that this type of regulation also occurs in the spicule protraction circuit , but we do not believe this regulation is abrogated by the sy574 allele . Although unc-43 ( sy574 ) males superficially display the same behavioral defects as unc-103 ( 0 ) males , unc-43 ( sy574 ) –induced spicule protraction can not be suppressed by food deprivation . A circuit that acts in suppressing unc-103 ( 0 ) –induced sex-muscle seizures incorporates the activity of the pharyngeal muscles and the pharyngeal NSM neurons . In standard food conditions , a missense mutation in tropomyosin/lev-11 can phenocopy the effects of food deprivation , and partially suppress the unc-103 ( 0 ) phenotype . The mutant tropomyosin reduces pharyngeal pumping , but the feeding defect is masked by activity from the pharyngeal NSM neurosecretory neurons . When the NSM neurons in lev-11 mutants are laser ablated , pharyngeal pumping rate decreases , and concurrently , unc-103 ( 0 ) –induced sex-muscle seizures are no longer suppressed . The NSM neurons are hypothesized to sense changes in pharyngeal pumping rate and secrete a factor ( s ) that attenuates muscle excitability , thus suppressing the effects of defective UNC-103 function . Interestingly , similar to depriving males of food , the lev-11 mutation , which can suppress unc-103 ( 0 ) , has no effect on unc-43 ( sy574 ) –induced spicule protraction [9] . Since the CaMKII/unc-43 ( sy574 ) phenotype is not suppressed by food deprivation or pharyngeal muscle–NSM neuron interaction , we suspected activated wild-type UNC-43 might be required to attenuate cell excitability during nonoptimal food conditions . The strong lf unc-43 ( e408 ) allele was used to test if CaMKII is required during food deprivation . The e408 mutation encodes a serine to leucine change at amino acid 179 nine residues from the glycine that is changed in the unc-43 ( sy574 ) allele . The unc-43 ( e408 ) allele induces pleiotropic behavioral abnormalities such as seizures and inhibited egg-laying , suggesting that UNC-43 ( e408 ) kinase activity is severely reduced . In contrast to the nonsense and sy574 mutations , under standard food conditions , the S179L change causes the UNC-43 ( e408 ) kinase to decrease the excitability of male sex muscles via EAG family K+ channels ( Figure 4B ) . The S179L change might reduce effector interactions so that the kinase cannot phosphorylate many of it substrates , except for those that directly or indirectly regulate K+ channels in the sex muscles . Thus , the unc-43 ( e408 ) allele disrupts kinase function in a manner different from the unc-43 ( n1186 ) nonsense allele , which has been shown to severely reduce the amount of CaMKII present in hermaphrodites [16] . Under food deprivation conditions , males containing the UNC-43 ( e408 ) kinase are not able to significantly suppress unc-103 ( 0 ) –induced muscle seizures . The inability of unc-43 ( e408 ) to reduce unc-103 ( 0 ) –induced muscle seizures suggests that the S179L substitution , which reduces CaMKII function in attenuating excitability of many behaviors , also disrupts how food deprivation attenuates the excitability of male muscles . Due to the broad effects of food deprivation , we hypothesize that multiple pathways are used to suppress sex-muscle excitability in response to low food conditions . Since neither unc-103 ( 0 ) ; unc-43 ( e408 ) double-mutant or unc-43 ( n1186 ) single-mutant males are significantly suppressed by food deprivation , it is possible that the majority of these pathways require CaMKII . However , since there is a slight reduction in each case , as well as a significant reduction in unc-103 ( 0 ) ; unc-43 ( sy574 ) males , there might be other pathways working in parallel that are CaMKII independent . Taken together , these data suggests that wild-type CaMKII is required to decrease cell excitability during dietary deprivation conditions , and the mechanism of regulation differs from conditions where food is available . We propose that calcium influx activates CaMKII and EAG K+ channels to depress cell excitability , which can compensate for ERG-like K+ channel dysfunction . In C . elegans , the ortholog of EAG is encoded by the gene egl-2 [30] . Synaptic plasticity studies in Drosophila have shown that neuronal CaMKII directly phosphorylates EAG K+ channels . This phosphorylation is presumed to maintain or enhance EAG K+ channel activity during calcium influx and after intracellular calcium has dropped to basal levels [14 , 15 , 31] . In C . elegans , EGL-2 is coexpressed with UNC-103 and CaMKII/UNC-43 in the spicule muscles . Under standard food conditions , the egl-2 ( 0 ) mutation induces no obvious abnormal defect either by itself or in conjunction with unc-103 ( 0 ) in regulating spicule muscle contraction . Similarly , the gain-of-function egl-2 ( n693 ) mutation , which hyperpolarizes sensory neurons , egg-laying , and defecation muscles , also does not enhance or suppress unc-103 ( 0 ) –induced seizures . Taken together , the genetic data suggest that under standard food conditions , EGL-2 K+ channels , although present in the spicule muscles , are not active . However , similar to males containing the severe lf unc-43 ( e408 ) mutation , deletion of egl-2 reduces the ability of food deprivation to suppress unc-103 ( 0 ) –induced muscle seizures . Our data suggest that under food deprivation conditions , calcium influx activates CaMKII and EAG K+ channels to depress cell excitability , which can compensate for ERG-like K+ channel dysfunction . We speculate that in well-fed wild-type males , ERG-like K+ channels act in parallel with a CaMKII-regulated pathway ( s ) to attenuate mating behavior until specific cues are presented ( Figure 4A ) . If one of these mechanisms is disrupted , the other can compensate in a percentage of animals ( Figure 4C ) . In the absence of food , CaMKII and EAG K+ channels additionally act to hyperpolarize the sex muscles , thus suppressing mating behavior so that the male will forage for food , rather than mate ( Figure 4D ) . We have demonstrated that one can take advantage of this dietary deprivation pathway to suppress spontaneous muscle seizures caused by defective ERG-like K+ channels . We suspect that similar mechanisms that occur during food deprivation in C . elegans males also function in neurons and muscles in higher animals . This is consistent with many studies that have shown that caloric restriction can reduce seizure susceptibility in epileptic mouse models [48–50] . Although the exact molecules will differ between neuronal types in various excitable cell circuitries , co-opting endogenous parallel regulatory pathways to compensate for ion channel dysfunction should be a feasible general strategy .
All strains contain him-5 ( e1490 ) ( LGV ) [51] and were maintained as described in [52] . The following strains were used . LGI: lev-11 ( rg1 ) [9]; LGIII: unc-103 ( n1213 ) [53] and pha-1 ( e2123 ) [54]; LGIV: unc-43 ( e408 ) and unc-43 ( e266 ) [52] , unc-43 ( n1186 ) , unc-43 ( n1179 ) , unc-43 ( n498 ) [53] , unc-43 ( sa200 ) [45] , and egl-19 ( n582 ) [55]; and LGV: unc-68 ( r1158 ) [28]; and egl-2 ( n693 ) [56] . The strain CB4856 was used for single nucleotide polymorphism mapping [17] . The isolation of sy574 has been previously described [7] . sy574 animals were out-crossed five times . sy574 contains two missense changes: sy574A changes the sequence CACGGATTT to CACGAATTT , and sy574B changes GCCGCGTGT to GCCGTGTGT . unc-43 was also sequenced in the strain used for mutagenesis , PS1385 , and no mutations were found . In unc-43 ( e408 ) , the e408 allele changes the sequence TTGTCGCCA to TTGTTGCCA . egl-2 ( rg4 ) was generated by trimethylpsoralen mutagenesis of egl-2 ( n693gf ) him-5 [57] . After mutagenesis , we selected worms that displayed normal egg-laying behavior . PCR analysis using primers that annealed to internal exons of egl-2 were used to screen lines for deletions in the egl-2 locus . egl-2 ( rg4 ) includes exons 1–7 , but does not include the pore region or the egl-2 ( 693gf ) mutation [30] . The egl-2 ( rg4 ) deletion ends before the start of the next gene pme-5 . egl-2 ( rg4 ) animals were out-crossed four times . L4 males were separated and allowed to develop into adults overnight on NGM plates seeded with OP50 . Generally , 20–30 worms were analyzed per plate . The adult males were scored as positive for displaying spontaneous spicule protraction if at least one spicule protruded from the cloaca . L4 males were isolated and allowed to mature overnight on NGM plates . The next day , males that did not display constitutive spicule protraction were placed , five at a time , in solutions of ARE ( Indofine Chemical Company , http://www . indofinechemical . com ) in Pyrex , round-bottom , three-well titer plates . The males were observed for 5 min and scored if their spicules remained protracted for at least 10 s . Curve fits and EC90 values were determined using GraphPad Prism ( version 4 . 03; http://www . graphpad . com ) . We used 10-mm NGM plates with an 8-M glycerol ring surrounding the outer edge of the plate . The 8-M glycerol ring served as a repellent during the assay to keep males from crawling up the inside edge of the plate; the 8-M glycerol ring induced no other significant changes in male behavior ( unpublished data ) . To measure the effects of food deprivation , we separated late L4–stage males to a clean NGM plate with no Escherischia coli OP50 and allowed them to crawl away from any OP50 transferred . Males were then picked up by mouth pipette , washed with M9 buffer , and transferred to a clean NGM plate with the glycerol ring described above . As a control , sibling males of the same stage were placed on 10-mm NGM plates seeded with OP50 . Males from both the control and experimental plates were then assayed 15–20 h later for the constitutive protraction phenotype . The details of the construction of the plasmids used in this study are listed in Text S1 and Table S1 . Plasmids containing unc-103 genomic DNA were created as previously described [23] . The plasmids pBL58 , pBL70 , pBL69 , pBL71 , pBL72 , and pBL80 contain the unc-43g cDNA expressed from the hsp-16 , aex-3 , lev-11 , unc-103E , tnt-4 , and acr-8 promoters , respectively . The plasmids pTG44 and pTG46 contain the egl-2 cDNA expressed from the unc-103E and lev-11 promoters , respectively . pBL68 contains unc-43 cDNA lacking the self-association domain expressed from the lev-11 promoter . pBL66 contains CFP expressed from the gtl-1 promoter . pLR16 contains yellow fluorescent protein ( YFP ) expressed from the egl-2 promoter . Worms expressing unc-103 transgenic lines were previously described [23] . L4 males were isolated and scored the next day to determine the number of males that displayed protruding spicules . To obtain unc-43 transgenic lines , DNA was injected into unc-43 ( sy574 ) ; him-5 ( e1490 ) hermaphrodites out-crossed three times using standard protocols [58] . The injection mixtures were created as follows: the injected concentrations of the unc-43 plasmids , pBL69 ( 50 ng/μl ) , pBL70 ( 50 ng/μl ) , pBL72 ( 50 ng/μl ) , pBL58 ( 26 ng/μl ) , and pBL71 ( 10 ng/μl ) were combined with the appropriate amount of pUC18 to bring the total concentration of DNA to 180 ng/μl . In addition , each mixture contained 20 ng/μl of pBL66 that was used as a marker to identify transgenic lines . pBL68 was injected along with pUC18 ( 92 ng/μl ) and the pha-1–rescuing plasmid pBX1 ( 100 ng/μl ) [59] into pha-1 ( e2131 ) ; unc-43 ( sy574 ) ; him-5 ( e1490 ) at a concentration of 23 . 6 ng/μl . F1 hermaphrodite progeny with CFP expression in their intestines were selected . For each injection , three to five lines were analyzed; one representative line is shown in the tables . Males from the transgenic lines were scored for spontaneous spicule protraction in the following manner: 6 L4 hermaphrodites for each line were placed on individual NGM + OP50 plates . Only first-generation L4 males of the isolated hermaphrodites were picked , allowed to mature to adults overnight at 20 °C , and then scored for constitutive spicule protraction . For males containing UNC-43 cDNA expressed from the heat-shock promoter , late-stage L4 males were heat-shocked at 33 °C for . 5 h and then scored for the instance of spicule protraction the next day . | We investigated the mechanisms used during dietary stress conditions that regulate the behavioral output of excitable cells . In the roundworm Caenorhabditis elegans , males must display the proper behavioral response to potential food sources or mating partners . This regulation is disrupted by loss of ERG-like K+ channel function . K+ channel defects cause a percentage of males to contract their genitalia muscles permanently , even in the absence of mating cues . For the percentage of K+ channel–defective males that do not display spontaneous sex-muscle seizure , we show that abnormal muscle contraction is attenuated by the calcium/calmodulin-dependent protein kinase II ( CaMKII ) . Under food-deprived conditions , we find that CaMKII acts with the ether-a-go-go K+ channel to further suppress spontaneous sex-muscle seizures . We speculate that in food-abundant conditions , ether-a-go-go–related gene–like K+ channels act in parallel with a CaMKII-regulated pathway ( s ) to attenuate sexual behavior until mating signals are presented . If one of these mechanisms is disrupted , the other can compensate . In the absence of food , CaMKII and ether-a-go-go K+ channels further hyperpolarize the genital neuromuscular circuitry , changing the behavioral state so that the male will forage for food rather than mate . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"caenorhabditis",
"neuroscience"
] | 2007 | Food Deprivation Attenuates Seizures through CaMKII and EAG K+ Channels |
The ‘communication through coherence’ ( CTC ) hypothesis proposes that selective communication among neural networks is achieved by coherence between firing rate oscillation in a sending region and gain modulation in a receiving region . Although this hypothesis has stimulated extensive work , it remains unclear whether the mechanism can in principle allow reliable and selective information transfer . Here we use a simple mathematical model to investigate how accurately coherent gain modulation can filter a population-coded target signal from task-irrelevant distracting inputs . We show that selective communication can indeed be achieved , although the structure of oscillatory activity in the target and distracting networks must satisfy certain previously unrecognized constraints . Firstly , the target input must be differentiated from distractors by the amplitude , phase or frequency of its oscillatory modulation . When distracting inputs oscillate incoherently in the same frequency band as the target , communication accuracy is severely degraded because of varying overlap between the firing rate oscillations of distracting inputs and the gain modulation in the receiving region . Secondly , the oscillatory modulation of the target input must be strong in order to achieve a high signal-to-noise ratio relative to stochastic spiking of individual neurons . Thus , whilst providing a quantitative demonstration of the power of coherent oscillatory gain modulation to flexibly control information flow , our results identify constraints imposed by the need to avoid interference between signals , and reveal a likely organizing principle for the structure of neural oscillations in the brain .
Task-dependent changes in the power and inter-region coherence of oscillatory network activity are observed in many brain regions and behavioral tasks [1]–[3] . A possible function of such activity is to modulate functional connectivity among anatomically connected regions [4]–[6] . This may play an important role in cognition by allowing the structure , and hence function , of brain networks to be dynamically reconfigured in response to different task demands . The ‘communication through coherence’ ( CTC ) hypothesis [5] , [7] proposes that selective communication is achieved through coherence between firing rate oscillation in the sending region and oscillatory gain modulation in the receiving region . This could , theoretically , allow a network to respond selectively to a task-relevant ‘target’ signal while ignoring other distracting inputs . However , the conditions under which accurate selective communication can be achieved by this mechanism remain unclear . Intuitively , it appears likely that the accuracy with which a target signal can be filtered from distractors will depend on how they differ with respect to the oscillatory modulations of their firing rates . Clearly , if target and distracting inputs have the same modulation , coherent gain modulation cannot separate them; but in what way and to what extent must their modulations differ in order for the target signal to be accurately recovered ? Understanding which structures of oscillatory activity can support accurate selective signal transmission is an important step in evaluating whether activity patterns observed in vivo are consistent with their proposed functional role in routing information flow . Despite extensive experimental [8]–[10] and computational work [11]–[14] , it remains unclear under what conditions the CTC mechanism could allow a network to distinguish among converging population-coded signals , and how its performance depends on the structure of their oscillatory modulations . We recently developed a convergent pathway model to investigate oscillatory routing of information flow [15] . In the present study we have used a similar paradigm to address these questions . Our results indicate that where inputs are distinguished by the frequency , phase or amplitude of their modulations they can be readily separated by coherent gain modulation , but that attempting to separate inputs that oscillate incoherently in the same frequency band results in greatly increased noise and reduced communication accuracy . Additionally , the oscillatory modulation of the target input must be strong to ensure a high signal to noise ratio relative to stochastic spiking of individual neurons . These constraints on patterns of activity that efficiently support flexible routing of information may be an organizing principle for the rich structures of neural oscillations observed in vivo .
We modeled a convergent pathway in which multiple input networks converged to a single receiving network . The task required of the model was to selectively route a behaviorally relevant signal encoded in an input network ( the ‘target’ input ) to the receiving network , while ignoring simultaneously active distracting inputs in other converging networks . While not specifically a model of any particular region , this network design minimally recapitulates many converging cortical and sub-cortical pathways where selective information flow may be required . Input selectivity was achieved by oscillatory gain modulation in the receiving network , coherent with oscillatory modulation of the firing rate of the target input network . To evaluate how accurately this gain modulation could filter the target input from distractors , the output of the receiving network was integrated over time and decoded to produce an estimate of the stimulus encoded by the target input . Each input network modeled a local population of neurons representing a separate one-dimensional circular variable using a firing rate population code ( e . g . a cortical hypercolumn ) . The average firing rates of individual neurons were given by bell-shaped tuning curves with respect to the stimulus orientation . Spike times in each neuron were determined by a Poisson process whose instantaneous rate could be simultaneously modulated for all neurons within a given network to simulate a population oscillation . This modulation was modeled as a Von Mises function of the phase of the oscillation , characterized by a modulation strength and frequency . As network oscillations in vivo are irregular in frequency and amplitude , we allowed the instantaneous strength and frequency to fluctuate around their mean values . These fluctuations were modeled as low-pass filtered Gaussian white noise . The resulting activity was consistent with in vivo data showing irregular spiking of single units [16]–[18] during sparsely synchronized oscillatory activity [19] . We initially considered a situation in which an external control input synchronized activity in the target and receiving networks , such that the oscillatory modulation of the target input firing rate was known by the receiving network . The receiving network must exploit this known temporal structure to generate a pattern of gain modulation that separates target from distracting inputs , and hence recover the spatial population code representing the target stimulus . We later compare the performance of models with and without such an external synchronizing input . Physiologically , gain modulation could be achieved by local interneuron circuitry modulating the distribution of membrane potentials , degree of shunting inhibition [20] , or synaptic noise [21] experienced by the principal neurons of the receiving network . Oscillatory gain modulation coherent with the target input could then be generated by driving such interneuron circuitry with the oscillating external control signal . Although this arrangement could , in principle , be implemented in a biophysical model , we required a model of the receiving network that could be optimized to generate the temporal pattern of gain modulation that best separated the target from distracting inputs . If the receiving network was not optimized , the results would be uninformative about the performance of the mechanism in general and would only shed light on the specific implementation . Two obstacles made optimizing a biophysical model intractable . Firstly , we do not know a priori what waveform the optimal gain modulation should take for a given temporal pattern of input activity . Secondly , it remains incompletely understood how neuronal and network parameters determine the response of networks to temporally structured inputs . Therefore , even if we knew what temporal pattern of activity the interneurons in the receiving network must generate in response to a given control signal , it would not be straightforward to design such a network with the appropriate dynamics . Given these difficulties with optimizing a biophysical model we instead developed an algorithmic description of the receiving network's operation that could be optimized with respect to the mean squared error of the target stimulus estimates decoded from its output . The receiving network model consisted of two components . The first component was a layer of projection units , which received convergent population-coded signals from the input networks and formed the output of the receiving network . Each unit in the projection layer represented a population of cells innervated by neurons with similar orientation preferences in each input network and whose output was an analog firing rate signal . The output of unit was given by:where was the spike input received by the unit and was a temporal pattern of gain modulation ( see below ) . We allowed the gain to take both positive and negative values ( corresponding to net excitatory and inhibitory output respectively ) such that spikes arriving during periods of negative gain contributed negatively to the integrated output . Though we do not explicitly model the circuitry , a simple micro-circuit supporting positive and negative net gain would be a pathway where excitation is balanced by feed forward inhibition , with gain modulation acting on the inhibitory neurons . The second component of the receiving network represented the local interneuronal circuitry , which received the oscillating top-down control signal and converted this into a temporal pattern of gain modulation , which was applied uniformly to all projection units . In an optimized filtering network the dynamics of this circuitry must be such that the pattern of gain modulation generated in response to a given control input is that which best filters the target from distracting inputs . Rather than model these dynamics directly we instead represented them as a filtering process . Optimizing the receiving network then became a problem of finding the filter that transformed the firing rate modulation of the target into the gain modulation that best separated the target from distracting inputs . This problem is closely analogous to that of matched filtering in the engineering literature in which a target signal of known waveform must be detected against a background of noise . We initially considered gain modulations that were linearly filtered versions of the firing rate modulation of the target input . For this model , we could rapidly optimize the frequency response of the filter using gradient descent on training data ( see Materials and Methods ) , allowing us to explore the parameter space of input activity patterns . We then verified that our key results obtained for this linear model held when we allowed the gain modulation to be an arbitrary function of the firing rate modulation of the target input . The output of the receiving network was integrated over 100 ms to give a spatial pattern of activity . This was then decoded to produce an estimate of the target stimulus . We therefore only considered the information contained in the average firing rates of the receiving network output units over the integration window . We report the lower bound on the Fisher information given by the reciprocal of the mean squared error of the stimulus estimates . Locally optimal linear estimators ( LOLEs ) were used for decoding . These decoders were sufficiently simple to permit optimization of the temporal filtering with respect to the root mean squared error of decoded stimulus estimates using gradient descent ( see Experimental Procedures ) . Under many noise distributions these decoders perform close to optimally , as indicated by the minimal difference in performance when compared with more sophisticated non-linear methods [22] , [23] . These decoders are , moreover , biologically plausible as their performance corresponds to that of de-noising by networks implementing line attractor dynamics [24] , [25] . input networks , each consisting of 10 , 000 Poisson neurons , represented independent orientation variables . The firing rate of the th neuron in input network was given by:Where is a firing rate tuning curve with respect to stimulus orientation ( range 0–180° ) and is an oscillatory firing rate modulation:Where is the firing rate of the th neuron , is the neurons preferred orientation and is the average firing rate across the population . The oscillatory modulation was a Von Mises function of the oscillation phase :Where is a concentration parameter that determines how tightly synchronized the activity is , and is the modified Bessel function of order 0 which normalizes the modulation such that its average value over time is 1 . A sinusoidal modulation was also used where indicated , given by the equation:To model the irregularity of network oscillations we allowed the oscillation angular frequency and concentration parameter to fluctuate around their mean values and . These fluctuations were modeled as:where , were low-pass filtered Gaussian white noise with amplitude 1 and a cut-off frequency of . The variability of the frequency and amplitude respectively were therefore determined by and , which were the standard deviation of fluctuations divided by the mean value . Where we report synchronization strength , we use the following measure:Where is the modified Bessel function of order . This measure is 1 if all spikes occur at the same phase and 0 if the firing rate is equal at all phases . The combined input spike rate impinging on the receiving network was the sum of activity in all input networks:The combined spike input was:The receiving network consisted of a layer of 8 units , each of which received spike input from neurons with similar orientation preference in each input network . The range of orientation preference from 0° to 180° was divided into 8 equal width bands and neurons in each input network with orientation preference in a given band projected to the same unit in the receiving network . The combined input to unit in the receiving network was:Where the sum was over those units in the input networks that projected to unit in the receiving network . The output of unit in the receiving network was: was integrated over time to produce a spatial pattern of activity :An estimate of the stimulus encoded in the target input network was decoded from the integrated activity using a LOLE . where are the weights for each unit of the receiving network and is a constant . The simulations were performed in discrete time with a resolution of 1 ms , such that the integral over time was computed as a sum over time bins:where is the spike count received by unit in time bin . Except where specified otherwise , the gain modulation was a linearly filtered version of the modulation of the target input . For this model we optimized the receiving network by using gradient descent to find the frequency response of the filter ( the gain and phase shift as a function of frequency ) which optimized decoding accuracy with respect to the variance of the stimulus estimates . We used Plancherel's theorem to express in terms of the discrete Fourier transforms ( DFTs ) and of the gain modulation and spike input . ( indicates complex conjugate , is the number of components in the Fourier transforms ) . As the gain modulation and spike input are both real valued , the imaginary parts of at positive and negative frequencies cancel and only the real parts contribute to the sum . Also , so the second half of the sum is redundant:where for , for . We expressed the Fourier transform of the gain modulation as the product of the frequency response of the filter and the Fourier transform of the target inputs firing rate modulation . The frequency response is a complex valued function of frequency where is the gain of the filter at frequency k and is the phase shift . We can then express the stimulus estimate as:We define This can be written in vector notation as:Where is a row vector whose components are the weights of the LOLE , and are matrices whose components are and , and and are column vectors whose components are respectively the real and imaginary parts of the filter frequency response and . To further simplify the expression we concatenate the vectors and to make a single vector containing both the real and imaginary parts of the filter frequency response and concatenate the matrices and to make a single matrix . We can now express the decoded stimulus estimate as:To use gradient descent to find the optimal LOLE weights and filter frequency response we define a cost function and calculate the gradient with respect to it . We define the cost function as the squared error between the true stimulus value and the decoded estimate , averaged over a training set of data:Where indicates the average over the training set . The following default parameter values were used except where stated otherwise: Average neuronal firing rate = 5 Hz , Synchronization strength = 0 . 5 , Modulation frequency = 50 Hz , Amplitude variability parameter = 0 . 1 . The frequency variability parameter was set to 0 . 1 for narrowband oscillations and 0 . 3 for broadband oscillations . Narrowband oscillations were used except those where distractors oscillated incoherently in the same frequency band as the target . The following procedure was performed to establish decoding accuracy for each set of input parameters . All simulations were performed in MATLAB . A training set and a test set of input activity were generated , each consisting of 5000 samples of 100 ms each . In each set , half of the samples had target stimulus orientation and the other half . The separation was chosen iteratively such that 75–80% of samples were correctly classified from the decoded stimulus estimates . The orientation of stimuli encoded in the distracting input networks were uniformly randomly distributed in all samples . To reduce spectral leakage due to finite integration times , we applied a Hann window to the spike activity in each sample . The weight vectors for the LOLE and the frequency response of the filter were optimized using a two-stage gradient descent procedure . Firstly we used gradient descent to find the filter frequency response that minimized the mean squared error between the output of the units comprising the receiving network and the firing rate each unit received from neurons in the target input network . This gradient descent stage was initialized with all components of the filter frequency response set to zero . We then performed gradient descent simultaneously on the LOLE weights and the filter frequency response to minimize the mean squared error of target stimulus estimates , using the gradients calculated above . This second stage of the gradient descent was initialized with the filter frequency response found in the first gradient descent and with the weights of the LOLE set to zero . We used this two stage procedure because it converged much more rapidly than initializing the simultaneous gradient descent for & with small random weights . To prevent over-fitting we evaluated the mean squared error for the test set and halted gradient descent when this started to rise . We evaluated the mean and variance of the stimulus estimates for both orientations on the test set:The lower bound on the Fisher information was given by:When examining decoding accuracy for integration times up to 1 , 000 ms , the gradient descent took a very long time because of the larger number of weights to be fitted . We had observed in other simulations that the amplitude of the filter frequency response was consistently zero at high frequencies where there was minimal power in the target modulation . For all simulations in this figure we therefore set the filter frequency response to zero for frequencies above 3 times the target modulation frequency , reducing the number of weights that had to be fitted . We verified for a subset of simulations that this minimally affected decoding accuracy . We also allowed the gain modulation to be an arbitrary function of the modulation of the target input . To do this we generated training and test sets as described above , but instead of generating a different modulation of the target input network for each sample , we used the same modulation of the target input network while generating different modulations for the distractors . Instead of optimizing the filter parameters that transformed the modulation of the target input into the gain modulation , we directly fitted the gain modulation using gradient descent . To do the gradient descent we rewrite the equation for the stimulus estimate in vector form:Where is a row vector with components , is a matrix with components and is a column vector with components . As this has identical form as , we can use the gradient descent procedure described above to find the gain modulation that minimizes the mean squared error of target stimulus estimates . Decoding accuracy varies somewhat depending on the precise waveform of the modulation of the target , so we repeated the procedure 100 times using different instances of the gain modulations of the target input , and report the mean and standard deviation of the decoding accuracy over these different modulations .
Figure 1 illustrates the power of coherent gain modulation to filter an oscillating population-coded target signal from distractors , and hence achieve selective communication . Though four different inputs of equal average firing rate converge on the receiving network , the integrated output reflects only the spatial pattern of activity in the target input , and the stimulus encoded by this input can be accurately decoded from the output . How does this selection occur ? The effective gain for each converging pathway is determined by the overlap between the input's firing rate modulation and the gain modulation in the receiving region , averaged over the integration window . In this case the gain modulation in the receiving region is approximately sinusoidal and in phase with the target input ( 2nd input network from the top in Figure 1 ) . The target input contributes strongly to the integrated output because periods of high firing rate occur concurrently with large positive gain . Because the average gain is zero , the distractor whose units fire asynchronously , without any population firing rate modulation ( first input network ) , contributes minimally to the integrated output . Likewise , for distractors oscillating at frequencies well separated from the target ( 3rd and 4th networks ) the average overlap between firing rate modulation and the gain modulation is very close to zero , and hence they also contribute minimally to the integrated output . Mathematically , the signal from distracting inputs is rejected because their firing rate modulations are either zero or orthogonal to the gain modulation in the receiving region . For accurate selective communication to be achieved by this mechanism , there must exist a pattern of gain modulation that is strongly driven by the modulation of the target input but close to orthogonal to the modulations of the distracting inputs . As we will see in the next section , this imposes constraints on the structure of oscillatory activity in the converging inputs . We evaluated the accuracy of selective communication for four different structures of oscillatory activity in the input pathways ( Figure 2A–D ) . We first considered a condition in which only the target input was oscillating while the units in the distractor networks fired asynchronously . The gain modulation produced by the optimized filtering network was near-sinusoidal , in phase with the oscillation in the target input ( Figure 2A ) . Decoding accuracy depended strongly on the strength of oscillatory modulation of the target input ( Figure 2F ) . Accuracy was high for strongly modulated input , dropping steeply as the oscillation strength decreased . We quantified the depth of modulation of the target input firing rate ( ‘synchronization’ ) using a metric that ranged from 0 for fully asynchronous activity to 1 if all spikes occur at the same phase of the modulation ( see Materials and Methods , Figure 2E ) . Over the range of synchronization from 0 . 1 to 0 . 9 , Fisher information increased by a factor of 95 . 7 , with the majority of this increase occurring in the range from weak to moderate synchronization ( Fisher information increased 26-fold when synchronization strength increased from 0 . 1 to 0 . 5 , and 3 . 65-fold as it increased from 0 . 5 to 0 . 9 ) . Weak target input modulation resulted in poor decoding accuracy because the signal read out by the receiving network was small relative to noise from stochastic spiking of distracting inputs . Across a wide range Fisher information increased with the average firing rate in the target and distractors ( Figure 2G ) . We next evaluated the proposal that changes in the inter-region coherence of oscillatory activity [5] , [8] , [10] ( as distinct from changes in frequency or amplitude , or changes in a consistent phase relationship ) could be used to switch on or off information propagation through a convergent pathway . In our model , this corresponds to distracting inputs oscillating irregularly in the same frequency band as the target ( Figure 2B ) . A corollary of this CTC scheme is ‘non-communication through non-coherence’ [5] whereby absence of a reliable phase relationship between firing rate modulation in the sending target network and gain modulation in the receiving region prevents information transmission . Fisher information was greatly reduced for pathways in which distracting inputs oscillated incoherently compared with pathways in which distracting inputs were asynchronous ( Figure 2F , G ) . The relative performance in the two conditions depended on the firing rate of the input networks . For pathways with asynchronous distractors , information increased linearly , but with incoherently oscillating distractors , information increased sublinearly with the firing rate of the input networks ( Figure 2G ) . For average firing rates of 1 Hz per neuron , the Fisher information was 5 . 7 times higher when distractors fired asynchronously than when they oscillated incoherently , and this ratio increased to 27 . 8 when the average firing rate was 10 Hz . Increasing the synchronization of oscillations in the input networks improved decoding accuracy ( Figure 2F ) , but across all oscillation strengths accuracy was much higher for pathways with asynchronous distractors . Two aspects of the input network activity were changed between the asynchronous and incoherent distractors conditions; the modulation of the distracting inputs but also the variability of the frequency of the target input which was narrow in the asynchronous distractors case but broad in the incoherent distractors case . To determine which of these changes degraded communication accuracy we evaluated a condition with broadband incoherent distracting inputs but narrowband modulation of the target input ( Figure S1A ) , and a condition with asynchronous distractors but broadband modulation of the target input ( Figure S1B ) . Changing from a narrowband to a broadband modulation of the target signal did not affect communication accuracy with either asynchronous or incoherent distractors ( Figure S1C ) . Changing from asynchronous to incoherent distractors dramatically reduced communication accuracy for both narrowband and broadband modulation of the target input ( Figure S1C ) . These results indicate that it is the incoherently oscillating distracting inputs that degrade communication accuracy . The poor communication accuracy in the incoherent distractors condition can be understood by considering the overlap between the gain modulation in the receiving network and the firing rate modulation of the distractors . From cycle to cycle the distracting inputs will drift in and out of phase with the target input , and hence with the gain modulation in the receiving region . This causes large fluctuations in the effective gain for distracting inputs , a source of ‘overlap’ noise quite distinct from that due to stochastic spiking of individual neurons . Although we have measured the accuracy of signal estimation for a given integration time , an alternative effect of this additional source of noise is an increase in the integration time required to reach a given decoding accuracy when compared with asynchronous distracting inputs , i . e . a decrease in the rate of information transmission through the pathway . The differential dependence of communication accuracy on firing rate in the asynchronous and incoherent distractors conditions can be understood by considering more closely the two sources of noise that degrade the stimulus estimate . Noise due to stochastic spiking of individual neurons occurs for both asynchronous and oscillating distractors , and becomes smaller relative to the signal as the firing rates of the input networks increase . Overlap noise , in contrast , occurs only for oscillating distractors and increases in proportion to the signal size with increasing input firing rates . With incoherently oscillating distractors , this second source of noise becomes dominant as the mean firing rate increases , and prevents a further increase in signal-to-noise ratio . At very low firing rates , noise in the output of the receiving network is dominated by stochastic spiking of individual neurons . In this regime we found decoding accuracy to be comparable for asynchronous and incoherently oscillating distractors . In supporting information we evaluate the firing rate threshold above which overlap noise dominates ( Text S1 and Figure S2 ) . Above this threshold ‘non-communication through non-coherence’ results in severe signal degradation compared with schemes in which distracting inputs are asynchronous or separated from the target in frequency or phase ( see below ) . This threshold is proportional to the oscillation frequency , but for physiological frequencies it is low relative to firing rates relevant for coding in cortex . Gain modulations generated by the receiving network in the incoherent distractors condition were often very different in shape from the firing rate modulation of the target ( Figure 3A , 6B ) . This is because the optimized frequency response of the filter that transformed the firing rate modulation of the target input into the gain modulation strongly emphasized the high and low frequency components of the target modulation ( Figure 3B ) . Because such gain modulation may be biologically implausible , we also evaluated the performance of a receiving network that applied a gain modulation that oscillated around 0 with the same waveform as the firing rate modulation of the target input ( Figure 3A ) . This considerably reduced decoding accuracy , resulting in ∼40% lower Fisher information than the optimized receiving network ( Figure 3C ) . We next tested whether separating distracting inputs from the target in frequency improved performance ( Figure 2C , F–H ) . Because the Von Mises modulations used in the rest of this study contain harmonics which broaden the frequency band occupied by the oscillation , we additionally considered input networks where the firing rate modulation was a sinusoidal function of oscillation phase ( Figure 2H ) . As in other simulations the frequencies of these sinusoidal modulations were allowed to fluctuate from cycle to cycle around their means . Decoding accuracy increased steeply as the average modulation frequency of the distracting inputs was moved to either higher or lower frequencies than that of the target input ( Figure 2H ) . When distracting inputs were well separated in frequency from the target , decoding accuracy was comparable to that for asynchronous distracting inputs , and Fisher information increased linearly with input network firing rate ( Figure 2G ) . For Von Mises modulations accuracy was reduced when the distractors' modulation frequency was half that of the target ( Figure 2F , H ) , as a result of interference between the first harmonic of the distractors' modulation and the fundamental frequency of the target modulation ( see spectra , Figure 2I ) . We compared Fisher information as a function of distractor frequency for narrowband oscillations , in which cycle-to-cycle frequency fluctuations were small , and for broadband oscillations , in which such fluctuations were large . With broadband oscillations the target and distractor inputs had to be more widely separated in frequency to avoid interference ( Figure 2H ) as the modulations occupied a broader frequency band ( Figure 2I ) . Separation in frequency works because sinusoids of different frequency are orthogonal under the overlap integral operation that separates target from distracting inputs . Distracting inputs that are well separated from the target in frequency therefore only contribute noise due to stochastic spiking and not due to fluctuations in the overlap of their firing rate modulation with the gain modulation . We next explored whether separating inputs in phase allowed the target signal to be accurately separated from distractors . Phase coding , in which assemblies of neurons fire at different phases relative to a global oscillation , has been reported in several neural systems [26]–[30] , most notably in the hippocampus where place cells representing past , present and future locations fire at progressively later phases with respect to the theta oscillation . We evaluated how accurately gain modulation could separate a target input from distractors oscillating coherently with it , evenly separated in phase ( Figure 2D ) . For strongly synchronized activity , decoding was highly accurate , with better performance than for asynchronous distractors ( Figure 2F ) . This was because the absolute value of the gain modulation was small except at the phase where the target but not distracting inputs were firing strongly , reducing noise due to stochastic spiking of distracting inputs . Performance dropped rapidly as modulation strength was decreased , such that for weakly and moderately synchronized inputs decoding accuracy was worse than for asynchronous distractors . Unsurprisingly , increasing the separation in phase between the peak firing of target and distractor inputs relaxed the degree of synchronization required to reach a given decoding accuracy ( data not shown ) . We examined how changing the duration over which the output of the receiving network was integrated affected the accuracy of selective communication . In the asynchronous distractors condition , Fisher information increased linearly with integration time as long as the integration time was greater than approximately twice the period of the target input modulation ( Figure 4A ) . Below this threshold integration time decoding accuracy dropped precipitously . For efficient selective communication , the integration time must be sufficiently long relative to the target input oscillation period to ensure that the contribution of the target input to the integrated output is minimally affected by the phase of the target modulation , and to ensure that the contribution of the distracting inputs consistently averages to zero over the integration window . In the incoherent distractors , phase separation , and frequency separation conditions ( Figure 4B–D ) , decoding accuracy also dropped dramatically when the integration time was reduced below approximately two cycles of the target input modulation . In the frequency separation case with target and distractors well separated in frequency , the target but not distractor modulation frequency determined the minimum integration time required . This frequency dependent minimum integration time required to achieve efficient communication has implications for how different signals may be distributed across different frequency bands . Low modulation frequencies require long integration times , and hence are not suitable for encoding signals that vary on a rapid timescale , whereas higher frequency modulations permit shorter integration times and hence can encode signals that vary on a shorter timescale . This suggests a possible principle contributing to the division of labor between different frequency bands of neural oscillations . In the model considered hitherto , an external control input imposed coherence between the firing rate modulation of the target input network and the gain modulation in the receiving network . An alternative approach to generating coherence would be for the interneuronal circuitry generating the gain modulation to entrain directly to the combined input in a ‘bottom-up’ fashion . Specifically , if no distracting inputs oscillate in the same frequency band as the target , a resonant interneuronal circuit with band-pass characteristics may be able to filter the modulation of the target signal from the combined input activity and use this to generate the appropriate gain modulation . To explore this possibility we considered an alternative version of the model ( Figure 5A ) . Rather than receiving an external control input , the circuitry generating the gain modulation received an input that was simply the summed spiking activity of all input networks . As in the original model , this was linearly filtered to produce a temporal pattern of gain modulation that was applied to the receiving network projection neurons . As before , we optimized the filtering such that the resulting gain modulation best separated target from distracting inputs , and evaluated the performance of the network by integrating the output for 100 ms and decoding the resulting spatial pattern to estimate the target stimulus . When distracting inputs were asynchronous ( Figure 5B ) , or well separated from the target in frequency ( Figure 5D ) , the performance of the bottom-up model was comparable with that of the top-down model . However , when distracting inputs oscillated in the same frequency band as the target , either incoherently with it ( Figure 5C ) or at different phases ( Figure 5E ) , the bottom-up model was unable to selectively propagate information about the target input . This was because , with no reference signal to provide information about the phase of target input modulation , there was no way for the receiving network to differentiate between target and distracting inputs . This bottom-up configuration removes the need for a synchronizing input to the receiving network by effectively hard-wiring in a frequency preference , such that it will respond only to inputs modulated at the correct frequency . Implementing flexible communication would however still require some control circuitry , either to manipulate the dynamical state of the input networks such that only the target oscillates in the pass band , or to shift the resonance of the receiving network to read out signals encoded at different frequencies . So far we have presented results for receiving networks which applied a gain modulation that was a linearly filtered version of the firing rate modulation of the target input . We asked if the performance could be improved by allowing the gain modulation to be an arbitrary function of the modulation of the target input . To do this , we took a single instance of the target firing rate modulation over the integration window and used gradient descent to find the pattern of gain modulation that maximized decoding accuracy for this particular target input waveform ( see Materials and Methods ) . Only the modulation of the target input was frozen; distracting input modulations varied as before for each sample in the training and test set . We repeated this for 100 individual instances of the target firing rate modulation , each of which was different because of random variation in its frequency and amplitude and phase . The shape of the gain modulations found by this approach was similar to that found by optimized linear filtering of the target firing rate modulation ( Figure 6A–D ) . For asynchronous distractors ( Figure 6A ) and those well separated from the target signal in frequency ( Figure 6C ) , gain modulations were close to sinusoidal with a mean value of zero . For distracting inputs oscillating incoherently in the same frequency band as the target , the optimized gain modulation again strongly emphasized frequency components above and below the central frequency of the target firing rate modulation ( Figure 6B ) . Allowing the gain modulation to be an arbitrary waveform did not qualitatively change the results . As before , the degree of synchronization strongly affected decoding accuracy , with weak modulation resulting in low Fisher information ( Figure 6E ) . Distracting inputs oscillating incoherently in the same frequency band as the target severely compromised accuracy when compared with asynchronous distractors ( Figure 6E , G ) . When distractor and target modulations were well separated in frequency , decoding accuracy was comparable to when distractors were asynchronous ( Figure 6E , F ) .
This study provides a quantitative assessment of the proposal that selective communication can be achieved by coherence between firing rate modulation in a sending region and gain modulation in a receiving region [5] . Our results demonstrate that this is a viable mechanism for gating functional connectivity , potentially allowing robust routing of population-coded information in convergent pathways . However , they show a strong and previously unrecognized dependence of the accuracy of information transmission on the structure and strength of oscillatory activity across a set of inputs . Our findings question the proposal that incoherent oscillations functionally decouple anatomically connected regions . While random variation in the phase between a firing rate modulation and a gain modulation can reduce the average gain for an input to arbitrarily low levels , this is achieved at the cost of large fluctuations in gain from cycle to cycle . Unless firing rates are very low , these fluctuations are the dominant source of noise in the recovered signal , and severely limit the fidelity with which information encoded by the target input can be recovered . These random fluctuations can be greatly reduced if distracting inputs are asynchronous , or separated from the target input in frequency or phase; these more structured arrangements for multiplexing population codes permit selective communication with much lower noise and higher accuracy . The fundamental reason for this is that where inputs are distinguished by the frequency , phase or amplitude of their oscillations , patterns of gain modulation exist which are strongly driven by the target input but consistently orthogonal to distracting inputs . This is not the case for distracting inputs oscillating incoherently in the same frequency band , in which case much greater interference occurs between the signals . As we have only considered receiving networks that use multiplicative linear gain modulation , we cannot completely rule out the possibility that a network implementing a more complex operation could more accurately separate signals oscillating incoherently in the same frequency band . If we wish to retain the basic mechanism of coherence between firing rate and gain modulation , an obvious extension is to consider a class of models in which the instantaneous input-output relationship of the receiving network is a non-linear function of the input , and the gain modulation acts by changing the shape of this nonlinearity . We have explored the performance of several models in this class , including those with threshold-linear , power law and threshold-power law nonlinearities ( data not shown ) . In these experiments the gain modulation could vary both a linear input gain and the threshold and/or exponent of the non-linearity . These extensions to the model did not , however , result in any improvement in performance over the linear gain modulation outlined above . Although we cannot claim to have exhaustively explored all possible models in this class , we think it is unlikely that any approach based on coherence between firing rate and gain modulations can efficiently separate signals oscillating incoherently in the same frequency band . Though it has not been conclusively demonstrated that oscillations play a causal role in controlling functional connectivity , if this hypothesis is correct , the requirement to avoid interference between signals oscillating with different modulations is a probable organizing principle for the richly structured oscillatory activity observed in the mammalian brain . Such activity spans several orders of magnitude in frequency [31] , and in several brain regions the phase of firing is actively modulated relative to a single coherent oscillation [26] , [28] , [29] . These data suggest that the brain can indeed exploit phase and frequency separation to minimize interference between oscillatory signals . Our findings can help identify whether observed task-dependent changes in oscillatory activity in vivo are consistent with a causal role in controlling effective connectivity . Signals must be differentiated by the frequency , phase or amplitude of their modulation to be efficiently separated by coherent gain modulation , so task-dependent changes in these aspects of oscillation structure are plausible signatures of oscillatory control of effective connectivity . Conversely , changes in coherence , i . e . the consistency of a phase relationship , alone do not efficiently support changes in effective connectivity by this mechanism , and hence in the absence of other changes in the structure of oscillatory activity are more likely to be a consequence rather than a cause of changes in signal flow . Striking bursts of transient task dependent oscillatory activity are well documented in many brain regions including in motor cortex during movement preparation [32] , [33] , the basal ganglia during cue utilization [34] , and in visual cortex during working memory [35] . These transient increases in oscillation amplitude may reflect mechanisms for transiently and selectively enhancing effective connectivity between those networks participating in the oscillation event . Various studies have reported switching between distinct oscillation frequencies in a local network [36]–[38] , potentially reflecting participation in distinct large scale networks utilizing different frequencies for communication . Systematic changes in the phase of neurons relative to the hippocampal theta rhythm have been observed both within the hippocampus [39] and in extra-hippocampal regions [28] , [29] . It is unclear whether these phase shifts should be thought of as a phase code operating in parallel to and separate from rate coding , or whether they are a mechanism for multiplexing multiple firing rate population codes into distinct phases as considered here , allowing functional interactions ( or plasticity , see below ) between assemblies to be controlled through changes in relative phase . A further interesting example of phase separation was recently identified in the projection from olfactory bulb , where activity in the spatially overlapping projections made by mitral and tufted cells is segregated into opposite phases of the sniff cycle , creating putative independently accessible information channels to cortex [30] . Changes in oscillatory activity dependent on visual spatial attention [3] , [16] , [40] , [41] have been proposed to underlie the selective processing of behaviorally relevant stimuli [5] , [7] . Multisite electrocorticographic ( ECoG ) recording in primates was recently used to evaluate oscillatory synchronization simultaneously in two V1 regions representing separate visual stimuli and a V4 region receiving converging input from these areas [41] . These data provide a detailed description of attention-related changes in gamma oscillations in a convergent pathway during stimulus selection . They show a small ( ∼4% ) increase in oscillation frequency of the V1 network representing the attended stimulus over the V1 network representing the unattended stimulus , comparable gamma amplitude in both V1 regions , and a striking increase in gamma coherence between the attended V1 network and V4 . Whether this activity is compatible with the constraints we have identified depends on whether a consistent phase relationship occurs between the two V1 sites , an aspect of the activity not directly explored in the paper . If the phase relationship between the V1 sites is random , variability in the phase between the unattended V1 site and V4 would act as a substantial source of ‘overlap’ noise , limiting the accuracy of selective communication . One possibility discussed by the authors is that theta frequency resetting of gamma oscillation phase [42] across V1 and V4 , combined with the frequency offset between the V1 networks , creates periods in which the attended V1 site consistently leads the unattended site . The consistent phase offset produced in such an arrangement could be efficiently exploited for selective communication . Further analysis will be needed to establish whether such structured activity is in fact generated across the V1 networks during attention . We note that gamma oscillations in V1 are particularly amenable to experimental phase manipulation as they are readily entrained by flickering visual stimuli [43] , as expected given the response dynamics of gamma oscillating networks in vitro [44] . A recent study found no effect of manipulating the relative phase of gamma frequency flicker between target and distracting stimuli on selective attentional processing [45] , although without concurrent electrophysiological data it is unclear how effectively cortical activity was manipulated . The combination of flicker manipulations with ECoG recordings is a potentially powerful way of testing the functional importance of attention dependent changes observed in V1–V4 gamma coherence . Our data indicate that the strength of oscillatory modulation of the target signal critically determines accuracy of selective communication and hence can serve as another important clue in evaluating whether in vivo oscillatory phenomena play a causal role in controlling effective connectivity . Weak oscillations result in poor signal-to-noise ratios because the firing rate modulation read out by the receiving network is small relative to noise from stochastic spiking of individual neurons . This conclusion is likely to generalize beyond CTC to other mechanisms in which the principal carrier of information is non-zero frequency components of the firing rate generated by oscillatory network activity [15] . Estimating the modulation depth of sparsely synchronized oscillatory activity is technically challenging . Individual neurons fire irregularly at rates potentially well below the oscillation frequency , such that the spike pattern of a single neuron provides little information about the population firing rate modulation . The widely used measures of spike-field and spike-spike coherence do not map directly onto modulation strength as they are confounded by firing rate [46] , which substantially impedes attempts to evaluate modulation strength from much of the published literature . A common approach to estimating modulation strength is to look at the distribution of spikes relative to the phase of a band-pass filtered local field potential ( LFP ) oscillation . This method can underestimate modulation strength if the LFP signal is corrupted by noise , for example from neurons not participating in the oscillation , or if the analysis combines activity from periods with and without strong oscillation . Despite these technical difficulties reported spike phase histograms show a wide range of modulation strengths across different oscillations , from very strong modulations during hippocampal theta oscillations [47] , and oscillations in the olfactory system of zebrafish [27] and locusts [48] , to apparently weaker modulation in some studies of gamma oscillations in the hippocampus [49] , [50] and entorhinal cortex [49] . Our results suggest that , where oscillations are genuinely weak , mechanisms exploiting them for selective routing of signals would recover only a tiny fraction of the information represented in the sending population . Our use of a highly simplified non-biophysical model in this work was necessary to permit model optimization and hence to find an upper bound on how accurately coherent gain modulation could separate target from distracting inputs . However , it raises the question of whether a biological network or biophysical model could achieve this performance . A biophysical implementation of the receiving network must generate approximately multiplicative gain modulation coherent with either a top-down control input , or with a particular frequency component of the combined input in a bottom-up fashion . Several biophysical mechanisms including shunting inhibition and synaptic noise are known to produce approximately multiplicative gain modulation in individual neurons [20] , [21] . Entrainment of oscillatory or resonant local circuitry in the receiving network is a plausible mechanism for generating the required temporal patterning of gain modulation . We recently demonstrated that the dynamical properties of gamma oscillations in the CA3 region support entrainment to periodic inputs [44] , though the consequences of such entrainment for the gain of signal transmission remain to be established . These data suggest that entrainment phase may be controlled by varying the natural frequency of the network relative to the input , or by varying the relative coupling of the input to excitatory and inhibitory populations . We previously developed a biophysical model [15] , which exploits network resonance effects at the boundary between asynchronous and oscillating states [51] to selectively respond to inputs oscillating at a specific frequency . While the biophysical model implemented a similar functionality to the ‘bottom-up’ coherence configuration considered here , there are some key differences in the mechanism of operation that bear outlining explicitly . In both models , information was represented in the input networks as spatial patterns of firing rate , while the target input was differentiated from distractors by multiplicative modulation ( here represented explicitly , in the biophysical model generated by sparsely synchronized network dynamics ) . Both models exploit the fact that multiplicative modulation selectively reproduces the spatial pattern of firing rate into those higher frequency components of the firing rate present in the modulation [15] . However , the models differ in the way the receiving network reads out the resulting spatial patterns of firing rate oscillation . The biophysical model essentially converts the amplitude of input firing rate oscillation at a given frequency into the average firing rate of the output neurons through a process of bandpass filtering followed by half-wave rectification . The bandpass filtering is implemented by a combination of resonant feed forward inhibition and synaptic filtering which ensures that the net input current received by the output neurons is a bandpass-filtered version of the input activity . The spike threshold then rectifies this input current to produce an output firing rate . In the current model , readout is by multiplicative gain modulation followed by integration over time , exploiting the orthogonality of different frequency components under overlap integration to separate target from distracting signals . Thus although the previous biophysical model utilizes a similar coding strategy to the current model and achieves similar functionality to the ‘bottom up’ configuration , the mechanisms underlying the filtering in the two models differ significantly . Implementation of biophysical models that operate on the same principle as the current model is a clear direction for future work . Several studies working in this direction [12] , [14] , [52] , [53] have demonstrated some degree of input selectivity on the basis of modulation , particularly for phase separated inputs [52] , [53] . Our understanding is that the selective communication performance achieved by these biophysical models is substantially lower than the current optimized model . Further work is required to establish how efficiently and robustly biophysical networks can utilize coherent gain modulation to extract information multiplexed into patterns of firing rate modulation , and what network architectures are effective in this task . Throughout this work we have discussed gain modulation that acts on the input-output relationship for the activity of a population of neurons . Oscillatory activity can also modulate the gain of synaptic plasticity [54] , [55] , and spike timing dependent plasticity with an oscillating post synaptic population will also produce periodic modulation of the gain for plasticity . Periodic modulation of the gain for plasticity coherent with the firing rate modulation of a target input could selectively enhance plasticity for that input just as coherent modulation of neuronal input-output gain can permit selective response to a target input . As our results are due to signal to noise considerations they are equally applicable to identifying which structures of activity permit accurate selective plasticity of a subset of inputs by oscillatory modulation of the gain for plasticity . In conclusion , accurate and selective communication can be achieved by coherence between gain and firing rate modulations . However , to achieve a high signal to noise ratio the oscillatory modulation of the target signal must be strong , and distracting inputs must be distinguished from the target by frequency , phase or amplitude of oscillation . Failure to satisfy these constraints greatly reduces the accuracy of information transmission . Where oscillatory activity plays a causal role in modulating functional connectivity we expect it to be organized to maximize the accuracy of signal propagation . | Distributed regions of mammalian brains transiently engage in coherent oscillations , often at specific stages of behavioral or cognitive tasks . This activity may play a role in controlling information flow among connected regions , allowing the brain's connectivity structure to be flexibly reconfigured in response to changing task demands . We have used a computational model to investigate the conditions under which oscillations can generate selective communication through a mechanism in which the excitability of neurons in one region is modulated coherently with a firing rate oscillation in another region . Our results demonstrate that this mechanism is able to accurately and selectively control the flow of signals encoded as spatial patterns of firing rate . However , we found that the requirement to avoid interference between different signals imposes previously unrecognised constraints on the structures of oscillatory activity that can efficiently support this mechanism . These constraints may be an organizing principle for the structured oscillatory activity observed in vivo . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"circuit",
"models",
"computational",
"neuroscience",
"biology",
"neuroscience",
"coding",
"mechanisms"
] | 2012 | Efficient “Communication through Coherence” Requires Oscillations Structured to Minimize Interference between Signals |
Sensory hair cells are essential for hearing and balance . Their development from epithelial precursors has been extensively characterized with respect to transcriptional regulation , but not in terms of posttranscriptional influences . Here we report on the identification and functional characterization of an alternative-splicing regulator whose inactivation is responsible for defective hair-cell development , deafness , and impaired balance in the spontaneous mutant Bronx waltzer ( bv ) mouse . We used positional cloning and transgenic rescue to locate the bv mutation to the splicing factor-encoding gene Ser/Arg repetitive matrix 4 ( Srrm4 ) . Transcriptome-wide analysis of pre–mRNA splicing in the sensory patches of embryonic inner ears revealed that specific alternative exons were skipped at abnormally high rates in the bv mice . Minigene experiments in a heterologous expression system confirmed that these skipped exons require Srrm4 for inclusion into the mature mRNA . Sequence analysis and mutagenesis experiments showed that the affected transcripts share a novel motif that is necessary for the Srrm4-dependent alternative splicing . Functional annotations and protein–protein interaction data indicated that the encoded proteins cluster in the secretion and neurotransmission pathways . In addition , the splicing of a few transcriptional regulators was found to be Srrm4 dependent , and several of the genes known to be targeted by these regulators were expressed at reduced levels in the bv mice . Although Srrm4 expression was detected in neural tissues as well as hair cells , analyses of the bv mouse cerebellum and neocortex failed to detect splicing defects . Our data suggest that Srrm4 function is critical in the hearing and balance organs , but not in all neural tissues . Srrm4 is the first alternative-splicing regulator to be associated with hearing , and the analysis of bv mice provides exon-level insights into hair-cell development .
Hair cells of the hearing and balance organs are specialized mechanoreceptors that convert mechanical stimuli to electrical signals . These signals are transmitted to the central nervous system via connecting afferent neurons . Hair cells of the hearing organ are specialized further as inner hair cells ( IHCs ) and outer hair cells ( OHCs ) . IHCs are the primary auditory receptors , whereas the electromotile OHCs ( and their neural feedback loops ) are amplifiers of the mechanical stimulus [1] . In mice , hair cells become responsive to mechanical stimuli between embryonic day ( E ) 17 and postnatal day ( P ) 4 [2] , [3] . During this period , the mechanosensing stereociliary bundles of the hair cells grow and become organized into rows of increasing height [4] . Defects in either stereocilium formation or afferent synaptogenesis lead to deafness and impaired balance [5]–[8] . The development of sensory hair cells is governed by several known transcription factors . For example , Sox2 , Eya1 , and the Notch effectors Hes1 and Hey1/2 are key transcriptional regulators of the specification process that guides the undifferentiated otocyst cells towards a prosensory fate [9] . The prosensory cells can then differentiate into either hair cells or supporting cells , depending on the presence or absence of the basic helix-loop-helix transcription factor Atoh1 within the cell . Genetic deletion of Atoh1 leads to the complete absence of hair cells [10] , whereas the ectopic expression of Atoh1 in supporting cells can induce the formation of stereociliary bundles and the expression of hair-cell markers [11]–[13] . Atoh1 also induces the expression of at least 2 other transcription factors ( i . e . Pou4f3 and Gfi1 ) that are required for the terminal differentiation of hair cells [14]–[17] . Proper hair-cell differentiation has also been shown to depend on microRNA-96 ( miR-96 ) , a post-transcriptional regulator of gene expression . Mutations in the miR-96 genes of both humans and mice have been associated with deafness [18] , [19] , and the analysis of mice harboring such mutations has demonstrated that this miR is required for the maturation of stereociliary bundles , as well as for the establishment of auditory nerve connections [20] . In addition , the analysis of knockout mice lacking the miR-processing protein Dicer1 in the inner ear supports the notion that miR-dependent regulation of gene expression plays a critical role in hair-cell differentiation [21] , [22] . In an effort to identify additional regulatory mechanisms that are necessary for hair-cell development , we analyzed the bv mouse line , whose inner ear pathology suggested that the bv mutation disrupts a gene that is key to the differentiation of most hair-cell types [23]–[26] . Although the hair cells of homozygous bv ( bv/bv ) mice are morphologically intact until E17 . 5 , neither the IHCs in the hearing organ nor the vestibular hair cells ( VHCs ) in the balance organs develop normally beyond this point . Specifically , the IHCs and VHCs fail to form synapses with afferent neurons , exhibit delayed stereociliary-bundle growth , and tend to degenerate by P3–5 [25]–[27] . These hair-cell defects are associated with the deafness and impaired balance observed in these animals . The bv/bv mouse is unique amongst the deaf mouse models in that IHC degeneration is not accompanied by the loss of OHCs [23] , [27] . In this study , we localize the deafness-causing gene defect of the bv mouse line to the splicing factor-encoding gene Srrm4 ( also known as nSR100 [28] ) . Because Srrm4 is expressed broadly in neural tissues , we used tissue-selective transgenic rescue to examine the biological importance of Srrm4 in and outside of the inner ear . The results of these rescue experiments indicated that defective Srrm4 function specifically in hair cells is the main , if not the only , cause of the bv phenotype . We evaluated the molecular function of Srrm4 using a transcriptome-wide approach , and found that it was required for neuron-like alternative splicing in the developing sensory patches of the inner ear but not in other Srrm4-expressing tissues that we examined . The majority of the affected pre-mRNAs encoded proteins with functions related to neurotransmission and secretion , confirming the notion that alternative splicing factors can selectively alter specific functional modules in the cell . Moreover , Srrm4-dependent splicing in hair cells affected transcriptional regulators that are known to control cell differentiation and presynaptic vesicle processing in neural tissues . Thus , our analysis of the bv mouse line suggests that Srrm4-dependent regulation of alternative exon choice has a profound effect on the differentiation program of sensory hair cells .
Although the gene affected by the bv mutation was not identified in previous studies , it had been mapped to a 4-mega base pair ( bp ) interval on chromosome 5 [29] . We examined the tissue expression profiles and putative functions of all 63 genes localized within this interval and selected 12 for further analysis ( Figure 1A ) . In amplifying the transcripts of the 12 genes , we found that one , which encodes the splicing factor Srrm4 , was abnormally short in bv/bv mice ( Figure 1B ) . Sequence analysis showed that the shortened Srrm4 transcript lacked hundreds of nucleotides and retained an intronic sequence ( Figure 1B ) , whereas the other amplified transcripts did not contain any mutations ( data not shown ) . Sequencing of the 3′ end of the Srrm4 gene in bv/bv mice revealed a 2 , 710-bp deletion that removed a portion of the last intron and the entire coding region of the last exon but left the polyadenylation site intact ( Figure 1C and Table S1 ) . The affected last exon of Srrm4 encodes potentially important domains of the Srrm4 protein , including the C-terminal SR repeats and a region that is highly conserved between Srrm4 and its closest paralogue , Srrm3 ( Figure 1D ) . We used Western blotting to examine expression of the wild-type Srrm4 ( Srrm4wt ) protein and that of the mutant form encoded by the bv mouse genome ( Srrm4bv ) . In nuclear pellets generated from Srrm4wt-transfected HEK293 cells and from the sensory regions of the balance organs ( i . e . vestibular maculas ) in wild-type mice , the protein was detected as multiple bands between 82 and 115 kDa ( Figure 1E ) . In contrast , in nuclear pellets generated from Srrm4bv-transfected HEK293 cells and the vestibular maculas of bv/bv mice , only a single , faint band was detected at ∼82 kDa ( Figure 1E ) . Thus , the bv mutation not only truncates the Srrm4 protein but also interferes with either the stability or synthesis of the truncated protein . Next , we used in situ hybridization to examine the expression pattern of wild-type Srrm4 in the inner ear , and found that the Srrm4 mRNA was detected in all sensory regions of hearing and balance organs ( Figure 2A ) . In the cochlea , the antisense Srrm4 probe labeled the IHCs , OHCs , and spiral ganglion ( Figure 2B ) . In the utricle , the most intensive staining was found at the periphery of the sensory macula ( Figure 2C ) , where the density of VHCs is highest [30] . In the crista ampullaris , the VHC-containing regions were strongly positive , whereas the non-sensory septum cruciatum ( present in the anterior but not the lateral crista [31] ) , was not labeled ( Figure 2D , asterisk ) . The negative-control , Srrm4 sense probe did not hybridize with any of the sensory regions in the inner ear ( Figure S1 ) . These data indicate that Srrm4 is expressed in the sensory hair cells and the spiral ganglion . RT-PCR experiments showed that the Srrm4 mRNA was also present in the brain but not in the kidney , liver , or spleen ( data not shown ) , consistent with the previously reported neural expression pattern of Srrm4 [28] . Although Srrm4 is expressed broadly in neural tissues , we hypothesized that the Srrm4 defect in hair cells is the cause of the bv phenotype . We tested this possibility by transgenic rescue . Specifically , a Myo7a-Srrm4 transgene ( Figure 3A ) was constructed in which the hair-cell specific promoter of Myo7a [32] controlled the transcription of wild-type Srrm4 . Transgenic founder mice were generated via pronuclear injection of the Myo7a-Srrm4 transgene , and bred into the bv/bv line . In situ hybridizations of inner ear samples with a probe corresponding to the coding region in Srrm4 exon 13 demonstrated that the Myo7a-Srrm4 transgene was expressed in the balance organs and IHCs of the transgenic bv/bv mice ( Figure S2 ) . In contrast , the inner ears of bv/bv mice did not hybridize with the Srrm4 exon 13 probe ( Figure S2 ) , confirming that exon 13 is missing from the Srrm4 transcript in the bv/bv mice . We evaluated the hearing of Myo7a-Srrm4 transgenic bv/bv mice and non-transgenic bv/bv littermates by measuring the auditory brainstem response ( ABR ) of these animals on P21–28 , using broadband sounds . The ABR measurements confirmed that the bv/bv mice were severely hearing impaired ( Figure 3B ) , whereas the hearing thresholds of several Myo7a-Srrm4 transgenic bv/bv mice were at the wild-type level ( Figure 3B and 3C ) . Nevertheless , the extent to which hearing was restored in individual Myo7a-Srrm4 transgenic bv/bv mice varied ( Figure 3C ) by transgenic lineage ( Figure S3 ) , and probably reflected differences in transgene insertion site or copy number . We also assessed balance in the transgenic bv/bv animals , by measuring the length of time they could remain on a horizontal rod . The performance of the Myo7a-Srrm4 transgenic bv/bv mice was similar to that of bv/+ animals , which were able to remain on the rod for the duration of the assay ( 60 s ) ; in contrast , most bv/bv mice fell within 10 s ( Figure 3D ) . We next analyzed the effects of the Myo7a-Srrm4 transgene on hair-cell survival in bv/bv mice . The sensory regions were dissected from the balance organs of bv/+ , bv/bv , and Myo7a-Srrm4 transgenic bv/bv mice on P5 , and the actin-rich stereociliary bundles of VHCs were visualized using fluorescently labeled phalloidin . Looking at the balance organs of bv/bv and Myo7a-Srrm4 transgenic bv/bv mice , we found that the vast majority of stereociliary bundles were absent in the former but present at nearly normal density in the latter ( see utricle in Figure 3E and quantitative analysis in Figure S4A ) . In the cochleas of bv/bv mice , both actin staining of stereocilia and immunofluorescence-based visualization of the hair-cell protein Myo7a indicated that 71% of the IHCs were absent on P5 . In contrast , in the Myo7a-Srrm4 transgenic bv/bv mice 63% of IHCs were present at this time ( Figure 3F and Figure S4 ) . These results indicate that the Srrm4 mutation is responsible for the hair-cell loss , deafness , and balance defect in the bv mouse line . Our data also support the notion that the inner ear pathology in bv/bv mice is caused by defects in the hair cells rather than in the neurons . Srrm4 belongs to the family of SR-related proteins , which act as regulators of alternative pre-mRNA splicing [33] , [34] . Therefore , we examined whether alternative splicing was altered in the embryonic hair cells of bv/bv mice , using a transcriptome-wide approach . Specifically , embryonic hair cells ( and the adjacent supporting cells ) were acquired from the vestibular maculas of bv/bv and bv/+ mice , by laser-capture microdissection , on E16 . 5 , i . e . ∼1 day before the onset of hair-cell degeneration . RNA from the captured tissue was analyzed using the new Affymatrix chip ‘Mouse Exon Junction Array’ ( MJAY , Figure 4A ) . MJAY contains more than half a million exon and exon-exon junction probe sets ( see probe-set design in Figure 4B ) , and interrogates all of the splicing events supported by mouse EST/mRNA evidence within the UCSC/Ensembl databases . Processing of the MJAY data was carried out largely in the Partek Genomics Suite ( see details in Methods ) , based on concepts that were previously described for the analysis of Human Exon Junction Array data [35] . The frequency of an alternative splicing event was considered to differ significantly between the bv/bv and bv/+ samples if the difference in normalized intensities for at least two probe sets per splicing event ( Figure 4C ) resulted in P-values less than 0 . 05 . Seventy-six candidate alternative splicing events were found and tested further by RT-PCR , using primers that annealed with the constitutive exons upstream and downstream of the alternative exons . These reactions validated 24 alternative splicing events in the vestibular maculas of bv/bv mice ( Figure 4D and Figure S5A ) . Notably , examination of these splicing defects indicated that , in the bv/bv cells , certain alternative exons were either spliced into the mature mRNA at reduced frequency or completely skipped . Common features of the affected exons included conservation among vertebrates ( data not shown ) and – with the exception of Add1 exon 15 – a neuron-specific inclusion pattern ( Figure S5C ) . Therefore , we used ‘conservation’ and ‘neuron-specific splicing’ ( based on EST evidence ) as new criteria with which to scrutinize the list of exons for which a single probe set suggested abnormal splicing in the bv/bv mice . RT-PCR revealed that , among the 283 new candidate exons , 30 were incorrectly spliced in the bv/bv sample ( Figure S5B ) . Thus , overall , RT-PCR verified 54 changes in splicing in the bv/bv mouse ( see Table S2 ) . We used the DAVID software [36] to analyze the gene ontology ( GO ) annotations of the encoded proteins , and found that the lowest P values were for those associated with the ‘transmission of nerve impulse’ ( Benjamini-Hochberg corrected P-value = 0 . 00047 ) , ‘secretion by cell’ ( P-value = 0 . 0064 ) , and other closely related GO terms ( e . g . ‘cell-cell signaling’ ) . The majority of splicing defects we found in the vestibular macula of bv/bv mice ( 81% ) had not been reported in an earlier study that examined Srrm4 function in the Neuro2A cell line [28] . Conversely , although nPTB ( exon 10 ) was found to be a key target of Srrm4 in the Neuro2A cells [28] , it was spliced normally in the vestibular maculas of bv/bv mice ( Figure S5D ) . Nevertheless , there were striking instances of overlap as well . For example , the RE1 silencing transcription factor ( Rest , exon 4 ) was reported as a target of Srrm4 in the Neuro2A cells [37] , and RT-PCR showed that the same Rest exon was differently spliced in the inner ears of bv/bv and bv/+ mice ( Figure S5E ) . Furthermore , the Srrm4-dependent splicing of Rest had been shown to affect the expression of numerous Rest-regulated genes in Neuro2A cells [37] , and our analysis of the MJAY data suggested that Rest-regulated genes [38] were overrepresented among those whose expression was reduced in the vestibular macula of bv/bv mice ( χ2 test P<0 . 0001 , Figure S6 ) . Notably , our data showed that the Phf21a/Bhc80 mRNA , which encodes a negative modulator of Rest-dependent transcriptional regulation [39]–[41] , was also differentially spliced in the vestibular maculas of bv/bv and bv/+ mice ( Figure 4D ) . These results support the notion that Srrm4 modifies gene expression in hair cells , probably through the alternative splicing of specific transcriptional regulators . We also wanted to test whether the bv mutation led to splicing alterations in Srrm4-expressing tissues other than the inner ear . We focused on the cerebellum , based on our RT-PCR analysis showing that the Srrm4 transcript is highly expressed in this tissue ( Figure S7A ) , and in situ hybridization data in the Allen Brain Atlas [42] indicating that the neuron-rich layers of the cerebellum contain large amounts of Srrm4 mRNA . Notably , although the analysis of MJAY data identified 18 alternative exons as potentially differently spliced in the cerebellums of bv/bv and bv/+ mice on P15 , anlaysis by RT-PCR failed to validate such an outcome ( Figure S7C and S7D ) . Furthermore , both the MJAY and RT-PCR data showed that , in the bv mouse line , inclusion rates for alternative exons that were abnormally spliced in the vestibular macula were unaltered in the cerebellum ( Figure 4E and 4F and Figure S7E ) . Given that the Srrm4 mRNA is highly expressed in the neocortex [28] , we used RT-PCR to test the inclusion rates of 10 Srrm4-regulated exons in this tissue . Again , we found no alterations in the inclusion rates of tested alternative exons in the investigated brain region of bv/bv mice ( Figure S8 ) . These findings are supported by the lack of obvious histological alterations in the cerebellum and neocortex of bv/bv mice ( Figure S7B and Figure S9 ) . In sum , the bv mutation does not lead to apparent defects in these Srrm4-expressing tissues . We next used a reconstituted system to evaluate whether the Srrm4bv protein retains molecular function . Specifically , HEK293 cells were transfected with Srrm4bv , Srrm4wt , or empty vector ( control ) alongside various minigenes consisting of exons and introns . Each minigene construct contained an exon that was incorrectly spliced in the vestibular macula of bv/bv mice , the flanking intronic sequences ( ∼300 bp ) , and two constitutive exons ( Figure 5A ) . Of the 54 alternative exons whose inclusion rates were found to be altered in the vestibular maculas of bv/bv mice , 12 were randomly selected for these minigene experiments . RT-PCR-based evaluation of pre-mRNA splicing demonstrated that all 12 exons required Srrm4wt for alternative splicing in the transfected cells , and that Srrm4bv was unable to promote such splicing ( Figure 5A and Figure S10A ) . When the minigenes were co-transfected with a construct encoding an SR protein other than Srrm4 ( i . e . Srsf1 ) , the inclusion rates of the alternative exons did not increase above background levels ( Figure S10A ) . Thus , the minigene experiments confirmed that splicing of the tested exons is dependent on Srrm4wt , and also indicated that the bv truncation prevents the expression of functional Srrm4 protein in transfected HEK293 cells . We also tested the functional status of Srrm4bv in vivo , using zebrafish as an animal model . The endogenous Srrm4 mRNA of zebrafish ( zSrrm4 ) was knocked down by injecting a previously described zSrrm4 morpholino ( MO ) [28] into fish eggs . Some of these eggs were also injected with either an mRNA encoding a MO-insensitive wild-type zSrrm4 ( zSrrm4wt ) or the zebrafish version of a MO-insensitive Srrm4bv ( zSrrm4bv ) . Three days later , the hair cells were visualized in the lateral line of zebrafish larvae using the fluorescent dye FM1–43 [43] , [44] . We found that in the zSrrm4 MO-injected fish , the body axis was abnormally curved ( Figure 5C versus Figure 5B , upper panel ) . This deformity in the body axis has previously been described for Srrm4 knock-down zebrafish , and has been attributed to neuronal defects [28] . In addition , we found that the number of hair cells was dramatically reduced in the zSrrm4 MO-injected fish ( Figure 5C versus Figure 5B , lower panel , and quantitative analyses in Figure S10B and S10C ) . Co-injection of the zSrrm4wt mRNA with the MO rescued both the body axis deformity and the hair-cell loss ( Figure 5D ) , whereas co-injection of the zSrrm4bv mRNA did not ( Figure 5E , and statistical analyses in Figure S5B and S5C ) . These data suggest that Srrm4bv is not functional , regardless of the expression system . Furthermore , our data show that although the loss of Srrm4 function has a broader phenotypic impact in zebrafish than in mice , Srrm4 is essential for hair-cell development in both species . We hypothesized that a unique sequence motif may mark the Srrm4-regulated exons for splicing . Initially , we focused on exon sequences , testing a 9-nucleotide long exon whose splicing we had found to be Srrm4 regulated ( i . e . Dtna exon 11 ) . However , random mutation of 5 consecutive nucleotides in the 9-nucleotide exon did not affect its Srrm4-dependent splicing ( ) . Next , we used the MEME software [45] to search for consensus motifs in both the Srrm4-regulated exons ( n = 54 ) and the 50-nucleotide long portions of introns that are directly adjacent to these exons . MEME identified 3 motifs with P-values lower than 0 . 05 , including a novel UGC motif and the known binding sequences of 2 splicing factors ( i . e . U2af1 and the U1 small nuclear ribonucleoprotein ) . Alignment of the intron sequences upstream and downstream of the UGC motif showed that it is located near the 3′ end of the polypyrimidine tract ( Figure 5F ) . To test whether UGC commonly occurs upstream of exons ( i . e . , regardless of Srrm4-dependent alternative splicing ) , we assessed all 50-bp regions that lie directly upstream of an exon in the Cacna1d and Ergic3 pre-mRNAs ( n = 60 ) . MEME did not detect UGC as a frequent motif in these sequences . Thus , UGC might be important for specifically Srrm4-dependent splicing . Next , we used G-to-A point mutations to disrupt the selected UGC motifs in 6 minigenes that were randomly chosen from those that require Srrm4 for alternative splicing . In all cases , the mutations inhibited Srrm4-dependent exon inclusion ( Figure 5G , Figure S11B–S11C ) . Mutagenesis experiments were also carried out to test the importance of the other two nucleotides in the UGC motif . We found that C-to-U substitutions ( Figure 5G and Figure S11C ) , but not U-to-C/G/A mutations ( data not shown ) inhibited Srrm4-dependent exon inclusion . Thus , although a U nucleotide most often precedes the functionally relevant GC motif , only the GC nucleotides are required for Srrm4-dependent alternative splicing . Blast searches revealed that the functionally relevant GC motifs are conserved among vertebrate species ( Figure S11D ) . To test whether these motifs interact with Srrm4 , we carried out streptavidin pull-down assays using two types of biotin-labeled RNA oligos . The ‘wild-type’ RNA oligo corresponded exactly to a 40-nucleotide long sequence around the splice acceptor site of the Srrm4-regulated exon in the Ergic3 pre-mRNA , whereas the ‘mutated’ RNA oligo contained a GC-to-AU substitution ( Figure 5H ) . Western blot analysis of cell lysates prepared from flag-Srrm4-transfected HEK293 cells showed that only the ‘wild-type’ RNA efficiently pulled down flag-tagged Srrm4 ( Figure 5H ) . These results suggest that conserved GC motifs upstream of the splice acceptor sites of Srrm4-regulated exons are necessary for the interaction between Srrm4 and the pre-mRNA .
In the present study , we show that the hearing and balance defects of bv/bv mice are caused by a mutation in the Srrm4 gene . Using the bv mouse line and a genome-wide screening method to analyze the molecular function of Srrm4 in vivo , we identified Srrm4 as a key regulator of pre-mRNA splicing in the inner ear . Moreover , we found that Srrm4 was required for the alternative splicing of a specific set of exons that are marked by GC motifs near the 3′ ends of the polypyrimidine tracts . The Srrm4 mutation in bv/bv mice also affected gene expression in the sensory patches of the inner ear , suggesting that Srrm4 controls a cascade of transcriptome-modifying events . Based on this analysis of the bv mouse line , we propose that Srrm4-regulated alternative splicing is critical for the differentiation of all sensory hair-cell types except the OHCs . Although Srrm4 is expressed broadly in neural tissues , we did not detect splicing defects in the cerebellum and neocortex of P15 bv/bv mice . Moreover , if neurogenesis is impaired in bv/bv mouse embryos , the consequences of this defect are not readily detectable by Nissl staining at P15 . Nevertheless , we cannot rule out the possibility that pre-mRNA splicing is affected at other time points or in other brain regions in these animals . Notably , a recent study showed that neurogenesis was impaired in E13/14 wild-type mice after neural progenitors in the ventricular zone were electroporated with vectors encoding an Srrm4-targeting shRNA [37] . Also , in examining the brains of bv mice , Matsuda and colleagues observed that immunofluorescence-based visualization of the parvalbumin-expressing GABAergic interneurons [46] detected abnormally few parvalbumin-expressing cells in the auditory cortex , somatosensory cortex , and anterior cingulate , whereas the visual cortex and the amygdala complex were unaffected . A possible interpretation of these data is that the Srrm4 defect in bv/bv mice directly affects the differentiation of interneurons in certain brain regions . Alternatively , some of the observed changes in the number of parvalbumin-expressing cells could be secondary to the hearing and balance defects in bv/bv mice . This possibility is consistent with the fact that congenital deafness has been shown to prevent the maturation of GABAergic transmission in the auditory cortex [47] , [48] , and sensory hearing loss has been associated with a decrease in the number of parvalbumin-positive cells in the superior olivary complex [49] . In addition , the lack of vestibular input has been reported to cause a reduction in the expression of various calcium-binding proteins , including parvalbumin , in the medial vestibular nucleus [50] . Thus , additional studies will be necessary to establish the etiology of the altered GABAergic interneuron density in certain brain regions of bv/bv mice . In zebrafish , MO-mediated knock-down of Srrm4 has an obvious effect on both neural differentiation [28] and hair-cell development ( Figure 5 ) . Why does Srrm4 deficiency have a much greater impact on neural differentiation in zebrafish than in mice ? One possible explanation is that splicing proteins other than Srrm4 have Srrm4-like functions in the mouse brain , but not in that of zebrafish . However , a more complex explanation is suggested by two findings . Firstly , whereas approximately 70% of the IHCs die between E18 and P5 in the bv mouse , this trend does not continue after P5 [27] . Secondly , the surviving IHCs are most likely functional because the bv/bv mice are not completely deaf . Together these data suggest that Srrm4 is not needed in the inner ear after a critical phase in development . This “critical phase” hypothesis is supported by the gene expression profile of the splicing suppressor polypyrimidine tract binding protein 1 ( PTBP1 ) , which has been shown to inhibit the constitutive inclusion of at least some Srrm4-regulated exons in Neuro2A cells [28] . PTBP1 is expressed in neural cells only during the early phases of differentiation [51] . Thus , Srrm4 may not be needed during the later phases of development when neural cells no longer contain PTBP1 . We speculate that although Srrm4 deficiency could possibly lead to splicing defects in the neurons of both mice and zebrafish during early development , Srrm4-independent regulatory mechanisms are sufficient to support neuron differentiation until the end of the critical phase in mice but not in zebrafish . Our finding that Srrm4-dependent exon inclusion requires the presence of a GC motif near the 3′ end of the polypyrimidine tract suggests that this motif serves as a cis-regulatory element for Srrm4-dependent splicing . Cis-regulatory elements are short sequence motifs that recruit RNA-binding proteins [52]; they can either enhance or suppress exon inclusion depending on which splicing factors are recruited and – in some cases – the position of the cis-regulatory element relative to the exon [53] , [54] . Pre-mRNAs co-regulated by the same RNA-binding protein usually contain the same cis-regulatory element . Thus , the presence of the same motif next to almost every affected exon in the Srrm4 mutant mouse suggests that the inclusion of these exons into the mRNA is regulated by the same Srrm4-dependent mechanism . Our RNA pull-down experiment suggested that the GC motif is necessary for the interaction between Srrm4 and the RNA . Whether this interaction is direct or mediated through other proteins remains to be determined . Notably , the GC motif is not the only sequence in the pre-mRNA that is important for the regulation of Srrm4-dependent splicing events . A previous study showed that pyrimidine-rich motifs are often present in introns that flank Srrm4-regulated exons [28] , and that these pyrimidine-rich motifs are binding sites for PTBP1 [28] . Because the GC motifs are located near the 3′ end of the polypyrimidine sequences , it is tempting to speculate that the recruitment of either Srrm4 or Srrm4-binding proteins to the pre-mRNA may interfere with the binding of PTBP1 . The fact that the Srrm4-regulated exons were found more frequently in the transcripts of proteins that are annotated with the GO terms ‘transmission of nerve impulse’ and ‘secretion by cell’ than in the transcripts of a random set of proteins suggests that the protein products of the Srrm4-regulated pre-mRNAs are functionally linked . We explored this possibility by collecting PubMed data on the subcellular localizations of , and interactions among , the affected proteins; given that we wanted to maximize the amount of information gathered , we did not restrict these PubMed searches to hair cell-related publications . Based on the information collected , we charted the likely subcellular localization of the affected proteins on a schematic model of the basolateral portion of a hair cell ( Figure 6 ) . This model illustrates that the majority of the proteins encoded by Srrm4-regulated transcripts may be associated with synaptic vesicles and the presynaptic plasma membrane . Notably , 42% of the proteins in this model that are encoded by Srrm4-regulated pre-mRNAs and have known protein-protein binding partners in the PubMed database interact with each other ( see reference list in Table S3 ) . Thus , both GO annotation analysis and the protein-protein interaction patterns suggest that the Srrm4-dependent modifications cluster predominantly in a single functional module of the proteome , and that this module is responsible for secretion and neurotransmission at the presynaptic side of the synapse . This analysis also suggests that the Srrm4-regulated proteins with uncharacterized molecular functions ( e . g . Plekha6 , 6330403A02Rik , and C230096C10Rik ) are more likely involved in secretion or neurotransmission than in other biological processes . Sustained high rates of neurotransmission from the IHCs and VHCs to their respective neural afferents require specialized presynaptic structures termed synaptic ribbons . Most OHCs contain synaptic ribbons only temporarily during differentiation; the only OHCs in which synaptic ribbons persist are those that are most apical [55] . Interestingly , in the bv/bv mice the OHCs are the only hair cells to remain intact , and many of the proteins with splicing defects are localized to synaptic ribbons ( i . e . Cacna1d , Cask , Erc2 , Rims2 , Snap91 , and Synj1 [56] ) . Thus , it seems plausible that the cell-type specificity of the synaptogenesis defect is due to an absence of protein isoforms that are specifically required for the formation of synaptic ribbons . Alternatively , it is possible that the mechanism that supports the inclusion of Srrm4-regulated exons in the cerebellum and neocortex of bv/bv mice also protects the OHCs from degeneration . These hypotheses could be tested by analyzing pre-mRNA splicing in the embryonic OHCs of bv/bv and control mice . However , RNA collection selectively from embryonic OHCs is technically challenging because of the physical proximity of OHCs and IHCs in the developing inner ear . Therefore , the analysis of pre-mRNA splicing in the OHCs of bv/bv mice is yet to be carried out . Although we found that the majority of the splicing defects in the bv mouse line were associated with the secretory and synaptic apparatuses , the alternative splicing of at least two ciliary protein-encoding mRNAs ( i . e . Bbs9 and Wdr35 ) were also altered . In addition , the Srrm4 mutation led to reduced expression of the receptor-like inositol phosphatase Ptprq , which is required for the development of stereociliary bundles in the cochlea [57] . Srrm4-dependent splicing also affected at least 3 mRNAs that encode nuclear proteins ( i . e . Rest , Bhc80 , and Mef2d ) . Two of these ( i . e . Rest and Bhc80 ) have been shown to have opposing effects on gene expression and have been reported to control vesicle processing and exocytosis through translational regulation [41] , [58] , [59] . We found that the genes regulated by Rest- and Bhc80 – but not those regulated by Mef2d – were overrepresented among the 44 whose expression was most reduced in the vestibular macula of bv/bv mice . Thus , the Srrm4-dependent splicing of the Rest and Bhc80 pre-mRNAs supports our hypothesis that Srrm4 plays a role in maturation of the regulated secretory apparatus in hair cells . Altered splicing of the Rest mRNA and reduced expression of the Rest target genes in the context of reduced Srrm4 function were described previously in Neuro2A cells subjected to RNA interference [37] . Thus , both in vivo and in vitro data suggest that the loss of Srrm4 function leads to a cascade of transcriptome alterations that affect both pre-mRNA splicing and gene expression . Further studies defining the importance of individual Srrm4-regulated exons in hair-cell development will enable us to elucidate the detailed pathogenesis of hair-cell degeneration in bv/bv mice . In summary , in analyzing the bv mouse line we have identified Srrm4 as a regulator of alternative splicing that is required for the differentiation of hair cells in the hearing and balance organs . We propose that a Srrm4-regulated cascade of transcriptome modifying events adjusts the proteome of differentiating hair cells such that they take on neuron-like functions . Our study adds alternative splicing to the list of mechanisms that are critical for hair-cell differentiation . Given that some deafness-causing mutations are known to be localized to alternative exons ( i . e . R643X in PCDH15 [60] and R500X in TRIC [61] ) , understanding the regulation of alternative exon choice in the inner ear is expected to create therapeutic opportunities for the prevention of deafness .
The bv mouse strain was recovered from cryopreserved sperm samples ( obtained from the European Mutant Mouse Archive ) , by intracytoplasmic sperm injection . All experiments and procedures were approved by the Animal Care and Use Committee of the University of Iowa . For mutation analysis , RNA was isolated from the inner ear of bv/bv and wild-type mice ( E16 . 5 ) , using the Trizol reagent . The coding regions of candidate mRNAs were amplified from the RNA samples using RT-PCR ( see primers in Table S4 ) , and the PCR products were sequenced . To amplify and sequence the genome adjacent to the deletion site in bv/bv mice , “genome walks” were carried out using the PCR-based Genome Walker Universal kit ( Clontech Laboratories , Inc . ) and a gene specific primer that anneals to the penultimate exon in Srrm4 ( 5′-ACGGGACCTAAAGTATGGTGAGAAAG-3′ ) . For genotyping , the presence or absence of the bv mutation was detected by PCR using tail DNA extracts , and 2 sets of primer pairs ( wild-type allele: 5′-GGGAAGAGGTGGAGTATGTTG-3′ and 5′-CCTCGTGCTGGCATAGCTTTC-3′; bv allele: 5′-GAAAGAACCACAGCCCCGAGAA-3′ and 5′-CTGGGCAGGAGGGTACTTCTATAC-3 ) . The Myo7a-Srrm4 transgene was constructed by subcloning the mouse Srrm4-encoding cDNA downstream of the mouse Myo7a promoter and upstream of the SV40 polyadenylation site in the pSTEC-1 vector , using standard PCR and subcloning methods ( see PCR primers in Table S5 ) . The Myo7a-Srrm4 expression cassette was isolated from pSTEC-1 by restriction digestion , and sent to Xenogen Corp . for the production of transgenic mice . The ABR thresholds of mice were measured at P21–28 , using a previously described open-field system and broadband click stimuli [62] . The ability of mice to balance ( P70–80 ) was evaluated by measuring the time each mouse could remain on a fixed horizontal rod ( 1 . 8 cm in diameter ) following two training trials . Actin and Myo7a staining of whole-mount preparations of PFA-fixed cochlear and vestibular tissues was carried out as previously described [63] , using the following reagent and antibodies: Alexa-488 labeled phalloidin ( Invitrogen Corp . ) , rabbit anti-Myo7a antibody ( Proteus Biosciences , Inc . ) , and Alexa-594 labeled anti-rabbit IgG ( Invitrogen Corp . ) . Digoxigenin-labeled antisense exon 13 probe ( coding nucleotides 1521–1827 in Srrm4 ) , sense and antisense Srrm4 riboprobes ( coding nucleotides 23–188 ) were generated using the DIG RNA Labeling Mix ( Roche ) , and hybridized to inner ear samples of mice of various genotypes as described previously [17] . Inner ears of mouse embryos ( E16 . 5 ) were embedded in Tissue-Tek O . C . T . Compound ( Sakura Finetech , Inc . ) , frozen in liquid nitrogen , and cryosectioned . Sections were further processed for laser-capture microdissection using the Arcturus Histogene Frozen Section Staining kit ( Applied Biosystems ) . The manufacturer's staining protocol was modified in that RNAse inhibitor ( ProtectRNA , Sigma ) was added to every solution in the kit that contains more than 5% water . The vestibular macula was captured from the inner ear sections using the Laser Capture Microdissection system ( Pixcell II , Arcturus , Mountain View , CA ) . RNA was isolated from the captured tissue using the PicoPure RNA isolation kit . RNA was also extracted from the cerebellums of mice at P15 , using the Trizol Reagent ( Invitrogen ) . The cerebellar RNA was treated with DNase and further purified using the RNeasy mini kit ( Qiagen ) . RNA samples for microarray analysis were processed using the NuGEN WT-Ovation Pico RNA Amplification System , NuGEN WT-Ovation Exon Module , and NuGEN FL-Ovation cDNA Biotin Module . Samples were hybridized to Mouse Exon Junction Microarrays ( MJAY , Affymetrix Inc . ) . MJAY were scanned with an Affymetrix Model 7G upgraded scanner , and data were collected using GeneChip Operating Software . Raw microarray CEL files were imported into Partek Genomics Suite ( Partek , Inc . ) . Signal intensities for the probe sets were quantile normalized and median polished using Robust Multichip Average background correction . The signal intensities of exon probe sets were used to calculate the overall expression level of each gene represented in MJAY . Normalized probe-set intensities ( Inorm ) were calculated by dividing the background-corrected signal intensities of exon and exon-junction probe sets by the background-corrected gene-expression signal of the corresponding gene . The Inorm in the bv/bv and bv/+ samples was analyzed by two-tailed Student t-test . Probe sets with significantly different Inorm ( P<0 . 05 ) were queried against the Affymetrix annotation map file ( which contains alternative/constitutive annotations for each measured splicing event ) using simple Visual Basic for Application scripts , and probe sets that measure constitutive events were filtered out . The remaining probe sets were queried against the “SIB Alt-Splicing track” in the UCSC Genome Browser to identify and eliminate those that show either more than 50% identity with more than one gene or measure alternative promoter activity . The sequences of the remaining probe sets were queried against the mouse genome to identify those that measure the same splicing events . We required that probe sets targeting competing isoforms have opposite Inorm trends . RT-PCR was carried out essentially as described previously [64] . We defined an alternative exon as ‘differently spliced’ in the bv/bv and bv/+ samples if the RT-PCR data indicated that the inclusion rates for the exon were at least 1 . 5-fold different between the compared samples ( Table S6 contains the inclusion rates calculated based on the RT-PCR data shown in Figure 3D and 3F , Figures S4 , S7 , and S8 ) . Table S7 lists all primers that were used to generate the data shown in Figure 3D and 3F , Figures S4 , S7 , and S8 . For minigene-based validation of Srrm4-dependent splicing events , alternative exons and adjacent ∼300 bp intronic sequences were PCR amplified and subcloned into the exon trap pET-01 vector ( Mobitec , see primers in Table S5 ) . Mouse Srrm4wt and Srrm4bv were amplified by RT-PCR ( see primers in Table S5 ) from inner ear RNA and subcloned into the pcDNA3 . 1 expression vector . The Srsf1 expression construct ( Addgene plasmid 17990 ) has been described previously [65] . The minigines , Srrm4-encoding constructs , and the Srsf1-encoding plasmid were transfected into HEK293 cells using the Lipofectamine LTX and PLUS reagents ( Invitrogen ) , and RNA was extracted from the cells 24 hours later using the RNeasy mini kit ( Qiagen ) . RNA was reverse transcribed with Superscript III , and analyzed by RT-PCR using primers that annealed to the constitutive exons ( primers: 5′-CACTTGGTGGAAGCTCTCTACC-3′ and 5′-CCACCTCCAGTGCCAAGGTC-3′ ) . Site-directed mutagenesis of minigenes was carried out using overlap-extension PCR . The Srrm4 knock-down experiments were carried out in a transgenic zebrafish line developed by Haas and Gilmour [66] . In the neuromasts of these transgenic zebrafish , the claudin B promoter drives the expression of a membrane-tethered GFP ( Tg[CldnB-mGFP] ) . zSrrm4 expression was knocked down by injecting the transgenic zebrafish ( 2-cell stage ) with a previously described zSrrm4 MO [28] ( 5′-TTCTCCCAAAAGTACGCCAGCCATG-3′ from Gene Tools , Philomath , OR; 5 ng zSrrm4 morpholino/embryo ) . Since injection of 5 ng of the zSrrm4 MO led to non-specific toxicity , a p53-targeting MO ( 5′-GCGCCATTGCTTTGCAAGAATTG-3′ from Gene Tools; 5 ng/embryo ) was co-injected . 3 days after injection , zebrafish larvae were incubated with 3 µM FM1–43 dye for 30 s to label the mechanosensing hair cells in the neuromasts . The FM1–43 staining led to a bright green signal that was much more intense than the GFP signal of the CldnB-mGFP transgene . After staining with FM1–43 , the zebrafish were rinsed , anesthetized ( 0 . 02% 3-aminobenzoic acid ethyl ester ) , mounted in 3% methylcellulose , and photographed . The zSrrm4wt and zSrrm4bv mRNAs used for rescue experiments were generated using the mMessage mMachine kit ( Ambion ) and CS2+ plasmids that contained the zSrrm4wt and zSrrm4bv cDNAs ( see cloning primers in Table S5 ) . The zSrrm4wt and zSrrm4bv mRNAs were injected into zebrafish embryos ( 4-cell stage , 10 ng mRNA/embryo ) that had previously been injected with MOs targeting zSrrm4 and p53 . After the embryos had been maintained for 3 days , the mechanosensing hair cells were stained using the FM1–43 dye as described above . Nuclear fractions were isolated from the vestibular maculas of 44 wild-type and 44 bv/bv mice , on E16 . 5 , using the Nuclear Complex Co-IP kit ( Active Motif ) according to the manufacturer's instructions . The obtained nuclear fractions were treated with Enzymatic Shearing Cocktail ( Active Motif ) and centrifuged at 16 , 000 g for 15 min at 4°C . The pellets were dissolved in SDS sample buffer , boiled for 3 min , resolved by SDS-PAGE , and electroblotted onto nitrocellulose membranes . Following a blocking incubation step , goat anti-Srrm4 antibody ( sc-139291 from Santa Cruz Biotechnology Inc . ) diluted 1∶200 or rabbit anti-Lamin B1 antibody ( ab16048 from Abcam ) diluted 1∶5 , 000 was added to the membranes for 14 hours . After multiple washing steps , membranes were incubated with HRP-conjugated secondary antibodies ( anti-goat IgG and anti-rabbit IgG ) . Immunoblot signals were visualized using an Enhanced Chemiluminescence Detection System ( Pierce Biotechnology ) . Flag-tagged Srrm4 was subcloned into the pcDNA3 . 1 expression vector and transfected into HEK293 cells . 24 hours after transfection , the cells were harvested and resuspended in buffer DG ( containing 80 mM Potassium Glutamate , 0 . 1 mM EDTA , 10% glycerol , 0 . 01% NP40 , 0 . 1 mM PMSF , 1 mM DTT , 16 µg/ml chymostatin , 10 µl/ml protease inhibitor cocktail [from Sigma] , and 20 mM Hepes-KOH , pH 7 . 9 ) . The cells were then sonicated , incubated on ice for 15 min , and centrifuged at 16 , 000 g for 15 min at 4°C . The supernatant was collected and diluted ∼5-fold in buffer DG supplemented with 2 . 2 mM MgCl2 , 0 . 1 mg/ml tRNA ( Invitrogen ) and 1 U/ml RNase OUT . Mixtures of biotinylated RNA oligos ( 4 µg ) and NutrAvidin agarose resin ( ∼45 µl from Pierce Biotechnology ) were added to the cell lysates . Following 1 . 5-h incubation at 4°C , the resin was washed 6 times with buffer DG ( supplemented with 2 . 2 mM MgCl2 , 0 . 1 mg/ml tRNA , and 1 U/ml RNase OUT ) , resuspended in 45 µl of 2×SDS sample buffer , and boiled for 5 min . After a brief centrifugation , the supernatant fraction was resolved by SDS-PAGE , and protein was electroblotted onto nitrocellulose membranes . The membranes were blocked and incubated with 1∶1 , 000 dilution of a monoclonal anti-flag antibody ( Sigma ) for 14 hours . Following multiple washing steps , membranes were incubated with a HRP-conjugated secondary antibody ( anti-mouse IgG ) . Signal was visualized with an Enhanced Chemiluminescence Detection System . Complete microarray datasets have been deposited at Gene Expression Omnibus under SuperSeries accession number GSE33591 . | The identification of novel deafness-causing mutations has been instrumental in revealing unexpected mechanisms that are required for development of the sound- and gravity-sensing hair cells of the inner ear . The Bronx waltzer ( bv ) mouse is characterized by defects in hair-cell development , as well as by deafness and impaired balance . Here , we report on our identification of a mutation in the Ser/Arg repetitive matrix 4 ( Srrm4 ) gene as the source of these defects . The encoded protein , Srrm4 , belongs to a family of RNA splicing factors that regulate the inclusion of certain genetic information ( i . e . alternative exons ) into the transcribed RNA . We analyzed the molecular function of Srrm4 by comparing the exon composition of RNAs in the inner ear of bv and control mice . This approach revealed that , in the bv mice , specific alternative exons were omitted from protein-encoding RNAs . The affected transcripts shared two features: they contained a short sequence motif that was required for Srrm4-dependent splicing , and they encoded proteins that were related predominantly to secretion and neurotransmission . In addition , RNAs of a few gene expression regulators contained Srrm4-regulated exons . Our data suggest that Srrm4-dependent alternative splicing has a profound effect on the developmental program of hair cells . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"auditory",
"system",
"developmental",
"biology",
"gene",
"identification",
"and",
"analysis",
"genetics",
"molecular",
"genetics",
"biology",
"sensory",
"systems",
"neuroscience",
"genetics",
"and",
"genomics",
"cell",
"differentiation",
"gene",
"function"
] | 2012 | A Mutation in the Srrm4 Gene Causes Alternative Splicing Defects and Deafness in the Bronx Waltzer Mouse |
Viruses have evolved several strategies to modify cellular processes and evade the immune response in order to successfully infect , replicate , and persist in the host . By utilizing in-silico testing of a transmembrane sequence library derived from virus protein sequences , we have pin-pointed a nine amino-acid motif shared by a group of different viruses; this motif resembles the transmembrane domain of the α-subunit of the T-cell receptor ( TCRα ) . The most striking similarity was found within the immunodeficiency virus ( SIV and HIV ) glycoprotein 41 TMD ( gp41 TMD ) . Previous studies have shown that stable interactions between TCRα and CD3 are localized to this nine amino acid motif within TCRα , and a peptide derived from it ( TCRα TMD , GLRILLLKV ) interfered and intervened in the TCR function when added exogenously . We now report that the gp41 TMD peptide co-localizes with CD3 within the TCR complex and inhibits T cell proliferation in vitro . However , the inhibitory mechanism of gp41 TMD differs from that of the TCRα TMD and also from the other two known immunosuppressive regions within gp41 .
The transmembrane domains ( TMDs ) of the T-cell receptor ( TCR ) play an important role in the assembly of the receptor complex . The mosaic assembly of the TCR complex , composed of the TCR-α and β subunits and the invariant CD3 co-receptor chains , gives rise to one of the most remarkable and complex receptor structures . A unique characteristic of the multi-protein receptor assembly is the presence of nine potentially charged and conserved amino-acid residues within the transmembrane helices of both the TCR and the CD3 co-receptor [1]–[3] . It has been shown that one basic and two acidic transmembrane residues , within these nine residues , are required for the assembly of each of the three CD3 signaling dimmers ( δ-ε , γ-ε , ζ-ζ ) with the TCR [4] . In concert with the latter , Manolios et al have shown that the stable interactions between the TCRα and CD3 are localized to a short region within the transmembrane domain of TCRα . Moreover , they have shown that the charged amino acids arginine and lysine are essential for this interaction [5]–[7] . In addition , a nine amino-acid peptide derived from the TCRα transmembrane domain , which includes these two charged amino acids ( TCRα TMD , GLRILLLKV ) , interfered and intervened in the TCR function when added exogenously [8] . An important question is whether viruses have exploited this mechanism to interfere with the TCR complex assembly in order to modify T-cell function and thereby evade immune surveillance and enhance viral infection , replication and persistence . Recent studies with HIV support this notion . Similar to other enveloped viruses , HIV requires a fusion of the viral membrane and the cellular membrane to initiate infection . For that purpose it employs a fusion protein , the transmembrane gp41 ( gp41 ) that is non-covalently linked to the surface glycoprotein 120 ( gp120 ) . Together , gp120 and gp41 form the enveloped glycoprotein complex , gp160 , [9] which is embedded in the viral membrane as a trimmer [10] . The surface gp120 is primarily involved in binding to cellular receptors , whereas gp41 is anchored to the viral membrane and mediates membrane fusion . Gp41 is composed of cytoplasmic , transmembrane , and extra-cellular domains [11]–[15] . The extra-cellular domain ( ectodomain ) contains four major functional regions including a stretch of 16 hydrophobic residues , located at the N terminus , referred to as the “fusion peptide” ( FP ) [12] , [13] ( Figure 1 ) . During the initial step of T-cell infection , gp120 binds to the CD4 molecule , which is part of the T-cell receptor complex . The FP then anchors gp41 to the target membrane and initiates the fusion process [15]–[19] . However , it has been shown that anchoring gp41 to the target cell serves not only the purpose of supporting a successful fusion , but also suppresses the ability of the T-cell to be activated and proliferate . FP gains this activity by specifically binding to the TMD of TCRα and interfering with the assembly of the TCR complex; [20]–[23] . Another known immunosuppressive ( ISU ) region is composed of amino acids 583–599 , but its mode of action is not known [24] ( Figure 1 and Table 1 ) . Here , we tested in-silico a TMD sequence library derived from a group of different viruses and compared it to the TMD of TCRα . The most striking resemblance to the TCRα TMD , including the important arginine residue , was found within the immunodeficiency virus ( SIV and HIV ) glycoprotein 160 TMD ( gp41 TMD ) . This suggests a role for the TM domain in T-cell suppression as well . To test this hypothesis , we synthesized a peptide comprising the gp41 TM region and examined its immunosuppressive activity and a plausible mode of action . The results indicate that gp41 TMD co-localizes with CD3 within the TCR molecule and inhibits T-cell proliferation in-vitro . This effect is specific to the TCR complex , since T-cell activation by PMA/ionomycin is not affected by gp41 TMD . Interestingly , the mechanisms by which the other two immunosuppressive peptides express their activities are different: FP inhibits antigen-specific T-cell proliferation by specifically interacting with the TCRα subunit [23] , whereas the ISU inhibits T-cell proliferation induced by anti-CD3 and PMA/ionomycin [24] . Detailed understanding of the molecular interactions mediating the immunosuppressive activity of the gp41 TMD should facilitate the evaluation of its contribution to HIV pathology . Disassociated from HIV , however , the gp41 TMD molecule provides a novel mechanism for down regulating undesirable responses and might be used as an immunotherapeutic tool .
The nine amino-acid sequence , GLRILLLKV , derived from the TCRα TMD ( designated TCRα CP ) could interfere with TCR function [8] . In order to examine whether viral protein repertoires contain TMD sequences that are homologous to this immunosuppressive sequence , we took advantage of the Uniport Knowledge Base to construct a viral TMD sequence library . To do this , we performed a systematic pairwise alignment using the EMBOSS package : Needle global alignment [25] . Our results indicated that the top-ranked sequence belongs to the TMD of the SIV gp160 Envelope protein ( Figure 2A . I , Entry , accession number: Q8AIH5 ) . This TMD ( 21 aa ) contains a 9 aa sequence with 88 . 9% similarity and 66 . 7% identity to the TCRα CP . All the sequence entries ( n>3 ) belonging to a distinct species have been grouped into 265 TMD clusters ( over all 2874 sequence entries ) . Figure 2B displays the averaged normalized Z-score distribution of each cluster . Interestingly , the top-four ranked viral protein clusters were composed of the TMDs of HIV-1 gp160 ( namely HIV-1 gp41 ) , Feline immunodeficiency virus ( FIV ) gp150 and two TMDs within the latent membrane protein 1 ( LMP1 ) from Epstein-Barr virus ( EBV ) . Despite the fact that a single sequence entry within the SIV gp160 cluster ( Accession number: Q8AIH5 ) was the top ranked alignment , the SIV gp160 cluster was ranked lower . To statistically test the observation that HIV-1 gp41 TMD cluster was significantly ranked higher , a Wilcoxon Rank Test was performed . As shown in Figure 2C , the HIV-1 gp41 TMD cluster was significantly higher ranked than all the other clusters that were not included within this top four group ( FIV gp150 and EBV LMP1 ) . Our in-silico analysis revealed that four viral protein clusters were significantly ranked in comparison with all the other clusters within our TMD library . As HIV/T-cell interactions are central to HIV infection , we reasoned that these findings may aid in further characterizing the process of HIV infection . The initial step was to examine whether the gp41 TM region , like the TCRα CP , is able to inhibit T-cell proliferation in-vitro . For that purpose we prepared lymph node cells from mice immunized with MOG 35–55/CFA ( MOG 35–55-immunized ) and studied their T-cell responses to the MOG 35–55 peptide . The MOG 35–55 peptide induced strong proliferative responses and cytokine release from T-cells in the draining lymph node cells of MOG 35–55-immunized mice . We then synthesized a peptide derived from one of the top ranked HIV-1 gp41 TM entries ( Figure 2AII , Accession number: P03378 , Table 1 ) . Next , we incubated the peptide in the presence of MOG p35–55 antigen with a T-cell line that is specific to MOG p35–55 and determined the proliferative responses after 3 days of incubation , using a H3-thymidine uptake assay . Figure 3 shows that the peptide exhibited dose-dependent inhibition of T-cell proliferation . The concentration of gp41 TMD in the cell membrane is very small . However , gp41 contains two more immunosuppressant regions ( FP and ISU ) which probably act synergistically and rapidly with the gp41 TMD . In addition , these regions are directed to the TCR complex after binding of gp120 to the CD4 and co-receptors . In contrast , the gp41 TMD peptide investigated here lacks a receptor , and in addition it is only part of the gp41 complex . This requires much higher concentrations of the peptide to obtain a significant immunosuppressant activity . To determine whether gp41 TMD can also inhibit T-cell activation other than that induced by APC presentation of specific antigen , we tested the effect of gp41 TMD on T-cell activation induced by a mitogenic monoclonal antibody to CD3 . TCRα CP served as a control peptide . Gp41 TMD unlike TCRα CP [8] , inhibits the activation of T-cells by mitogenic anti-CD3 ( Figure 4 ) . Mitogenic anti-CD3 antibodies activate CD3 signaling regardless of the presence or absence of TCR [22] . These results suggest that gp41 TMD interacts with the CD3 , although we cannot rule out other , indirect interactions . To learn whether gp41 TMD can also inhibit T-cell activation other than that induced by APC presentation of specific antigen or mitogenic monoclonal antibody to CD3 ( the TCR complex ) , we tested the effect of gp41 TMD on T-cell activation induced by PMA/ionomycin . Gp41 TMD , like TCRα CP , did not inhibit the activation of T-cells by PMA/ionomycin ( Figure 5 ) . PMA/ionomycin activates the T-cell downstream to the membrane regardless of the presence or absence of the TCR [3] . These findings suggest that gp41 TMD inhibits T-cell activation by interfering with the TCR and CD3 proper function . We utilized confocal microscopy to further support our findings that the gp41 TMD peptide interacts specifically within the TCR complex and does not home to the cytoplasm . It has been reported that the TCR , CD3 , and CD4 receptors , among other components , are localized in micro-domains in the membranes of activated CD4+ T cells [2] . Using Sy5 labeled antibodies to the CD3 molecules , we examined the localization and distribution of the CD3 molecules in the membranes of resting and activated T-cells . In resting T-cells , the CD3 molecules were found in micro-domains all around the T-cell , while the distribution of the CD3 molecules resembled a capping shape in activated T-cells ( Figure 6A ) . In order to examine the distribution of gp41 TMD in the membrane of resting and activated mice T cells , we used gp41 TMD peptide conjugated to the fluorescent probe 4-chloro-7-nitrobenz-2-oxa-1 , 3-diazole ( NBD; gp41 TMD-NBD ) . Rather then uniformly labeling the T-cell membrane , the gp41-TMD NBD showed a heterogeneous membrane distribution and a capping shape on the membrane of activated T-cells . This distribution in membrane domains contrasted with that of a control membrane-active amphipathic peptide conjugated to NBD ( AMP-NBD ) , which demonstrated a uniform distribution and no capping shape on activated T-cell membranes ( Figure 6A ) . In contrast to the AMP-NBD control peptide , the gp41-TMD conjugates exhibited the same distribution pattern as the CD3 molecules . In the next step , we demonstrated that the gp41 TMD not only showed the same distribution pattern but also had a significantly higher percentage of co-localization with the CD3 molecules ( 58 . 8% , Figure 6B ) compared to the AMP-NBD control peptide ( 27% , Figure 6C ) ( p<0 . 05 ) . We labeled the TCRα CP peptide with NBD as the donor fluorophore , and both gp41 TMD and TCRα CP peptides with Rho-TAMRA as the acceptor fluorophore . In addition , we also labeled an unrelated transmembrane peptide derived from the E . coli aspartate receptor ( TAR/PS ) as a control hydrophobic peptide ( Table 1 ) . We then measured the fluorescence energy transfer ( FRET ) between fluorescently labeled TCRα CP-NBD peptide and gp41 TMD – Rho ( 7A ) , TCRα CP-Rho ( 7B ) or TAR/PS – Rho ( 7C ) . The assay was performed in a model lipid environment of large unilamellar vesicles ( LUV ) composed of PC phospholipids . Four ratios of Rho-peptide∶NBD-peptide were used: 0∶4 , 1∶4 , 2∶4 and 3∶4 . The TCRα CP-NBD peptide showed about ∼50% energy transfer in the presence of the gp41 TMD - Rho peptide at an acceptor-to-lipid ratio of 1∶1000 ( Figure 7A ) , indicating an interaction between the two peptides . When we examined the energy transfer of TCRα CP-NBD peptide in the presence of TCRα CP-Rho peptide , there was lower energy transfer ( ∼37% , 7B ) . The TAR/PS – Rho control peptide did not show energy transfer ( 7C ) .
During evolution , viruses have evolved various strategies to modify cellular processes and evade immune responses , which allow them to successfully infect , replicate , and persist in the host . Here we utilized bioinformatic tools to identify a new region corresponding to the TMD of the HIV-1 gp41 ENV glycoprotein that enables the virus to inhibit T-cell proliferation . We identified viruses that contain sequences within their TM region that are similar to TCRα CP , an immunosuppressant domain within the TCRα TMD ( Figure 2 ) . The results revealed four viral protein TMD clusters , namely , HIV gp160 , FIV gp150 and two TMDs within EBV LMP1 , that were significantly ranked in comparison with all other TMDs within our library ( Figure 2C ) . Dukers et al . have reported that purified recombinant EBV LMP1 suppress activation of T-cells . They found that a 7-amino acid peptide ( LALLFWL ) within the TMD1 of EBV LMP1 has immunosuppressant activity [26] . However , this region does not overlap with the EBV LMP1 TMD1 cluster that emerged from our bioinformatic analysis ( GLALLLLLL ) . Interestingly , within our top four highest ranked clusters we identified another EBV LMP1 TMD region within the TMD3 ( GLGLLLLMV ) . However , since HIV/T-cell interactions are central to HIV infection , it was most interesting to investigate whether the HIV-1 gp41 may be exploited by HIV to modulate and interfere with T-cell activation . Gp41 TMD is one of the most conserved regions within the gp41 sequence [27] . Studies have suggested that this region is involved in many important biological functions such as anchoring of the Env glycoprotein in both viral and cellular membranes [28] , cell-cell fusion [29] and in the mixing and fusion of phospholipids between two lipid vesicles [30] . The fact that T-cells are targets of HIV-1 raised the question whether this TMD is also exploited by the virus to modulate the T-cell immune response . Strikingly , we found that a synthetic peptide resembling the gp41 TMD is able to co-localize with the TCR complex and suppress T-cell proliferation in vitro , and , therefore , may play a role in infection as an immunosuppressive region . It has been shown previously that the HIV-1 gp41 has two other immunosuppressive regions: FP [21]–[23] and ISU [24] . The mechanisms by which these two immunosuppressive regions express their activity are different: FP inhibits antigen-specific T-cell proliferation by specifically interacting with the TCRα [23] , whereas ISU inhibits T-cell proliferation induced by anti-CD3 and PMA/ionomycin [24] . To better understand the mechanism by which the TM region exerts its immunosuppressive activity , we examined the abilities of the gp41 TM peptide to inhibit T-cell proliferation when triggered through different signal-transduction cascades . Interestingly , we found that gp41 TMD , unlike the FP [21] , [23] or TCRα CP [8] , inhibits T-cell activation induced by mitogenic antibodies specific for the CD3 molecule ( Figure 4 ) . This interaction activates T cells regardless of their TCR and is downstream of the TCR activation pathway . This is despite the fact that this TM peptide has high homology to TCRα CP as found through bioinformatic analysis . To follow the mode of action underlying the latter inhibition , we further examined the interaction between gp41 TMD and TCRα CP by using FRET . The data summarized in Figure 7 reveal a direct interaction between these two TMDs . Note that gp41 TMD did not inhibit T-cell activation triggered by PMA/ionomycin . Overall the data support the notion that gp41 TMD most likely interacts with the TCR complex in a way that interferes with both the TCR and CD3 proper function . Our confocal microscopy results indicate that the gp41 TMD-mediated interference is likely to occur within the cell membrane since the peptide is not detectable elsewhere ( Figure 6 ) . As discussed previously , the immunosuppressive domain ( ISU ) in gp41 ( Figure 1 and Table 1 ) is capable of inhibiting T-cell activation [24] . Similarly to gp41 TMD , the ISU inhibits T-cell activation triggered by CD3-specific antibodies [24] . However , in contrast to gp41 TMD , ISU can also inhibit T-cell activation triggered by PMA/ionomycine [31] . Thus , ISU simultaneously targets protein kinase C activity [32] and the events related to T-cell activation that occur within the cell membrane [33] . In contrast , gp41 TMD only targets T-cell activation within the cell membrane . Hence , gp41 equips HIV-1 with at least 3 different inhibitors of T cell activity: gp41 TMD , FP and ISU . Each of them uses a different mode of action , which we propose can function at different time points during viral infection . In addition to the inhibitors described above , it has been also reported that HIV-1 impairs T-cell activation and immunological synapse formation between infected lymphocytes and antigen-presenting cells ( APCs ) by Nef ( negative factor ) mediation [34]–[36] . One may reason that inhibitors of T-cell activation in general , such as that mediated by Nef and gp41 TMD in particular , may also inhibit the successful replication of the virions , since the magnitude of viral replication in CD4+ cells is directly linked to their activation state [37] . However , despite the fact that infection is much more efficient in activated T-cells , replication of HIV-1 and simian immunodeficiency virus ( SIV ) in-vivo occurs in T-cells that display a low activation profile as well [38]–[41] . Since we used a high concentration of gp41 TMD peptide ( 5–50µg/ml ) , we believe that the effect in vivo might be a modulation of the T-cell activity rather than total inhibition . It has been reported that lymphocyte activation , through TCR ligation or other stimuli , frequently leads to homeostatic programmed apoptosis [42] , [43] . Therefore , it is not unlikely that the gp41 ENV expression in vivo may lead to apoptosis due to TM-TM interactions of the gp41 and the TCR complex that modulate the appropriate signaling . The rapid death of the infected cells will most certainly limit the production of the virus . For that reason , in order to ensure its spread , the virus must establish a balance between the apoptosis-prone activation state and the replication-unfavorable environment of resting cells . Although it is still unclear how these target cell effects are related to the replication efficiency of the virus , various viral proteins are known to be involved in these phenomena . For example , Nef and Tat ( transcriptional transactivator ) are known to modulate T-cell activity , which is likely to facilitate viral replication [35] , [44]–[46] . Therefore , the fact that gp41 TMD inhibits or modulates T-cell activity does not necessarily imply that viral replication is decreased . The presence of gp41 on the cell membrane prior to and during budding may interfere with the ability of infected T cells to proliferate , thus allowing the virus to harness most cellular resources for its own needs . Hence , we suggest here that HIV-1 can utilize gp41 TM inhibitory mechanism to facilitate viral replication and to enable an efficient budding of the virions . As we reported earlier , HIV-1 uses the FP in the N terminus of gp41 for anchoring to the target cell and to inhibit T cell activation [21] , [23] . We show here that gp41 TMD interacts with CD3 and TCRα . Therefore , it is likely that gp41 TMD is also located within the TCR-FP region during the membrane fusion phase and for that reason may have a role in an efficient fusion process and as an alternative mechanism to ensure that the T cell will not be activated during this central step of infection . However , further experiments are required to better understand the mechanisms by which the virus modulates signaling within the infected cell . Note that an interesting theoretical model published recently try to explain TCR modulation by the appearance of two electro-positively charged residues within virus fusion peptides causing electrostatic disturbance within the complex [47] . It has been argued recently that HIV-1 and microvsicles from T cells share a common glycome , which may indicate a common origin [48] . Our in silico results pinpoints a high similarity between the gp41 TM region and the TCRα TMD region ( CP domain ) . In addition , the similarity between gp41 TMD and TCRα CP was found to be of far greater significance than that between other virsues' TMD and TCRα CP . Therefore , the question regarding the origin of this sequence is still unclear . In summary , the present study demonstrates a new weapon that HIV-1 uses to penetrate into the host cell and modulates its immune response . This immunosuppressive activity of gp41 TMD might be exploited in the future for the design of new therapies for autoimmune disease .
All animal experiments were conducted at the Weizmann Institute of Science and approved by the Weizmann Institutional Animal Care and Use Committee ( IACUC ) according to the Israel law and the National Research Council guide ( Guide for the Care and Use of Laboratory Animals 2010 ) . Rink amide MBHA resin and 9-fluorenylmethoxycarbonyl ( Fmoc ) amino acids were purchased from Calibochem-Novabiochem AG ( Switzerland ) . Other reagents used for peptide synthesis include N , N-diisopropylethylamine ( DIEA , Aldrich ) , dimethylformamide , dicheloromethane , and piperidine ( Biolab , IL ) . Egg phosphatidylcholine ( PC ) was purchased from Lipid Products ( South Nutfield , UK ) . 4-chloro-7-nitrobenz-2-oxa-1 , 3-diazole fluoride ( NBD-F ) and rhodamine-N-hydroxysuccinimide ( Rho-N ) were purchased from Molecular Probes ( Junction City , OR , USA ) . The myelin oligodendrocyte glycoprotein ( MOG ) p35–55 antigen used for the specific activation of the T-cell line was synthesized using the Fmoc technique with an automatic multiple peptide synthesizer ( AMS 422 , ABIMED , Langenfeld , Germany ) . The Hamster anti-mouse anti-CD3 antibody was collected by trichloroacetic acid ( TCA ) precipitation from 2C11 hybridoma supernatant . Peptides were synthesized using the Fmoc solid phase method on Rink amide resin ( 0 . 68 meq/gm ) , as previously described [49] . The synthetic peptides were purified ( greater than 98% homogeneity ) by reverse phase high performance liquid chromatography ( RP-HPLC ) on a C4 column using a linear gradient of 30–70% acetonitrile in 0 . 1% trifluoroacetic acid ( TFA ) for 40 minutes . The peptides were subjected to amino acid and mass spectrometry analysis to confirm their composition . To avoid aggregation of the peptides prior to their use in the cell culture assays , the stock solutions of the concentrated peptides were maintained in dimethyl sulfoxide ( DMSO ) . The final concentration of DMSO in each experiment was lower than 0 . 25% vol/vol and had no effect on the system under investigation . For NBD-F fluorescent labeling , resin-bound peptides were treated with NBD-F ( 2-fold excess ) dissolved in dimethyl formamide ( DMF ) , leading to the formation of resin-bound N-terminal NBD peptides [50] . After 1 h , the resins were washed thoroughly with DMF and then with methylene chloride , dried under nitrogen flow , and then cleaved for 3 h with TFA 95% , H2O 2 . 5% , and triethylsilane 2 . 5% . For Rho-N fluorescent labeling , the Fmoc protecting group was removed from the N-terminus of the resin-bound peptides by incubation with piperidine for 12 min , whereas all the other reactive amine groups of the attached peptides were kept protected . The resin-bound peptides were washed twice with DMF , and then treated with rhodamine-N-hydroxysuccinimide ( 2-fold excess ) , in anhydrous DMF containing 2% DIEA , leading to the formation of a resin-bound N-rhodamine peptide . After 24 h , the resin was washed thoroughly with DMF and then with methylene chloride , dried under nitrogen flow , and then cleaved for 3 h with TFA 95% , H2O 2 . 5% , and triethylsilane 2 . 5% . The labeled peptides were purified on a RP-HPLC C4 column as described above . Unless stated otherwise , stock solutions of concentrated peptides were maintained in DMSO to avoid aggregation of the peptides prior to use . Thin films of PC were generated after dissolving the lipids in a 2∶1 ( v/v ) mixture of CHCL3/MeOH and drying them under a stream of nitrogen gas while rotating them . The films were lyophilized overnight , sealed with argon gas to prevent oxidation of the lipids , and stored at −20°C . Before the experiments , films were suspended in the appropriate buffer and vortexed for 1 . 5 min . The lipid suspension underwent five cycles of freezing-thawing and extrusion through polycarbonate membranes with 1- and 0 . 1-µm diameter pores to create large unilamellar vesicles . The FRET experiments were performed by using NBD and Rho labeled peptides . Fluorescence spectra were obtained at room temperature , with excitation set at 467 nm ( 10-nm slit ) and emission scan at 500–600 nm ( 10-nm slits ) . In a typical experiment , a NBD-labeled peptide was added first from a stock solution in DMSO ( final concentration 0 . 1 µM and a maximum of 0 . 25% ( v/v ) DMSO ) to a dispersion of PC LUV ( 100 µM ) in PBS . This was followed by the addition of Rho labeled peptide in several sequential doses ranging from 0 . 025 µM to 0 . 075 µM ( stock in DMSO ) . Fluorescence spectra were obtained before and after addition of the Rho labeled peptide . The fluorescence values were corrected by subtracting the corresponding blank ( buffer with the same vesicles concentration ) . The statistical analysis was performed using ANOVA for the pick measurements at 535nm ( n = 3 , * p<0 . 05 ) . Resting MOG35–55 T-cells or after activation for 72 h with MOG35–55 and APC were blocked with 1% BSA at room temperature to block non-specific binding . After 30 min the cells were washed and divided into aliquots containing 100 , 000 cells per 100 µl , and either gp41 TMD or a control membrane-binding peptide ( AMP-scr ) was added ( final concentration of 2µM ) for 1 h at 37°c . The cell were then washed and labeled with the tested antibody for 25min at RT followed by biotin-conjugated anti-hamster IgG 25min at RT and Streptavidin PE-Cy-5 10min at RT ( all from eBioscience San-Diego CA , USA ) . Anti-hamster IgG followed by Streptavidin-Cy-5 served as a measure of non-specific binding . AMP peptide served as low TCR affinity control . The cells were analyzed by confocal fluorescence microscopy using Lab-Tek 8 chambers cover-glass ( nunc ) with living cells . The labeled cell samples were observed under a fluorescence confocal microscope . PE-Cy-5 excitation was done with HeNe laser 633nm ( emission data was collected with filter BA660IF , 660nm long pass ) . NBD excitation was done with Ar laser 488nm ( emission data was collected from 505–525nm ) . In order to quantify the co-localization percentage we utilized the “Simple PCI” software . The co-localization percentage was calculated as described below: T-cells were plated onto round 96-well plates in medium containing RPMI-1640 supplemented with 2 . 5% fetal calf serum ( FCS ) , 100 U/ml penicillin , 100 µg/ml streptomycin , 50µM 2β-mercaptoethanol , and 2mM L-glutamine . We used 12×104 cells of the T-cell line specific to MOG p35–55 , 5×105 irradiated ( 3000 rad ) spleen cells ( APC ) , and 10 µg/ml of MOG p35–55 were added to each well . In addition , peptide corresponding to the gp41 TM region was added . Each determination was made at least in triplicate . In order to exclude interaction between the examined peptides and the MOG p35–55 antigen , we initially added the MOG p35–55 antigen to the APCs in a test tube , and in a second test tube we added the examined peptides to the T-cells . After 1 hour , we mixed the APCs with the T-cells and incubated them for 72h in a 96 well round bottom plate . For some experiments , T cells were activated with immobilized anti-CD3 antibodies [51] or PMA/ionomycin as described [3] . After 72 hours , at 37°C in a 7 . 5% CO2 humidified atmosphere , the T-cells were pulsed with 1µCi ( H3 ) thymidine , with a specific activity of 5 . 0 Ci/mmol , for 7 hours , and ( H3 ) thymidine incorporation was measured using a 96-well plate beta-counter . The mean cpm ± Standard Deviation ( SD ) was calculated for each quadruplicate or more . The results of T-cell proliferation experiments are shown as the percentage of T-cell proliferation inhibition triggered by the antigen in the absence of gp41 TM peptide . The statistical analyses were performed using ANOVA . C57BL/6J mice were purchased from Harlan Olac ( Bicester UK ) . The mice were maintained in a specific pathogen-free facility . To evaluate the occurrence of TCRα CP-like TMDs in viruses , a dataset of putative viral TMDs was constructed based on the viral sub-division within the Uniport Knowledge Base , consisting on the intensively annotated Swiss-Prot database ( version 57 . 10 , total of 29 , 252 entries ) [52] . All sequences containing at least one TM annotation in the FT field were extracted from the dataset to create a library of TM viral domains , where every entry is composed from a distinct putative TMD within a protein . Overall , the library contained 6175 entries distinct at the sub-species/variants level . Entries belonging to the same taxonomic species were grouped into clusters which contained multiple sequences derived from several sub-species or variants . For our statistical analysis ( Wilcoxon Rank Test ) only clusters in which n>3 were used ( overall 2874 entries were grouped into 265 clusters ) . Taxonomic clustering of results was conducted according to the tax ID lineage of each distinct entry . In the next step , a pairwise alignment of the TCRα CP was performed against each of the dataset sequences utilizing the EMBOSS package Needle pairwise global alignment [25] at the ( http://www . ebi . ac . uk/Tools/emboss/align/index . html ) server . Alignments parameters were set using the Blosum40 matrix with gap opening/cost of 10/10 respectively . Results were ranked according to the clusters' Z-score ( normalized by the mean and standard deviation of the 265 clusters alignment scores ) [53] and were analyzed using Matlab software ( MathWorks , Natick , Mass ) . Statistical significance was determined according to the Benjamini-Hochberg method ( E ( FDR ) <0 . 05 ) [54] . | HIV uses several mechanisms that allow it to evade immune control , in order to successfully infect , replicate , and persist in the host . Here we report a new mechanism . We utilized bioinformatics and identified a region within the transmembrane domain ( TMD ) of the envelop proteins of viruses that has high similarity with the α subunit of the T-cell receptor ( TCR ) TMD . A striking similarity was found within the immunodeficiency virus ( SIV and HIV ) glycoprotein 41 ( gp41 ) . TCR TMDs play an important role in the assembly of the receptor complex composed of the TCR subunits and the CD3 co-receptor chains . We show that a synthetic peptide derived from gp41 TMD co-localizes with CD3 and inhibits T-cell proliferation in vitro . Biophysical studies suggest a specific interaction between gp41 TMD and the TMD of the TCRα subunit . Importantly , the inhibitory mechanism of gp41 TMD differs from that of the other two known immunosuppressive regions within gp41 . Overall , the present study demonstrates a new weapon that HIV-1 uses to penetrate into the host cell and modulates its immune response . Disassociated from HIV , however , HIV TMD molecule provides a novel mechanism for down regulating undesirable responses and might be used as a new therapy for autoimmune diseases . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"infectious",
"diseases/hiv",
"infection",
"and",
"aids",
"immunology/immune",
"response",
"microbiology/immunity",
"to",
"infections"
] | 2010 | HIV-1 gp41 and TCRα Trans-Membrane Domains Share a Motif Exploited by the HIV Virus to Modulate T-Cell Proliferation |
Ethosuximide is a medication used to treat seizure disorders in humans , and we previously demonstrated that ethosuximide can delay age-related changes and extend the lifespan of the nematode Caenorhabditis elegans . The mechanism of action of ethosuximide in lifespan extension is unknown , and elucidating how ethosuximide functions is important for defining endogenous processes that influence lifespan and for exploring the potential of ethosuximide as a therapeutic for age-related diseases . To identify genes that mediate the activity of ethosuximide , we conducted a genetic screen and identified mutations in two genes , che-3 and osm-3 , that cause resistance to ethosuximide-mediated toxicity . Mutations in che-3 and osm-3 cause defects in overlapping sets of chemosensory neurons , resulting in defective chemosensation and an extended lifespan . These findings suggest that ethosuximide extends lifespan by inhibiting the function of specific chemosensory neurons . This model is supported by the observation that ethosuximide-treated animals displayed numerous phenotypic similarities with mutants that have chemosensory defects , indicating that ethosuximide inhibits chemosensory function . Furthermore , ethosuximide extends lifespan by inhibiting chemosensation , since the long-lived osm-3 mutants were resistant to the lifespan extension caused by ethosuximide . These studies demonstrate a novel mechanism of action for a lifespan-extending drug and indicate that sensory perception has a critical role in controlling lifespan . Sensory perception also influences the lifespan of Drosophila , suggesting that sensory perception has an evolutionarily conserved role in lifespan control . These studies highlight the potential of ethosuximide and related drugs that modulate sensory perception to extend lifespan in diverse animals .
Pharmacological compounds that extend lifespan could delay the progression of age-related degenerative changes and age-related illnesses such as Alzheimer's disease and cardiovascular disease . In addition , the characterization of drugs that extend lifespan can elucidate endogenous mechanisms involved in lifespan determination , since the targets of these drugs are likely to influence normal aging . The short lifespan and rapid aging of invertebrates makes them powerful models for the identification of drugs that extend lifespan and for the characterization of the mechanism of action of these drugs [1]–[3] . The free-living soil nematode Caenorhabditis elegans has been a leading system for studying genetic and pharmacologic influences on lifespan . Four categories of compounds have been reported to extend C . elegans lifespan: a variety of antioxidant compounds [4]–[9]; complex mixtures derived from plants [10] , [11]; resveratrol , a potential modulator of Sir2 activity [12] , [13]; and medications such as heterocyclic anticonvulsant medications that may act by affecting neural activity [14]–[17] . Compounds that extend the lifespan of vertebrates have not been well characterized . However , a recent report showing that resveratrol can extend the lifespan of a short-lived fish suggests that compounds that extend invertebrate lifespan may be relevant to vertebrate biology [18] . By screening 19 drugs from different structural and functional classes that are FDA-approved for human use , Evason et al . ( 2005 ) discovered that ethosuximide can extend the lifespan of C . elegans [14] . Ethosuximide is a small , heterocyclic ring compound of the succinimde class that is approved for human use as an anticonvulsant [19] . Trimethadione is a structurally related anticonvulsant that is a member of the oxazolidinedione class , and trimethadione also extends C . elegans lifespan [14] . Ethosuximide is commonly used in clinical practice whereas trimethadione is rarely used due to the potential for adverse side effects . Ethosuximide extended the mean adult lifespan of wild-type animals grown on agar dishes by 17% [14] . The effect is dose dependent , and at high doses ethosuximide causes toxicity . In addition , ethosuximide extends the span of time that animals display fast body movement and pharyngeal pumping , demonstrating that ethosuximide delays age-related functional declines in addition to extending lifespan . Ethosuximide has been shown to affect the activity of multiple ion channels in vertebrate cultured cells , including T-type calcium channels [20]–[22] . The relationship between these activities in cultured cells and the anticonvulsant activity in whole animals has yet to be defined fully . Furthermore , the mechanism of action for lifespan extension in worms is not well characterized . To elucidate the mechanism of action of ethosuximide , we conducted a genetic screen for mutations that cause resistance to the drug . Screening for drug resistance is a well-established approach in C . elegans [23] . A mutation can cause resistance to a drug for several different reasons such as altering the molecular target of the drug , the cellular target of the drug , or the metabolism of the drug . An example of resistant mutants that identified the molecular target include the ivermectin resistant locus avr-15 , which encodes a glutamate gated chloride channel that binds ivermectin [24] and the α-amanitin resistant locus ama-1 , which encodes a RNA polymerase that binds α-amanatin [25] , [26] . An example of a gene that affects drug metabolism and was identified in a screen for drug resistant mutants is nrf-6 , which functions in intestinal cells to promote fluoxetine sensitivity [27] . Here we analyze the mechanism of action of ethosuximide in lifespan extension and show that it is related to the activity of chemosensory neurons , indicating that these neurons are the cellular target of ethosuximide in lifespan extension . C . elegans and other animals live in complex environments that can change rapidly , and therefore these animals have evolved the ability to respond quickly to changing conditions . The ability to perceive chemosensory cues and mount behavioral responses enables an animal to adjust to environmental changes . C . elegans uses ciliated chemosensory neurons located in the anterior and posterior of the animal to respond to numerous soluble and volatile cues [28]–[30] . Mutations that cause defects in cilia structure or sensory signaling within chemosensory neurons disrupt chemotaxis towards soluble and volatile chemicals [29]–[31] . These chemosensory neurons also influence adult lifespan , since mutations that cause defects in the structure of cilia or mutations that cause defects in sensory signaling can extend lifespan [32] , [33] . Furthermore , laser ablation of the chemosensory neurons ASI , AWA , and AWC , separately or together , can increase adult lifespan [34] . These results indicate that the activity of certain chemosensory neurons promotes a short lifespan . To characterize the mechanism of action of ethosuximide , we conducted a genetic screen and identified mutations in two genes , che-3 and osm-3 , that cause resistance to ethosuximide-mediated toxicity . Mutations in che-3 and osm-3 cause defects in overlapping sets of chemosensory neurons and can extend lifespan [32] . These findings indicate that ethosuximide extends lifespan by inhibiting a subset of chemosensory neurons . Here we present results that strongly support this model . Ethosuximide treated wild-type animals displayed numerous phenotypic similarities with mutants that have chemosensory defects , indicating that ethosuximide inhibits chemosensory function . Importantly , the long-lived osm-3 mutants did not respond to lifespan extending doses of ethosuximide . These studies demonstrate a novel mechanism of action for a lifespan extending drug and demonstrate the potential of pharmacologically targeting the sensory system as a means to extend animal lifespan .
Studies using vertebrate cultured cells have led to the proposal that T-type calcium channels may be the molecular target of ethosuximide in controlling seizures [20] , [22] , [35] . To determine if ethosuximide inhibits T-type calcium channels to extend lifespan , we analyzed a null mutation in the gene encoding the C . elegans orthologue of the mammalian T-type calcium channel , cca-1 ( gk30 ) [36] , [37] . If ethosuximide inhibits T-type calcium channels to extend lifespan , then we predict that ( 1 ) a cca-1 loss-of-function mutant will be long-lived and ( 2 ) a cca-1 loss-of-function mutant will not respond to the lifespan extension caused by ethosuximide . cca-1 ( gk30 ) mutants displayed a mean adult lifespan of 15 . 2 days , which was not significantly different from the 16 . 1 day mean adult lifespan of wild-type animals ( Figure 1A ) . cca-1 ( gk30 ) hermaphrodites treated with ethosuximide displayed a robust lifespan extension , indicating that ethosuximide does not require T-type calcium channels to extend C . elegans lifespan ( Figure 1A ) . In addition to extending lifespan , ethosuximide also stimulates the rate of egg-laying [14] . To determine if ethosuximide requires T-type calcium channels to stimulate egg-laying , we utilized a well-described assay that involves counting the eggs laid by a young adult hermaphrodite that is transferred from a Petri dish with abundant food to M9 liquid culture with no food [38] . We analyzed a second putative cca-1 null allele , cca-1 ( ad1650 ) [36] , [37] . Wild-type and cca-1 ( ad1650 ) hermaphrodites transferred to M9 buffer laid zero eggs in 90 minutes ( Figure 1B ) , demonstrating that egg-laying is strongly inhibited by these conditions . Wild-type and cca-1 ( ad1650 ) hermaphrodites displayed a similar , dose-dependent increase in egg-laying in response to ethosuximide ( Figure 1B ) , indicating that T-type calcium channels are not required for ethosuximide stimulated egg-laying . Collectively , these results suggest that ethosuximide modulates lifespan and egg-laying in C . elegans by acting on molecular targets distinct from the T-type calcium channel CCA-1 . The identification and characterization of mutants that are resistant to the activity of a drug has been a useful approach for elucidating the mechanism of drugs in C . elegans [24] , [39]–[41] . The analysis of candidate genes , including cca-1 and genes that influence longevity [14] , did not clearly identify the target of ethosuximide . Therefore , we conducted a genetic screen for mutations that cause resistance to ethosuximide . To identify a phenotype for conducting a genetic screen that is more suitable than lifespan , we performed a dose response analysis . The optimal dosage for lifespan extension by ethosuximide was 2–4 mg/ml in the culture media [14] . Concentrations of ethosuximide greater than 4 mg/ml caused dose dependent larval lethality ( data not shown ) . The minimum dosage that caused fully penetrant lethality was 12 mg/ml ethosuximide . The larval lethal phenotype is ideal for the isolation of resistant mutants , because it is possible to screen large numbers of mutant animals for survival in 12 mg/ml ethosuximide . We mutagenized wild-type animals with EMS or ENU , screened approximately 200 , 000 haploid genomes for mutants that survived when cultured on 12 mg/ml ethosuximide , and isolated 48 mutants . Forty-two mutant strains were successfully backcrossed to wild-type animals , indicating that ethosuximide resistance in these strains is caused by a mutation in one gene . In all cases the mutant strains displayed partially penetrant survival in 12 mg/ml ethosuximide ( Table 1 and data not shown ) , indicating that these mutations cause a shift in the dose response to ethosuximide and do not abrogate sensitivity to this drug . Here we describe six of these mutations that represent two complementation groups . To characterize the mutations that cause ethosuximide resistance , we positioned these mutations on the C . elegans genetic and physical maps . The mutations am178 , am165 , and am162 exhibited tight linkage to a single nucleotide polymorphism ( SNP ) marker positioned at the center of Chromosome I ( see Materials and Methods ) . Three-factor mapping positioned am178 between dpy-5 and unc-75 on Chromosome I ( Figure 2 ) . High resolution mapping positioned am178 between SNP markers uCE1-952 and snp_T01G9[1] , a ∼250 kB interval on Chromosome I ( Figure 2 ) . To identify the gene affected by am178 , we analyzed candidate genes in the interval by measuring the ability of existing mutations to cause resistance to ethosuximide lethality . To quantitatively assess the drug resistance of a mutant , we exposed embryos to 12 mg/ml ethosuximide for five to six days and measured the number of surviving animals that developed past the L1 larval stage . che-3 ( e1124 ) mutants displayed resistance to 12 mg/ml ethosuximide that was 31 percent penetrant ( Table 1 ) . Furthermore , che-3 ( e1124 ) failed to complement am178 for resistance to ethosuximide , indicating that am178 is a mutation in che-3 ( see Materials and Methods ) . To test the prediction that the am178 mutation reduces the activity of the che-3 gene , we generated transgenic am178 mutants containing 32 kB of genomic DNA in a fosmid clone that included the che-3 gene . This wild-type copy of the che-3 gene rescued the am178 mutant phenotype ( see Materials and Methods ) . To identify the molecular lesion , we determined the DNA sequence of the predicted che-3 exons and splice junctions using DNA from am178 mutants . We identified a single base change that affects the splice junction following exon 9 ( the 5′ splice site was changed from GTATG in wild-type to GTATA in am178 mutants ) . Analysis of transcripts demonstrated that am178 mutants produce an aberrantly spliced che-3 mRNA that is predicted to encode a truncated CHE-3 protein , suggesting that am178 is a loss-of-function mutation of the che-3 gene . che-3 encodes a cytoplasmic dynein heavy-chain [42] . Dynein is a component of the intraflagellar transport ( IFT ) machinery , and this machinery builds and maintains the structure of ciliated neurons in C . elegans . Mutations that disrupt IFT components , such as CHE-3 dynein , result in defects in the structure of ciliated neurons [43] . C . elegans contains 60 ciliated neurons , and these neurons function in the perception of chemical and mechanical cues [28]–[31] , [44] . The most extensively characterized ciliated neurons are present in the amphid organ , a bilaterally symmetrical neural structure located in the anterior of the animal . Some amphid neurons have ciliated dendrites that are exposed to the external environment , since they extend through a channel in the cuticle [28] , [45] . Since these neurons are exposed to the environment , treating animals with lipophilic dyes such as DiO specifically stains these neurons ( Figure 3A ) [28] , [46] . Mutants with severe defects in the structure of these neurons display no staining , a phenotype referred to as dye-filling defective ( Dyf ) ( Figure 3D ) [28] , [31] . Many che-3 mutations cause a robust Dyf phenotype [42] . To characterize che-3 ( am178 ) and two candidate che-3 alleles that were positioned near the center of Chromosome I , am162 and am165 , we performed the dye-filling assay . am162 and am165 both caused a Dyf phenotype ( Table 1 ) , and both mutations failed to complement che-3 ( e1124 ) for dye-filling defects ( Materials and Methods ) , indicating that am162 and am165 are alleles of che-3 . che-3 ( am178 ) mutants cultured at 20°C displayed dye-filling that was not significantly different from wild-type ( Table 1 , Table 2 and Figure 3B ) . However , che-3 ( am178 ) mutants cultured at 27°C displayed a highly penetrant dye-filling defective phenotype ( Table 2 ) , indicating that the am178 mutation partially reduces the activity of the che-3 gene at the permissive temperature of 20°C and strongly reduces the activity of the che-3 gene at the restrictive temperature of 27°C . To determine the temporal requirement for che-3 gene activity , we cultured hermaphrodites at 20°C until the L4 stage of development and then shifted the animals to 27°C for 24 hours before scoring DiO staining . Using this regimen che-3 ( am178 ) mutants displayed a significant defect in DiO staining , indicating that che-3 function is necessary after the L4 stage to maintain cilia structure ( Table 2 ) . The findings that che-3 ( am178 ) mutants cultured at 20°C displayed robust resistance to ethosuximide toxicity but did not display dye-filling defects indicate that gross morphological defects in amphid neuron structure are not required for che-3 mutants to be resistant to ethosuximide . Since che-3 mutations cause both ethosuximide resistance and DiO staining defects , we characterized dye-staining of the remaining ethosuximide resistant mutants . Three mutations that exhibited tightest linkage to a marker positioned at −3 . 7 on Chromosome IV , am161 , am172 and am177 , caused defects in DiO staining ( Table 1 ) . We investigated the possibility that these are osm-3 alleles because osm-3 is located at −2 . 2 on Chromosome IV and can be mutated to produce defective DiO staining ( Figure 3D ) . am161 , am172 , and am177 each failed to complement the osm-3 reference allele p802 for DiO staining defects , suggesting that each mutation affects the osm-3 gene ( see Materials and Methods ) . Furthermore , the osm-3 ( p802 ) mutation caused 80 percent resistance to ethosuximide ( Table 1 ) , consistent with the model that am161 , am172 , and am177 are osm-3 alleles . Like the CHE-3 dynein , the OSM-3 kinesin is a component of the IFT machinery that is essential for the formation of ciliated nerve endings . The OSM-3 kinesin has been divided into four domains: motor , neck , rod , and tail ( Figure 4 ) [47] . Previous molecular genetic studies identified two missense mutations that affect the motor region and one missense mutation that affects the rod region . To identify molecular lesions in the new osm-3 alleles , we determined the DNA sequence of osm-3 exons using DNA from am177 mutants . We identified a single base change in the coding region of the am177 allele; this missense mutation changes codon 329 from asparagine to histidine ( Figure 4 ) . The am177 mutation in the osm-3 gene is distinct from previously characterized missense mutations and affects the neck region , thus identifying a residue in the neck region that is critical for protein function ( Figure 4 ) . The demonstration that che-3 and osm-3 are necessary for sensitivity to ethosuximide-induced lethality suggests that the function of ciliated neurons is necessary to mediate the toxic effects of ethosuximide . To investigate this model , we exploited the well-characterized collection of mutations that affect ciliated neurons [28] , [31] . che-13 , daf-10 and osm-5 encode IFT components , and loss-of-function mutations of these genes cause robust defects in DiO staining . Loss-of-function mutations of che-13 , daf-10 and osm-5 caused significant ethosuximide resistance of 62 percent , 23 percent , and 37 percent , respectively ( Table 1 ) . These results support the hypothesis that the function of ciliated neurons is necessary to mediate the effects of ethosuximide . We identified five different genes ( che-3 , osm-3 , daf-10 , osm-5 , and che-13 ) that influence ethosuximide toxicity . Because these genes encode diverse protein products , it is unlikely that these proteins are the direct molecular target of ethosuximide . However , these genes are all required for the structure or function of ciliated neurons , indicating that ciliated neurons may be an important cellular focus of ethosuximide function . The che-3 , che-13 , daf-10 , and osm-5 genes are necessary for the structure or activity of all 60 ciliated neurons [28] . By contrast , the osm-3 gene is necessary for only a subset of ciliated neurons . osm-3 expression is restricted to amphid , phasmid and inter labial neurons that function in chemosensation , and osm-3 loss-of-function mutants are specifically defective in behaviors mediated by these neurons [28] , [48] , [49] . These observations define a subset of chemosensory neurons that are defective in osm-3 mutants as critical for the response to ethosuximide toxicity . In principle , mutations that disrupt cilium structure could cause resistance to ethosuximide toxicity because they affect the molecular target , the cellular target or the metabolism of the drug . Mutations that disrupt cilium structure prevent the lipophilic dye DiO from staining the membranes of ciliated neurons . This raises the possibility that these mutations also disrupt absorption of water soluble drugs , such as ethosuximide [24] . If this were the case then these mutants are predicted to be resistant to a wide variety of drugs . To investigate this possibility , we analyzed resistance to serotonin , a well-characterized compound that stimulates egg-laying [38] . We compared the resistance of wild-type hermaphrodites and osm-3 ( p802 ) mutants . osm-3 ( p802 ) has been extensively characterized using behavioral assays and electron microscopic analyses [28] . Genetic studies indicate that osm-3 ( p802 ) is a strong loss-of-function mutation , and molecular studies demonstrate that the allele contains a nonsense change at codon 346 [47] ( Figure 4 ) . Our analysis also supports the conclusion that osm-3 ( p802 ) is a strong loss-of-function mutation , since the osm-3 ( p802 ) mutation caused a significantly greater resistance to ethosuximide toxicity than the newly isolated osm-3 ( am177 ) and osm-3 ( am161 ) mutations . Wild-type animals and osm-3 ( p802 ) mutants displayed similar sensitivity to the stimulation of egg-laying caused by serotonin ( Figure 5A ) . These results indicate that entry of serotonin into the animal is normal in osm-3 mutants despite the defects in cilium structure . To directly investigate the absorption of ethosuximide by cilium structure mutants , we analyzed a separate phenotype caused by ethosuximide . If drug absorption is mediated by ciliated neurons , then mutants with defective cilium structure are predicted to be resistant to all the effects of ethosuximide . By contrast , if drug absorption is not mediated by ciliated neurons , then mutants with defective cilium structure are predicted to be sensitive to the effects of ethosuximide mediated by non-ciliated cells . The ethosuximide effect we chose to analyze is the stimulation of egg-laying , since we previously showed this effect required the activity of non-ciliated HSN neurons that innervate the egg laying muscles [14] . We chose the osm-3 ( p802 ) mutant for these analyses because it has severe defects in cilium structure and is highly resistant to ethosuximide toxicity ( Table 1 ) . Wild-type and osm-3 hermaphrodites transferred to M9 buffer laid an average of 0 . 17 and 0 . 29 eggs in 90 minutes , respectively ( Figure 5B ) . Wild-type and osm-3 hermaphrodites displayed a similar , dose-dependent increase in the rate of egg-laying in the presence of ethosuximide ( Figure 5B ) , indicating that ethosuximide stimulates egg-laying in these conditions . These results demonstrate that osm-3 mutants have a normal response to ethosuximide-stimulated egg-laying , indicating that the entry of ethosuximide into the animal is normal in these mutants . Our findings suggest that ethosuximide affects the activity of chemosensory neurons . To directly investigate this model , we compared the effects of ethosuximide treatment and mutations that cause defects in chemosensory neurons . Animals with defects in amphid neurons resulting from mutations or physical ablation do not respond properly to chemical cues from the environment and display defective chemotaxis [28]–[31] . To analyze how ethosuximide treatment affects chemotaxis , we measured chemotaxis towards the volatile odorant isoamyl alcohol which is mediated by AWC amphid neurons [29] . Untreated wild-type animals displayed a chemotaxis index towards isoamyl alcohol of about 0 . 8 , and ethosuximide treatment significantly reduced this to about 0 . 3 ( Figure 6A ) . This result indicates that ethosuximide disrupts the activity of AWC amphid neurons , resulting in defective chemotaxis . To determine if ethosuximide treatment affects additional amphid neurons , we analyzed chemotaxis of ethosuximide treated animals to the volatile attractant diacetyl , an odorant detected by AWA amphid neurons [29] . Ethosuximide treatment significantly diminished chemotaxis toward diacetyl ( Figure 6B ) , indicating that ethosuximide disrupts the activity of AWA neurons . These results suggest that treatment with ethosuximide disrupts multiple chemosensory amphid neurons that mediate chemotaxis , including AWC and AWA . Amphid neurons play important roles in the developmental decision between forming an L3 larva that matures to a reproductive adult or a dauer larva that persists until environmental conditions improve . Laser ablation studies indicate that specific chemosensory amphid neurons are necessary to inhibit dauer formation ( ASI , ADF , and ASG ) , and to promote dauer formation ( ASJ ) [50] , [51] . Consistent with these observations , mutations that cause defects in multiple chemosensory neurons have complex effects on dauer formation . These mutations can cause dauer defective ( Daf-d ) phenotypes or dauer constitutive ( Daf-c ) phenotypes depending on the environmental conditions [32] , [52]–[54] . To investigate the effects of ethosuximide on chemosensory neuron activity , we analyzed the effects of ethosuximide treatment on dauer formation in environmental conditions previously described to cause Daf-d and Daf-c phenotypes in animals with chemosensory neuron defects . The combination of high temperature , high population density and low food availability promotes the formation of dauer larvae in wild-type animals [23] . Mutants with defects in chemosensory neurons display a Daf-d phenotype in this combination of culture conditions , indicating that these sensory neurons are necessary to respond to these environmental cues and promote dauer larvae formation [54] . To determine if ethosuximide treatment causes a similar Daf-d phenotype , we analyzed animals cultured at the high temperature of 25°C , at a high population density , and with low food availability . The osm-3 ( p802 ) mutation significantly reduced the average number of dauer larvae by 10-fold , from 250 to 26 ( p<0 . 0001 ) ( Table 3 ) , consistent with previous observations [54] . Ethosuximide treatment of wild-type animals significantly reduced the average number of dauer larvae by 16-fold , from 250 to 16 ( p<0 . 0001 ) ( Table 3 ) . These results indicate that ethosuximide inhibits chemosensory neurons that are necessary to promote dauer larvae formation . By contrast to the culture conditions described above , culturing animals in low density populations with abundant food at the extremely high temperature of 27°C reveals that chemosensory neuron function is necessary to inhibit dauer formation [32] , [52] . To determine if ethosuximide treatment and mutations that disrupt chemosensory neurons cause similar defects , we analyzed dauer larvae formation at 27°C . The osm-3 ( p802 ) mutation significantly increased the fraction of dauer larvae from 0 to 7 percent ( Table 4 ) . Ethosuximide treatment of wild-type animals significantly increased the fraction of dauer larvae from 0 to 10 percent ( p<0 . 0001 ) ( Table 4 ) , indicating that ethosuximide treatment and an osm-3 mutation caused similar defects in the ability to inhibit dauer formation . To investigate the relationship between the dauer promoting effects of ethosuximide and the osm-3 ( p802 ) mutation , we treated osm-3 ( p802 ) mutants with ethosuximide . Ethosuximide treatment did not enhance the dauer arrest caused by osm-3 ( p802 ) ( Table 4 ) , suggesting that ethosuximide and osm-3 mutations promote dauer arrest by a similar mechanism . Collectively , these results indicate that ethosuximide treatment can cause Daf-c and Daf-d phenotypes in specific environmental conditions , and this result supports the model that ethosuximide disrupts the activity of multiple chemosensory neurons . Insulin/IGF-1 signaling is a critical regulator of dauer formation and an important modulator of adult lifespan [55]–[57] . daf-2 encodes a receptor tyrosine kinase similar to the insulin receptor [56] , and age-1 encodes a phosphatidylinositol-3-kinase that functions downstream in the signaling pathway [58] . Loss-of-function mutations in daf-2 and age-1 promote dauer formation . In standard culture conditions that do not promote dauer formation , 2 mg/ml ethosuximide treatment enhanced the dauer arrest phenotype of daf-2 ( m41 ) mutants from 3 . 4 percent ( N = 263 ) to 91 . 3 percent ( N = 304 ) , suggesting an interaction between ethosuximide and insulin/IGF-1 signaling . In addition , the insulin/IGF-1 signaling pathway regulates the ability of animals that hatch in the absence of food to arrest at the L1 stage of development [59] , [60] . Loss-of-function mutations in daf-2 and age-1 cause animals to arrest at the L1 stage inappropriately when food is present [55] , [61] . The activity of chemosensory neurons modulates this L1 arrest , since mutations in genes such as osm-3 and che-3 enhance the L1 arrest phenotype of daf-2 and age-1 mutants [54] , [61] . To test the effects of ethosuximide treatment on L1 arrest , we treated daf-2 mutants with the lifespan-extending dose of 4 mg/ml ethosuximide . Treatment of daf-2 ( e1370 ) animals with ethosuximide significantly increased L1 arrest from 19 . 2 percent in untreated animals to 47 . 5 percent in drug treated animals ( Table 5 ) . The ability of mutations that affect chemosensory neurons to enhance the daf-2 L1 arrest phenotype requires the activity of daf-16 , since the addition of a daf-16 mutation abrogates this effect [62] . Similarly , the daf-16 ( mu86 ) mutation significantly reduced the effect of ethosuximide treatment from 47 . 5 percent in daf-2 ( e1370 ) mutants to 2 . 3 percent in daf-2 ( e1370 ) ; daf-16 ( mu86 ) mutants ( Table 5 ) . These results indicate that ethosuximide treatment and mutations that disrupt chemosensory neurons have a similar effect on the L1 arrest phenotype of an insulin-signaling mutant , consistent with the model that ethosuximide inhibits the activity of chemosensory neurons . One interpretation of these findings is that daf-2 ( e1370 ) mutants are hypersensitive to ethosuximide toxicity , since wild-type animals treated with 12 mg/ml ethosuximide also arrest development at the L1 stage . In addition to functioning in chemotaxis , dauer formation , and L1 larval arrest , chemosensory neurons play a role in adult lifespan determination [32]–[34] . The finding that ethosuximide treatment extends the adult lifespan and affects the activity of chemosensory neurons , suggests that ethosuximide extends adult lifespan by affecting the activity of chemosensory neurons . To investigate this model , we monitored the lifespan of osm-3 ( p802 ) hermaphrodites . osm-3 mutants display a significantly extended lifespan [32] ( Figure 7 and Table 6 ) . If ethosuximide extends the adult lifespan by modulating the activity of chemosensory neurons , then osm-3 ( lf ) animals are predicted to be resistant to the lifespan extension caused by ethosuximide . Treatment of wild-type hermaphrodites with 2 mg/ml or 4 mg/ml ethosuximide significantly extended the mean lifespan by 16 percent or 13 percent , respectively , and the maximum lifespan by 29 percent or 21 percent , respectively ( Figure 7 and Table 6 ) . By contrast , treatment of osm-3 ( p802 ) hermaphrodites with 2 mg/ml or 4 mg/ml ethosuximide did not cause a statistically significant extension of mean or maximum lifespan ( Figure 7 and Table 6 ) . These results support the model that ethosuximide extends lifespan by inhibiting chemosensory neurons .
A variety of drugs have been demonstrated to extend the lifespan of invertebrates , including compounds that are proposed to act as antioxidants [4]–[9] , complex chemical mixtures derived from plants [10] , [11] , resveratrol [12] , [13] , histone deacetylase inhibitors [63] , [64] , and compounds that influence the vertebrate nervous system [14]–[17] . In each case , the understanding of how these drugs act to extend lifespan is limited . In particular , the cellular target has not been established for any of these compounds . Advancing this understanding is challenging because of the complexity of the aging phenotype and the fact that many determinants influence lifespan . Characterizing the mechanism of these drugs and defining the cellular target of action is critical for elucidating the endogenous pathways that are affected by the drugs , and ultimately using these drugs in therapeutic applications . Here we demonstrate that chemosensory neurons are the cellular target of ethosuximide in lifespan extension . Two lines of evidence support this conclusion . First , ethosuximide treatment caused the same defects as mutations that result in structural defects in chemosensory neurons , including defective chemotaxis , abnormal dauer arrest , abnormal L1 arrest and extended lifespan . Second , osm-3 mutants that are defective in a subset of chemosensory neurons were resistant to the lifespan extension caused by ethosuximide treatment and the toxic effects of ethosuximide . Together these results suggest that ethosuximide inhibits chemosensory function to extend C . elegans lifespan . This conclusion is important because it defines a specific cellular target for a lifespan extending drug and it demonstrates for the first time that a drug can act on specific cells in the nervous system to extend lifespan . Our previous studies of the related compound trimethadione demonstrated that osm-3 mutants are partially resistant to the mean lifespan extension caused by trimethadione and are fully resistant to the maximum lifespan extension cause by trimethadione [14] . These results are consistent with the conclusion that the lifespan extension caused by trimethadione is partly caused by inhibiting the function of chemosensory neurons . The finding that part of the lifespan extending activity of trimethadione was not suppressed by an osm-3 mutation raises the possibility that trimethadione functions by additional mechanisms to extend lifespan . We demonstrated that resistance to ethosuximide lethality can be caused by mutations in several different genes that disrupt ciliated neuron structure and function , including che-3 , osm-3 , che-13 , osm-5 and daf-10 . Ciliated neurons in the amphid sheath have access to the environment so that they can detect chemicals and function in sensory perception . Because these neurons are exposed to the environment the cell membranes can be stained with lipophilic dyes . These observations raise the possibility that ciliated neurons absorb water-soluble chemicals , like ethosuximide , from the environment and mutations that disrupt cilium structure reduce drug absorption [24] . However , several lines of evidence suggest that mutants with defects in ciliated neurons are normal for drug absorption . First , we demonstrated that che-3 ( am162 ) mutants were weakly resistant to ethosuximide lethality and strongly defective in DiO staining . By contrast , che-3 ( am178 ) mutants raised at 20°C were strongly resistant to ethosuximide lethality but displayed relatively normal DiO staining , indicating that the ciliated neurons in these mutants have access to the environment . The lack of a correlation between dye-filling defects and ethosuximide resistance suggests that the ethosuximide resistance of che-3 mutants is not caused by a defect in the ability of the drug to enter the animal . Second , we demonstrated that osm-3 mutants are not resistant to drugs in general . Cilium structure mutants are resistant to the effects of the nematicidal drug ivermectin [24] . Therefore , we directly tested the possibility that cilium structure mutants are resistant to drugs in general by analyzing sensitivity to a third compound , serotonin . osm-3 mutants and wild-type animals displayed similar sensitivity to egg-laying stimulation caused by serotonin , indicating that the mutation does not impair absorption of serotonin . In addition , screens for mutations that cause resistance to fluoxetine [27] , [39] , levamisole [65] , benzimidazole [66] , aldicarb [67] , nemadipine-A [41] , BMS-192364 [40] , and α-amanatin [68] have been described . Despite extensive screening and the successful identification of mutants resistant to these compounds , none of these reports describe the isolation of mutations that disrupt the structure of ciliated neurons . While the failure to identify a class of mutations in a genetic screen is not a definitive finding , these results suggest that mutations that disrupt cilium structure do not cause significant resistance to these compounds . Third , we took advantage of our observation that ethosuximide treatment stimulates egg-laying and this effect requires the non-ciliated HSN neurons [14] . osm-3 ( p802 ) mutants that are highly resistant to the lethality and lifespan extension caused by ethosuximde were nonetheless sensitive to the egg-laying effects of the drug , indicating that these mutants are not defective in ethosuximide absorption . A simple model to explain our findings is that ethosuximide is absorbed by animals independent of the structure of ciliated neurons and the drug influences the activity of multiple neurons throughout the body . The activity of ethosuximide on HSN neurons that mediate egg-laying is intact in osm-3 mutants whereas the activity on ciliated neurons that mediate toxicity and lifespan extension is altered . Chemosensory neurons have been extensively characterized in C . elegans to understand behavior , and Kenyon and colleagues demonstrated that chemosensory neurons play an important role in controlling adult lifespan [32] , [34] . Mutants that have defects in chemosensory neuron function display an extended lifespan [32] , and ablation of specific neurons such as ASI , AWA , and AWC extends the adult lifespan [34] . Here we provide an independent line of evidence that these neurons control lifespan . Our results indicate that ethosuximide inhibits the activity of chemosensory neurons and thereby causes a lifespan extension . These results provide independent support the model that high levels of chemosensory neuronal activity promote a reduced lifespan and inhibiting these neurons causes a lifespan extension . An important issue that is raised by these observations is how does the activity of chemosensory neurons influence lifespan ? One possibility is that the chemosensory neurons regulate the activity of the insulin/IGF-1 signaling pathway . Amphid neurons are the site of expression of many insulin/IGF-1 ligands [69] raising the possibility that these neurons release ligand in response to environmental cues and thereby influence lifespan by an endocrine mechanism . Consistent with this model , ethosuximide enhanced L1 arrest in daf-2 mutants , similar to osm-3 mutations [54] , [61] . An important test of this model is the dependence of these effects on DAF-16 , a transcription factor that mediates effects of insulin/IGF-1 signaling . Ethosuximide enhancement of L1 arrest was dramatically reduced by a daf-16 ( lf ) mutation . Furthermore , ablation of ASI amphid neurons extends lifespan in a manner the requires the DAF-16 FOXO transcription factor [34] . These results suggest that amphid neurons are important for insulin/IGF-1 signaling and that ethosuximide may extend lifespan by disrupting this pathway . However , ethosuximide and osm-3 mutations extend the lifespan of daf-16 mutants , indicating loss of insulin/IGF-1 signaling cannot fully explain these lifespan extensions [14] , [32] . Furthermore , ablation of AWA and AWC amphid neurons extends the lifespan of daf-16 mutants [34] . Therefore , our data support the model that ethosuximide acts through insulin-dependent and insulin-independent pathways to extend C . elegans lifespan . Restricting dietary intake can extend lifespan in many animal models [70]–[73] , and these observations document the critical role of food availability in controlling lifespan . An important question is how do animals assess food availability to control lifespan ? In particular , is food availability determined metabolically , by monitoring nutrients that are ingested , or is food availability determined by neural sensation , by monitoring food derived cues in the environment ? Recent studies in Drosophila indicate that the effects of dietary restriction are mediated , in part , by the perception of food-derived cues [74] . Exposure of flies to food-derived odors can partially suppress the lifespan extension caused by dietary restriction . Therefore , mutations that extend lifespan by disrupting chemosensation may block perception of food-related chemical cues and thereby activate pathways that respond to dietary restriction . Here we demonstrate that the anticonvulsant ethosuximide extends C . elegans lifespan by inhibiting sensory neurons that are hypothesized to mediate attraction to environmental food sources . Hermaphrodites treated with ethosuximide appear to ingest a normal amount of food since they do not display a diminished body size or progeny production , the characteristics of dietary restricted animals [72] . These results suggest that inhibiting the sensation of food is sufficient to extend lifespan even in the presence of normal food ingestion . These observations raise the exciting possibility that inhibiting the sensation of food may be a conserved mechanism of lifespan extension . In addition to controlling seizures , a common side effect of ethosuximide treatment in humans is loss of taste sensation [75] . Therefore , targeting sensory mediated processes with drugs like ethosuximide may present a means to extend mammalian lifespan .
C . elegans strains were cultured at 20°C on 6 cm Petri dishes containing nematode growth media ( NGM ) agar and a lawn of E . coli strain OP50 unless otherwise noted [76] . Unless otherwise noted , preparation and storage of dishes containing pharmacological compounds was performed as previously described [14] . We used the following C . elegans mutations that are described in Riddle et al . [23] or in this study: che-3 ( e1124 ) I , che-3 ( am165 ) I , che-3 ( am162 ) I , che-3 ( am178 ) I , che-13 ( e1805 ) I , daf-16 ( mu86 ) I , unc-11 ( e47 ) I , dpy-5 ( e61 ) I , unc-29 ( e1072 ) I , unc-75 ( e950 ) I , daf-2 ( e1370 ) III , daf-2 ( m41 ) III , osm-3 ( p802 ) IV , osm-3 ( am161 ) IV , osm-3 ( am177 ) IV , osm-3 ( am172 ) IV , daf-10 ( e1387 ) IV , osm-5 ( p813 ) X , lin-15 ( n765 ) X , cca-1 ( ad1650 ) X , and cca-1 ( gk30 ) X . The following well-characterized mutations were used in this study: che-3 ( e1124 Q2233stop ) is a probable null mutation in the CHE-3 dynein [42]; che-13 ( e1805 Q219stop ) is a loss-of-function mutation that affects the IFT57 protein [77]; osm-3 ( p802 Q346stop ) is a strong loss-of-function mutation that disrupts the OSM-3 kinesin [47]; daf-10 ( e1387 Q892stop ) is a loss-of-function mutation that disrupts the IFT122 protein [78]; osm-5 ( p813 Q473stop ) is a loss-of-function mutation that disrupts the Tg737/Polaris protein [79]; daf-2 ( e1370 P1465S ) is a partial loss-of-function mutation that affects the kinase domain of the DAF-2 receptor tyrosine kinase [56]; daf-2 ( m41 G383E ) is a weak loss-of-function mutation that affects the ligand-binding domain of the DAF-2 receptor tyrosine kinase [80]; daf-16 ( mu86 ) is a probable null mutation that results in deletion of most of the DAF-16 coding sequence including all of the forkhead domain [57]; cca-1 ( ad1650 ) and cca-1 ( gk30 ) are strong loss-of-function mutations that result in deletions of different portions of the CCA-1 coding sequence [37] . N2 hermaphrodites ( P0 ) were mutagenized as described by Brenner [76] with either 50 mM EMS or 0 . 5 mM ENU . 100 , 000 F1 hermaphrodites were treated with hypochlorite , and F2 eggs were plated on NGM containing 10–12 mg/ml ethosuximide ( Sigma , St . Louis , MO ) . Since E . coli failed to form a thick lawn on dishes containing 12 mg/ml ethosuximide , these dishes were seeded with E . coli OP50 that had been concentrated 10-fold . F2 progeny that matured to the L4/adult stage were picked as ethosuximide resistant mutants , and populations derived from these individuals were retested for ethosuximide resistance . 48 independently derived ethosuximide resistant mutants were isolated . These mutant strains were backcrossed at least twice to wild-type ( N2 ) to remove extraneous mutations . Forty-two mutant strains were backcrossed successfully , indicating that ethosuximide resistance was caused by a single mutation . Single nucleotide polymorphism ( SNP ) mapping analysis was performed on a subset of resistant mutants by mating ethosuximide resistant mutants ( P0 ) to the divergent CB4856 strain ( P0 ) , selecting F1 outcross progeny , and scoring F2 self-progeny for ethosuximide resistance . F2 animals that displayed resistance were judged to be homozygous for the mutation , and F3 progeny were harvested for DNA . SNPs distributed throughout the C . elegans genome [81] were scored using Pyrosequencing ( Biotage Foxboro , MA ) or direct DNA sequencing . Linkage values were calculated by determining the ratio of N2 DNA to CB4856 DNA at each polymorphism . The am178 mutation was tightly linked to a SNP marker at the center of Chromosome I . Three factor mapping experiments with visible markers yielded the following results . From am178/unc-11 dpy-5 hermaphrodites , 5/6 Unc non Dpy self progeny segregated am178 and 0/9 Dpy non Unc self progeny segregated am178; From am178/dpy-5 unc-75 hermaphrodites , 3/8 Unc non Dpy self progeny segregated am178; From am178/dpy-5 unc-29 hermaphrodites , 7/8 Dpy non Unc self progeny segregated am178 . These results position am178 right of dpy-5 , left of unc-75 , and probably left of unc-29 . For high resolution SNP mapping of am178 , we mated dpy-5 ( e61 ) am178 and am178 unc-75 ( e950 ) homozygotes to CB4856 males , picked F1 outcross progeny , and selected non-Dpy and non-Unc F2 self-progeny resistant to 12 mg/ml ethosuximide . We prepared DNA from strains homozygous for the recombinant chromosome and scored SNP markers . Because che-3 and osm-3 mutations interfere with male mating , che-3 ( e1124 ) and osm-3 ( p802 ) were maintained over hT2 and nT1 myo-2::GFP balancer chromosomes , respectively . For complementation analysis with am178 and che-3 ( e1124 ) , one am178 hermaphrodite was mated to five che-3 ( e1124 ) /hT2 males and outcross progeny were scored for ethosuximide resistance . Non-GFP , outcrossed am178/che-3 ( e1124 ) animals survived 12 mg/ml ethosuximide treatment , indicating that che-3 ( e1124 ) and am178 fail to complement for ethosuximide resistance . Neither che-3 ( e1124 ) /+ nor am178/+ animals survived 12 mg/ml ethosuximide treatment . For complementation analysis with dye-filling defective alleles , ethosuximide resistant mutants that displayed a Dyf phenotype were mated to either che-3 ( e1124 ) /hT2 or osm-3 ( p802 ) /nT1 males , and non-GFP outcross progeny were subjected to dye-filling analysis . 9/9 am172/osm-3 ( p802 ) animals , 15/15 am177/osm-3 ( p802 ) animals and 20/25 am161/osm-3 ( p802 ) animals displayed a Dyf phenotype , indicating that these three alleles fail to complement osm-3 ( p802 ) for dye-filling . 14/15 am165/ che-3 ( e1124 ) animals and 9/9 am162/ che-3 ( e1124 ) animals displayed a Dyf phenotype , indicating that these two alleles fail to complement che-3 ( e1124 ) for dye-filling . To analyze the ability of genomic DNA to rescue the mutant phenotype of am178 , we generated transgenic animals containing extrachromosomal arrays using standard procedures [82] . Fosmid clone WRM0637cB0 contains the entire che-3 coding region and only one other entire gene , F18C12 . 4 . To identify transgenic animals , we co-injected the fosmid clone WRM062bF09 that rescues the lin-15 ( n765 ) multi-vulval ( Muv ) phenotype ( M . Nonet , personal communication ) . These two fosmids were co-injected at a concentration of 20 ng/µl each into am178; lin-15 ( n765ts ) animals to generate the transgenic array amEx100 . Hermaphrodites containing the amEx100 array transmitted it to about 73% of self-progeny , since 27% of self-progeny were Muv ( n = 100 ) . When self-progeny of am178; lin-15 ( n765 ) ;amEx100 hermaphrodites were plated on media containing 12 mg/ml ethosuximide , the only animals that survived to adulthood after 5–6 days were Muv ( N = 59 ) , indicating that they did not contain amEx100 . These results indicate that the amEx100 array rescues the am178 resistance to ethosuximide phenotype , suggesting that the gene affected by am178 is contained in fosmid WRM0637cB0 . DNA sequencing was performed using standard procedures . We PCR amplified predicted exons and splice junctions ( www . wormbase . org ) from the che-3 and osm-3 genes using DNA from am178 and am177 mutant animals , respectively . The am178 strain contained a G to A change at the fifth base of intron 9 . To analyze the che-3 messenger RNA , we prepared RNA from a population of am178 mutants ( TRizol , Invitrogen ) , DNase treated the RNA ( DNA-free , Ambion ) , and generated cDNA by performing reverse transcription ( Retroscript , Ambion ) . We PCR amplified the junctions between exons 9 and 10 and determined the DNA sequence of the PCR product to infer the splicing pattern of the mRNA . The che-3 mRNA derived from am178 mutants displayed a deletion of 17 nucleotides ( GTCAGCTTGGTTTTTGC ) in exon 9 . Staining with DiO was performed using previously described methods [31] , [83] . L4 hermaphrodites were placed in 100 µL of M9 buffer containing 20 µg/ml DiO ( Molecular Probes ) in a microtiter dish and incubated for ∼2 hrs at room temperature . To remove non-specifically bound DiO , we transferred the animals to a NGM dish seeded with E . coli OP50 and cultured for 20–30 minutes at room temperature . For analysis of strains in Table 1 , five to fifteen animals were observed using a Zeiss Axioplan 2 microscope equipped for fluorescence microscopy at 400× magnification . Animals were categorized in the following classes: ( I ) No amphid neurons stained; ( II ) 1–2 amphid neurons stained; or ( III ) more than 2 amphid neurons stained . All the strains described in Table 1 displayed only the class I or only the class III pattern of staining . For analysis of che-3 ( am178 ) mutants in Table 2 , DiO staining was observed using an Olympus SZX12 dissecting microscope equipped for fluorescence microscopy at 144× magnification . Animals were scored as Dyf if no amphid neurons stained robustly with DiO . To quantify the penetrance of the ethosuximide resistance phenotype , we adapted an assay method described by Rand and Johnson [84] . Eggs were picked to a Petri dish with agar containing 12 mg/ml ethosuximide and counted . Five to six days later we counted the number of animals that matured past the L1 stage . For resistant strains most animals were L4 larvae or adults . The remaining eggs produced animals that did not mature past the L1 stage or left the agar surface . The number of animals that matured past the L1 stage was divided by the total number of eggs to determine the percent resistant to ethosuximide . For pharmacological analysis of chemotaxis , dishes were prepared by adding powdered ethosuximide directly to 60°C chemotaxis agar , agitating to dissolve the compound and dispensing to 10 cM Petri dishes . Chemotaxis assays were performed as described previously [85] with two minor modifications . First , to obtain a large number of well-fed animals for the assay , we cultured animals from conception to adulthood in the presence of ethosuximide or no drug on 6 cM Petri dishes containing OP50 that was concentrated 10-fold . Second , we added ethosuximide to the molten chemotaxis agar before dispensing into 10 cM Petri dishes and used these dishes after 3–8 hours . To conduct chemotaxis assays , we pipeted animals in a small volume of liquid onto the center of a 10 cM Petri dish that had been pre-treated with 1 µL of 1M Na azide , a paralyzing agent , at two diametrically opposed spots on the plate . At this time , we added 1 µL of volatile odorant to one azide treated spot and 1 µL of ethanol as a control to the other azide treated spot . Isoamyl alcohol and diacetyl were diluted in ethanol at concentrations of 1∶10 and 1∶1000 , respectively . After 60 minutes we scored the number of paralyzed animals at each azide spot and the number of moving animals on the plate . The sodium azide spots were positioned at the edge of the dish , and animals that desiccated on the side of the dish at the positions of the sodium azide spots were attributed to those categories; otherwise such animals were attributed to the moving animal category . Chemotaxis index was determined using the following formula: The analysis of egg-laying was performed as described previously [85] . To determine the effect of ethosuximide on egg-laying , we picked L4 hermaphrodites , incubated them at 20°C for approximately 20 hours on NGM plates with abundant food , individually placed them in 100 µL of M9 buffer in a microtiter dish plus or minus ethosuximide , and counted the number of eggs laid in 90 minutes . We used a similar method to analyze serotonin , except animals were placed in 50 µL buffer . To analyze dauer formation at 27°C , we allowed 8–10 gravid hermaphrodites to lay eggs on one Petri dish for 4–8 hours at room temperature and then incubated the dish at 27°C for precisely 44 hours [52] . We counted the total number of animals on the dish , flooded the dish with 1% SDS to kill all animals except for dauer larvae , and counted the number of live dauer larvae after 15 minutes . To analyze dauer formation at 25°C , we prepared NGM media without peptone , the main carbon source for the E . coli OP50 . The omission of peptone from the media allowed us to provide a consistent amount of E . coli OP50 food , and made the experiment independent of the effects of ethosuximide on bacterial proliferation . We aliquoted 200 µl of an overnight E . coli OP50 culture that had been concentrated 10-fold on each dish . We placed two L4 hermaphrodites on each dish , incubated the animals at 25°C , and monitored the dishes daily for starvation [86] . Five days post-starvation we scored the number of dauer larvae by flooding the dishes with 1% SDS and counting the number of live animals after 15 minutes . Dauer formation of daf-2 ( m41 ) mutants was analyzed as described previously [16] . To analyze the L1 arrest phenotype , we cultured L4 hermaphrodites 1–2 days on NGM dishes with 4 mg/ml ethosuximide or no drug , transferred these adults to fresh dishes with 4 mg/ml ethosuximide or no drug for 4–8 hours at room temperature , and removed the adults . These dishes containing freshly deposited eggs were cultured at 25 . 5°C for 48 hours . We counted the number of eggs , L1 larvae , and older larval stages . For a typical lifespan experiment , parental worms were cultured in the presence of the drug , and progeny were selected at the L4 stage for lifespan analysis . Thus , these progeny were exposed to drug from the time of conception until death . For measurements of lifespan , hermaphrodites were chosen for analysis at the L4 stage ( defined as day 0 ) and analyzed every 1–2 days from day 3 until death . Approximately 15 hermaphrodites were cultured on each Petri dish . Hermaphrodites were transferred to fresh Petri dishes about every two days until the cessation of progeny production and about every week thereafter . Animals were scored as dead if they displayed no spontaneous movement or response when prodded . Dead worms that displayed internally hatched progeny , an extruded gonad or desiccation due to crawling off the agar were excluded from the data . Lifespan is the number of days from the L4 stage to the average of the last day a worm was observed to be alive and the first day a worm was observed to be dead . Lifespan experiments involving pharmacological compounds were always done in parallel with a control group . For each experimental group , comparisons were made to a control group maintained in the same incubator and analyzed at the same time points . Mean , standard error , P values , and other statistical parameters were calculated using InStat 2 . 03 software ( Graphpad Software ) or Microsoft Excel . | Aging is a major factor that contributes to disease and disability in humans , but no medicines have been demonstrated to delay human aging . We previously conducted a screen for FDA-approved drugs that can extend the lifespan of the nematode worm C . elegans , resulting in the identification of ethosuximide , a medicine used to treat epilepsy . To elucidate the mechanism of action of ethosuximide in lifespan extension , we conducted a genetic screen for C . elegans mutations that cause resistance to ethosuximide . Here , we describe the identification of genes that are critical for ethosuximide sensitivity . These genes are necessary for the function of neurons that mediate sensory perception . Furthermore , ethosuximide treatment caused defects in sensory perception . These results indicate that ethosuximide affects lifespan by inhibiting neurons that function in the perception of sensory cues . These studies highlight the importance of sensory neurons in lifespan determination and demonstrate that a drug can act on specific cells within the nervous system to extend lifespan . Sensory perception also modulates Drosophila lifespan , suggesting this is an evolutionarily conserved relationship . Our results indicate that sensory perception may be a promising target for pharmacological extension of lifespan in a variety of animals . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genetics",
"and",
"genomics",
"developmental",
"biology/aging"
] | 2008 | The Anticonvulsant Ethosuximide Disrupts Sensory Function to Extend C. elegans Lifespan |
Infections of Echinococcus granulosus sensu stricto ( s . s ) , E . multilocularis and E . shiquicus are commonly found co-endemic on the Qinghai-Tibet plateau , China , and an efficient tool is needed to facilitate the detection of infected hosts and for species identification . A single-tube multiplex PCR assay was established to differentiate the Echinococcus species responsible for infections in intermediate and definitive hosts . Primers specific for E . granulosus , E . multilocularis and E . shiquicus were designed based on sequences of the mitochondrial NADH dehydrogenase subunit 1 ( nad1 ) , NADH dehydrogenase subunit 5 ( nad5 ) and cytochrome c oxidase subunit 1 ( cox1 ) genes , respectively . This multiplex PCR accurately detected Echinococcus DNA without generating nonspecific reaction products . PCR products were of the expected sizes of 219 ( nad1 ) , 584 ( nad5 ) and 471 ( cox1 ) bp . Furthermore , the multiplex PCR enabled diagnosis of multiple infections using DNA of protoscoleces and copro-DNA extracted from fecal samples of canine hosts . Specificity of the multiplex PCR was 100% when evaluated using DNA isolated from other cestodes . Sensitivity thresholds were determined for DNA from protoscoleces and from worm eggs , and were calculated as 20 pg of DNA for E . granulosus and E . shiquicus , 10 pg of DNA for E . multilocularis , 2 eggs for E . granulosus , and 1 egg for E . multilocularis . Positive results with copro-DNA could be obtained at day 17 and day 26 after experimental infection of dogs with larval E . multilocularis and E . granulosus , respectively . The multiplex PCR developed in this study is an efficient tool for discriminating E . granulosus , E . multilocularis and E . shiquicus from each other and from other taeniid cestodes . It can be used for the detection of canids infected with E . granulosus s . s . and E . multilocularis using feces collected from these definitive hosts . It can also be used for the identification of the Echinococcus metacestode larva in intermediate hosts , a stage that often cannot be identified to species on visual inspection .
In the most recent taxonomic revision , nine species were recognized in the genus Echinococcus [1] . Of these , the most important and widespread are E . granulosus sensu stricto ( genotypes G1-G3 ) and E . multilocularis , which cause cystic echinococcosis ( CE ) and alveolar echinococcosis ( AE ) , respectively . The former is commonly associated with livestock and human infections worldwide whereas the latter is primarily found in voles and humans and is geographically limited to the northern hemisphere [2] . To date , E . granulosus s . s . , E . canadensis ( G6 ) , E . multilocularis and E . shiquicus have been identified in China [3–5] . Both E . multilocularis and E . granulosus s . s . are particularly widespread in western China , including Qinghai , Ningxia , Gansu , Xinjiang and Sichuan provinces/autonomous regions , and are well known as major public health and medical threats . Unlike the other species , E . shiquicus has a very restricted distribution , being reported only from Qinghai Province , China . This species is not known to cause human echinococcosis . The intermediate hosts are plateau pikas ( Ochotona curzoniae ) , in which unilocular cysts occur . For Echinococcus species in general , dogs , wolves , other canids and cats are definitive hosts in which adult worms cause sub-clinical infections [6–9] . However , larval Echinococcus spp . can cause morbidity and mortality in their intermediate hosts which include cattle , sheep , small mammals ( including rodents , plateau pikas , etc . ) and humans [10 , 11] . It can be difficult to discriminate morphologically adults of some Echinococcus species , such as E . multilocularis and E . shiquicus [12] . To replace traditional morphological methods , a number of molecular approaches targeting parasite DNA have been developed for identification/discrimination of different life stages of Echinococcus species in definitive and intermediate hosts [13–15] . Multiplex PCR approaches , simultaneously using multiple specific primers in a single tube and detecting more than one target species , are material- and time-saving , precise , efficient and cost-effective when DNA from a mixture of pathogens may be present in a sample . This approach is also suitable for mass-screening of samples that may be generated from epidemiological investigations in endemic areas . Several multiplex PCR methods have been developed for identifying certain Echinococcus species , but none for the identification of E . shiquicus [16–17] . Based on interspecific variation in mitochondrial genes of the genus Echinococcus , we designed a multiplex PCR assay with three pairs of specific primers in a single reaction tube for rapid identification of E . granulosus s . s . , E . multilocularis and E . shiquicus originating from either intermediate or definitive hosts . Further assessment of the sensitivity and specificity of the multiplex PCR assay was performed using metacestode DNA and copro-DNA to determine the reliability and accuracy of the new diagnostic tool developed in this study .
Dogs and mice used in this study were handled in strict accordance with good animal practice according to the Animal Ethics Procedures and Guidelines of the People's Republic of China ( Regulations for Administration of Affairs Concerning Experimental Animals , China , 1988 ) . No endangered/protected species were involved in this study . The dogs and mice used were also treated in accordance with the institutional procedures and guidelines for animal husbandry issued by the Ethics Committee of Lanzhou Veterinary Research Institute , Chinese Academy of Agricultural Sciences ( Approval No . LVRIEC2010-005 ) . Adult worms were collected from stray dogs during routine work of the endemic echinococcosis prevention and control program in Dari County , Qinghai Province , P . R . China . A total of 86 Echinococcus spp . metacestode samples from yaks , sheep , Qinghai voles ( Microtus/Neodon fuscus ) and plateau pikas were collected on the Qinghai-Tibet plateau , P . R . China . Ten yak lungs and 16 sheep livers harboring hydatid cysts were collected from abattoirs in Maqu County , Gansu Province and Xining City , Qinghai Province , respectively . Thirty Qinghai vole livers and 30 plateau pika lungs harboring hydatid cysts were provided by the epidemic prevention station of Dari County , Qinghai Province . Parasite materials were dissected from the host tissue and stored either in 70% ethanol before molecular analyses , or temporarily stored at 4°C prior to experimental infections of dogs . Fifteen dogs ( mixed breeds ) aged 6–8 months were purchased in Lanzhou City , Gansu Province , China . These were de-wormed using praziquantel and confirmed to be free of intestinal parasites by examination of their feces two weeks later . Samples of these feces were retained as negative controls for the multiplex PCR assay . Live protoscoleces ( 100 , 000 ) of each Echinococcus spp . were fed independently to five dogs after their viability for dog challenge was confirmed by microscopy . Dogs were euthanized three months after challenge with protoscoleces . Fecal samples were collected from the dogs each day prior to sacrifice . After removal of the coarse gut contents , the small intestine was cut into 15–20 cm lengths and opened to expose the mucosa . Samples , taken by scraping the mucosa with glass strips , were placed in petri dishes in bio-safety containers [18] . After addition of a small volume of sterile phosphate-buffered saline ( PBS , pH 7 . 2 ) , the contents were checked for the presence of worms ( intact or fragmented ) and/or eggs . Adult worms were removed using a glass needle and washed in PBS three times . All procedures were performed following appropriate bio-safety conditions [19] . Ten stray dogs , provided by the epidemic prevention station in Dari County , Qinghai Province , were processed as above to obtain mucosal samples , worms and eggs . Additionally , five fecal samples from captive foxes were collected from a fur farm in Lanzhou City , Gansu Province . All the collected fecal samples were frozen at -80°C for at least seven days for bio-safety reasons . Worm samples were preserved either in 70% ethanol or frozen at ( -80°C ) in PBS for further analyses . DNA samples , extracted from a variety of cestodes ( identities confirmed by sequencing and morphology ) , were used to determine the specificity of the newly developed multiplex PCR assay ( Table 1 ) . They were kindly provided by the Key Laboratory of Veterinary Parasitology of Gansu Province , Lanzhou Veterinary Research Institute , CAAS . DNA extracted from host tissues was used to check for nonspecific reactions or assay interference that might be caused by contamination of parasite samples with host DNA . Host tissues included dog intestines , and liver and lung samples from cattle , sheep , Qinghai voles and plateau pikas . Two hundred mg of each metacestode sample was frozen in liquid nitrogen and ground to powder after removal of ethanol or PBS by rinsing with ddH2O . Total genomic DNA was extracted using a QIAGEN DNeasy Blood & Tissue Kit ( QIAGEN , Hilden , Germany ) according to the manufacturer’s instructions and stored at -20°C until use . To minimize the impact of inhibitors on PCR using copro-samples as template , an additional step of stool flotation in saturated zinc chloride solution was used before copro-DNA extraction [20] . Briefly , about 20 g ( 20 ml ) fecal material was placed in a 50 ml centrifuge tube , which was then filled with zinc chloride solution . The tube was vortexed until the fecal material was completely broken up . The tube was then centrifuged at 1000 ×g for 5 min . Five hundred μl of the supernatant ( usually containing helminth eggs , proglottids or cells of parasites ) was transferred to a 2 ml centrifuge tube , 1 . 5 ml ddH2O was added to dilute the solution , and the tube was centrifuged at 12 , 000 ×g for 10 min . The supernatant was carefully discarded and 200 μl ddH2O added to suspend the sediment for DNA extraction . Total genomic DNA was extracted using a QIAGEN QIAamp DNA Stool Mini Kit ( QIAGEN , Hilden , Germany ) following the manufacturer’s instructions , and the DNA concentration was determined using a spectrophotometer ( Thermo , NanoDrop 2000 , USA ) after elution in 50 μl ddH2O for use in the PCR assay . Genomic DNA was extracted from host tissues using a QIAGEN DNeasy Blood & Tissue Kit ( QIAGEN , Hilden , Germany ) , according to the manufacturer’s instructions , and stored at -20°C until use . The complete mt genomes ( mtDNA ) of various cestodes ( Table 1 ) available in GenBank ( http://www . ncbi . nlm . nih . gov/ ) were retrieved to facilitate design of primers specific for E . granulosus s . s . , E . multilocularis and E . shiquicus ( Table 1 ) . The sequences were aligned automatically using Clustal in MEGA5 . 0 [21] . Primer pairs , expected to be specific for E . granulosus s . s . ( S1 Fig ) , E . multilocularis ( S2 Fig ) and E . shiquicus ( S3 Fig ) , were thus obtained . After some preliminary experimentation , one pair of primers specific for each Echinococcus spp . was selected for inclusion in the multiplex PCR assay . Sequences of these primers , target genes and other related information are presented in Table 2 . PCR amplification was carried out in a 25 μl mixture containing 2 μl dNTPs ( 2 . 5 mM of each ) , 2 . 5 μl 10× ExTaq Buffer ( Mg2+ free ) , 2 μl MgSO4 ( 25 mM ) , 0 . 25 μl ExTaq DNA polymerase ( 5U/μl ) ( TaKaRa , Dalian , Liaoning ) , 100 pg DNA template of each Echinococcus sample , and all three primer pairs were added according to the final concentrations given in Table 2 . Fragments were amplified using the following optimized thermocycling conditions: 95°C/5 min for denaturation; 30 cycles of 94°C/30 sec , 55°C/30 sec , 72°C/40 sec; and 72°C/10 min extension . For all the multiplex PCR assays , positive DNA ( DNA templates of the three Echinococcus spp . ) and negative ( no-DNA ) controls were included . Amplicons were visualized by electrophoresis in 2 . 0% ( w/v ) agarose gels in 1×TAE ( 40 mM Tris-acetate , 2 mM EDTA , pH 8 . 5 ) , stained with ethidium bromide ( EB ) , and viewed under UV light . The fragments were purified using an agarose Gel DNA Purification Kit ( TaKaRa , Dalian , Liaoning ) , and then cloned into pMD18-T Simple vectors using a TA cloning strategy . The recombinant vectors were identified by enzyme digestion and at least two clones for each DNA region were sequenced by the Shanghai Invitrogen Biotechnology Co . Ltd .
Infections of E . granulosus s . s . and E . multilocularis were successfully achieved in all the experimentally infected dogs with 5539 , 8562 , 12535 , 18932 and 20775 E . granulosus s . s . and 2893 , 3153 , 3762 , 3864 and 5322 E . multilocularis adult worms being recovered from each group of 5 dogs that were fed with protoscoleces of each species . No adult worms were found in any of the 5 dogs fed larval E . shiquicus . None of the stray dogs was found harboring E . shiquicus or E . multilocularis; only E . granulosus s . s . adult worms were found in their intestinal contents ( identity confirmed by both morphology and cox1 sequencing ) . Worm burdens were relatively low ( circa 100–200 worms ) in the ten stray dogs examined . Expected PCR products of 219 , 584 and 471 bp were obtained for E . granulosus s . s . ( nad1 ) , E . multilocularis ( nad5 ) and E . shiquicus ( cox1 ) , respectively ( Fig 1 ) , and products of mixed templates of the three Echinococcus species are shown in Fig 2 . The multiplex PCR products contained 3 DNA bands ( 219 , 471 and 584 bp ) with mixed DNA templates of E . granulosus s . s . , E . multilocularis and E . shiquicus; 2 DNA bands ( 219 and 584 bp ) with E . granulosus s . s . and E . multilocularis DNA templates; 2 DNA bands ( 219 and 471 bp ) with E . granulosus s . s . and E . shiquicus DNA templates; and 2 DNA bands ( 471 and 584 bp ) with E . multilocularis and E . shiquicus DNA templates . DNA sequences of these products corresponded in each case with the relevant reference sequences in GenBank: E . granulosus ( G1 ) ( NC_008075 ) [22] , E . multilocularis ( NC_000928 ) [23] and E . shiquicus ( NC_009460 ) [24] .
China is the most severe pandemic country for cystic echinococcosis ( CE ) , in humans and livestock , due mainly to E . granulosus s . s . , and for alveolar echinococcosis ( AE ) due to E . multilocularis in humans and small wild mammals . E . shiquicus is also endemic although it has not been reported to infect humans . Dual infections of animal hosts with different Echinococcus spp have been reported in the eastern Qinghai-Tibet plateau region of China [4 , 25] . The very close relationship between dogs and humans can lead readily to human infection . The increasing number of human AE and CE cases in northwestern China , where considerable numbers of dogs are present , places a heavy burden on public health and veterinary services . To aid surveillance , management and diagnosis , effective methods are needed for rapid and accurate detection and identification of different life cycle stages of the three Echinococcus spp . simultaneously . The multiplex PCR assay developed in this study provides such a method . Traditional epidemiological surveys for tapeworms often involve recovery of eggs from feces of potential definitive hosts . However , morphological identification of Echinoccocus eggs to species level is practically impossible , prompting the development of several molecular approaches [26 , 27] . Inhibitors present in fecal material that co-purify with parasite DNA extracted from feces often present a problem for PCR-based methods [28] . In this study , the QIAGEN QIAamp DNA Stool Mini Kit , containing InhibitEX tablets for removing inhibitors in fecal samples , was used to purify copro-DNA . The sieving-flotation method was helpful in overcoming this problem due to its enrichment of worm eggs [29] . The positive control ( protoscolex DNA in fecal samples ) used in this study demonstrated the lack of inhibitor effects in our copro-multiplex PCR assay . E . granulosus s . s . has been reported as having a pre-patent period of 6 weeks ( 42 days ) [30 , 31] , while E . multilocularis eggs have been observed in feces at 42–46 days post infection [32] . However , in the current study we first identified eggs of E . granulosus s . s . at 47–56 days post-challenge and those of E . multilocularis at 36–44 days post-challenge by microscopy similar to reports by others [30 , 33] . The discrepancies between these studies may be due to the use of different dog-breeds , ages , nutrient status or the conditions under which the dogs were maintained . We were unable to experimentally infect dogs with E . shiquicus although the viability of the challenge sample of protoscoleces was confirmed by microscopy . PCR-positive signals in this study were obtained from dog fecal samples much earlier ( 17 days for E . multilocularis and 26 days for E . granulosus ) than any other previous studies using microscopy as a method of detecting infected canid hosts . The much earlier detection of an Echinococcus infection by the multiplex PCR method compared with egg recovery from feces and microscopic examination is a marked improvement that can aid surveillance programs aimed at preventing echinococcosis transmission . The method developed in this study has achieved high species specificity because it produced no amplicon from any other helminth ( including several that might dual infect with Echinococcus species in dogs ) or from the negative copro-samples ( no-DNA ) . The primer set ( three pairs of primers ) multiplex reaction in a single tube worked well with all templates tested and yielded specific amplicons of the expected length for each of the three Echinococcus spp . examined . E . granulosus s . s and E . multilocularis are of major public health concern in many endemic countries globally [34] . A cost effective diagnostic tool is required for echinococcosis surveillance of definitive and intermediate hosts , and for monitoring the effectiveness of control programs . The multiplex PCR assay developed in this study provides an effective method that can be applied in both clinical and epidemiological settings for the identification of Echinococcus spp in diverse hosts , and would be particularly useful for identifying infected hosts in areas co-endemic for AE and CE . In this study , we focused on Echinococcus samples collected from the Qinghai-Tibet plateau region of China , where three species ( E . granulosus , E . multilocularis and E . shiquicus ) are known to be endemic . In total , nine species are now recognized in the genus Echinococcus , including E . granulosus sensu stricto ( genotypes G1-G3 ) , E . equinus , E . canadensis ( genotypes G6 , G7 , G8 and G10 ) , E . ortleppi , E . multilocularis , E . shiquicus , E . vogeli , E . oligarthrus and E . felidis [1] . None of the three specific pairs of primers developed in this study produced a PCR-amplified product using DNA isolated from E . canadensis ( G6 genotype ) showing in Fig 3 ( the lane 2 with non-band as a negative result ) . This is supported by inspection and comparison of the primer target sequence for the G6 genotype with those of the three Echinococcus spp . , which showed six base pair differences between them ( S1 Fig , S2 Fig and S3 Fig in the Supporting Supplementary Information ) . Furthermore , six or more base pair differences are apparent between the target sequences for E . equinus , E . canadensis ( genotypes G7 , G8 and G10 ) , E . ortleppi , E . vogeli , E . oligarthrus and E . felidis . Therefore , it is highly unlikely that any amplicon would be produced from these species during the multiplex PCR due to its high species specificity . | The canid adapted intestinal tapeworms , Echinococcus granulosus , E . multilocularis and E . shuiqucus are well known to be endemic in Northwestern China . The first two species can cause fatal disease in humans . Although E . shiquicus has not been reported to infect humans , all three species can be transmitted by dogs . The very close relationship between dogs and humans can readily lead to human infection . To aid the surveillance and management of echinococcosis , effective diagnostic approaches are urgently needed . We developed a single tube multiplex PCR assay for the accurate identification and discrimination of the three Echinococcus species for use in both clinical diagnosis and epidemiological studies . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | Discrimination between E. granulosus sensu stricto, E. multilocularis and E. shiquicus Using a Multiplex PCR Assay |
Tonotopy is a fundamental organizational feature of the auditory system . Sounds are encoded by the spatial and temporal patterns of electrical activity in spiral ganglion neurons ( SGNs ) and are transmitted via tonotopically ordered processes from the cochlea through the eighth nerve to the cochlear nuclei . Upon reaching the brainstem , SGN axons bifurcate in a stereotyped pattern , innervating target neurons in the anteroventral cochlear nucleus ( aVCN ) with one branch and in the posteroventral and dorsal cochlear nuclei ( pVCN and DCN ) with the other . Each branch is tonotopically organized , thereby distributing acoustic information systematically along multiple parallel pathways for processing in the brainstem . In mice with a mutation in the receptor guanylyl cyclase Npr2 , this spatial organization is disrupted . Peripheral SGN processes appear normal , but central SGN processes fail to bifurcate and are disorganized as they exit the auditory nerve . Within the cochlear nuclei , the tonotopic organization of the SGN terminal arbors is blurred and the aVCN is underinnervated with a reduced convergence of SGN inputs onto target neurons . The tonotopy of circuitry within the cochlear nuclei is also degraded , as revealed by changes in the topographic mapping of tuberculoventral cell projections from DCN to VCN . Nonetheless , Npr2 mutant SGN axons are able to transmit acoustic information with normal sensitivity and timing , as revealed by auditory brainstem responses and electrophysiological recordings from VCN neurons . Although most features of signal transmission are normal , intermittent failures were observed in responses to trains of shocks , likely due to a failure in action potential conduction at branch points in Npr2 mutant afferent fibers . Our results show that Npr2 is necessary for the precise spatial organization typical of central auditory circuits , but that signals are still transmitted with normal timing , and that mutant mice can hear even with these deficits .
The sense of hearing is mediated by precisely organized neural circuits that encode the frequency content , timing , and intensity of sounds . Frequency information is encoded in the spatial organization of hair cells in the cochlea , with high frequencies detected in the base and low frequencies in the apex . SGNs transmit this information to the cochlear nuclei , where their axons bifurcate into an ascending branch that innervates the aVCN and a descending branch that targets the pVCN and DCN . In each of these regions , the systematic innervation by SGN fibers forms frequency maps that maintain the tonotopic order that is established in the cochlea and that is preserved along the auditory pathway . Tonotopy also governs intrinsic connections between neurons in the cochlear nuclei , including tuberculoventral cell projections from the DCN to the VCN [1] , [2] . SGN axons are responsible for delivering all acoustic information from the cochlea to the cochlear nuclei . By contacting a variety of target neurons with distinct projection patterns , each SGN feeds information to parallel pathways in the brainstem [3] . Through their ascending branches , SGNs convey auditory signals to bushy cells that are involved in comparing interaural time and intensity for localizing sounds in azimuth [4] , [5] , [6] , [7] , as well as to some T stellate cells . Through their descending branches , SGNs innervate T stellate cells that encode the spectrum of sounds [8] , [9] , [10] , octopus cells that mark the onset of sounds [11] , [12] , [13] , and fusiform and giant cells of the DCN that use spectral cues to localize sounds monaurally in the vertical plane [14] . Together , the activation of these diverse populations of cochlear nuclear neurons by SGN axons enables animals to detect , recognize , and locate sounds in their environment . In order to make the precise pattern of diverse connections that enable the interpretation of sound , developing SGN axons must elaborate a variety of synapses that are tonotopically organized but that show distinct signaling properties depending on the nature of the target neuron . For instance , within one isofrequency lamina of the VCN , SGN axons contact bushy cells , T stellate , and octopus cells and form functionally distinct synapses with each cell type . The branches that innervate bushy cells terminate in unusually large and complex endbulbs of Held that mediate large post-synaptic responses that depress , yet still signal with high temporal precision [15] , [16] . In contrast , on T stellate cells , SGN axons form typical bouton endings that induce smaller post-synaptic responses with less depression . Traditionally , the mechanisms that control axon guidance and synaptic function have been studied independently . However , recent evidence indicates that these two events can in fact be linked , as maturation of the calyx of Held does not progress normally in bushy cell axons that fail to cross the midline [17] . Whether the spatial organization of SGN axons is similarly coordinated with subsequent synaptogenesis remains unclear . One of the primary obstacles towards understanding how functional auditory circuits are assembled is the lack of genetic mutations that disrupt SGN central wiring . As a result , there are no clear predictions for how changes in the pattern of central innervation might impact hearing either in mice or humans . Indeed , although a growing number of genes affecting cochlear function have been implicated in sensorineural deafness in humans [18] , almost nothing is known about the genetic basis of central auditory processing disorders , which disrupt central auditory circuit function without obvious loss of hearing sensitivity [19] , [20] . Identifying and characterizing genetic mutations that affect the formation of central auditory circuits in mice is an important step towards understanding how these disorders may arise . During the course of a screen to identify genes required for auditory circuit assembly , we discovered that the natriuretic peptide receptor Npr2 is required for central axon bifurcation in SGNs [21] . Npr2 is a receptor guanylyl cyclase that activates a cGMP-dependent protein kinase signaling cascade upon binding to the C-type natriuretic peptide ( CNP ) , which is expressed dorsally along the length of the embryonic neural tube [22] , [23] . In Npr2 mutant mice , the axons of both dorsal root ganglion ( DRG ) and cranial ganglion neurons fail to bifurcate [24] , [25] . However , interstitial branches can still form from the unbifurcated axons . These observations suggest that Npr2 ensures that bifurcations form in an orderly manner as sensory axons enter the spinal cord and encounter CNP , but that other mechanisms determine how and when axons arborize within their targets [26] . Although Npr2 signaling leads to changes in cytoskeletal-associated proteins [27] , [28] , [29] , how Npr2 affects growth cone behavior is unclear , as CNP can also act as a chemoattractant [28] . In addition , the full extent of Npr2's effects on the organization and function of the mature nervous system has not yet been defined . Since NPR2 is mutated in human patients with achondroplasia [30] , a deeper understanding of the Npr2 phenotype in an animal model is needed . Here , we seek to gain insight into the mechanisms that govern SGN central wiring by characterizing the long term effects of Npr2 mutations on both the spatial organization and functional maturation of synaptic connections between SGN axons and their cochlear nuclear targets .
Previous studies established that Npr2 is absolutely required for bifurcation of sensory neuron axons with no obvious effects on the peripheral processes [24] , [25] , highlighting the utility of the Npr2 mutant mouse as a model for central auditory wiring defects . However , the peripheral auditory system has not yet been examined . To investigate whether there are any obvious changes in the innervation of the mature cochlea , we evaluated the gross organization of SGN peripheral processes after the onset of hearing , which is at about postnatal day 12 ( P12 ) in mice [31] . Neurofilament immunostaining revealed no obvious differences in the overall pattern of cochlear innervation at P18 , with orderly arrays of radial bundles present in both Npr2 mutants ( n = 2 ) and controls ( n = 2 ) ( Fig . 1A , B ) . Moreover , individual SGN fibers , as visualized at P14 by crossing Neurog1-CreERT2 to AI14-tdTomato reporter mice , exhibited normal morphologies , with single , unbranched peripheral processes extending directly towards the organ of Corti both in controls ( n = 3 ) and mutants ( n = 5 ) ( Fig . 1C , D ) . Similarly , the SGN central processes in the auditory nerve peripheral to the bifurcation in the nerve root seemed normal in Npr2 mutant mice at P21 , as assessed by electron microscopy ( Fig . 1E , F ) . The g-ratio , the ratio of the axon diameter to the total myelinated fiber diameter , did not differ significantly between control ( n = 3 ) and Npr2 mutant ( n = 3 ) animals ( P = 0 . 87 ) ( Fig . 1G ) . In both groups , the observed ratio was near the optimal for conduction [32] . Hence , development of the peripheral auditory system appears normal in the Npr2 mutant strain , which therefore offers a useful model for examining the anatomical and functional consequences of central auditory wiring defects . Previously , we showed that Npr2 is expressed in SGNs and is required for the bifurcation of central SGN axons at E12 . 5 [21] , and independent studies have shown a complete absence of axon bifurcation in these and other sensory neuron populations in Npr2 mutants [24] , [25] . Although DRG bifurcation defects have been shown to persist and ultimately disrupt the functional connectivity of the mature spinal cord [24] , the long term effects of the Npr2 mutation for central auditory wiring remain poorly characterized . To determine whether SGN central axons acquire additional defects , we characterized Npr2 mutants at E16 . 5 , when both branches have formed and projected tonotopically within the developing cochlear nuclei ( Fig . 2A ) [33] . Lipophilic dye labeling revealed that in Npr2 mutants ( n = 4 ) , the cochlear nerve root lacked the Y-shaped morphology typical of control animals ( n = 4 ) ( Fig . 2B , C ) , consistent with a persistent bifurcation defect . In addition , whereas SGN axons were neatly bundled both proximal and distal to the bifurcation of the nerve in control animals , Npr2 mutant axons were disorganized in the region where they would normally branch ( Fig . 2B , arrow ) , resembling the first exploratory SGN axons that reach the hindbrain at E11 . 5 in wild-type embryos [21] . We next examined SGN central projections in Npr2 mutant animals between P14–P18 , after the auditory circuit is fully formed and functional hearing has begun . To label SGN axons , Neurog1-creERT2 mice were crossed to AI14 tdTomato reporter mice , resulting in offspring with tdTomato expression in a random subset of SGNs due to leaky Cre activity in the absence of tamoxifen . In Neurog1-CreERT2;AI14 mice , tdTomato expression in SGNs is sparse enough that neuronal morphology can be examined , but is distributed uniformly along the length of the cochlea so that the overall pattern of SGN innervation throughout the intact cochlear nuclear complex can be visualized and qualitatively assessed in cleared tissue . In control animals ( n = 5 ) , auditory nerve fibers projected in a highly stereotyped and ordered manner to the aVCN , pVCN , and DCN ( Fig . 2D ) . In contrast , SGN afferent innervation was consistently disrupted in all Npr2 mutants ( n = 6 ) ( Fig . 2E ) . SGN axons were able to reach all three divisions , but exhibited several signs of disorganization that were not observed in controls . First , whereas control axons formed distinct bifurcations that aligned with each other within the nerve root , we could not recognize an obvious zone of bifurcation in Npr2 mutants . Moreover , a closer look at sections through the cochlear nuclei revealed a severe disorganization of projections in the aVCN and pVCN in Npr2 mutants ( Fig . 2G ) compared to controls ( Fig . 2F ) . Control axons exhibited clear bifurcations in the VCN ( Fig . 2H ) , resulting in neat bundles of axonal branches in the aVCN ( Fig . 2H′ ) . In contrast , Npr2 mutant SGN axons generally turned instead of bifurcating ( Fig . 2I ) and often followed aberrant paths in the aVCN ( Fig . 2I′ ) . Additionally , some patches of the aVCN appeared to be underinnervated ( Fig . 2E , arrowheads ) . Thus , loss of axon bifurcation and abnormal trajectories persist even after the onset of hearing in Npr2 mutant mice . Although the overall pattern of innervation was clearly abnormal , SGNs projections were nevertheless present in all divisions of the Npr2 mutant cochlear nuclei ( Fig . 2E ) . Given that Npr2 is required for axon bifurcation but not for development of collaterals [24] , [25] , we hypothesized that unbifurcated SGN axons might eventually form interstitial branches that are able to grow into other regions of the cochlear nuclei . To determine whether SGNs innervating the aVCN can still form branches that project to the pVCN , we labeled fibers with biocytin injections into the aVCN and searched for labeled fibers in the pVCN . In control animals ( n = 49 ) , such injections labeled the ascending branch retrogradely to the nerve root and they labeled the descending branch anterogradely through the pVCN and into the DCN ( Fig . 3A ) . The labeled descending branches formed a tight bundle in the octopus cell area , where the SGN fibers converge on their way to the DCN ( Fig . 3A′ ) . In Npr2 mutant animals ( n = 28 ) , obvious bundles were never seen . However , a few widely spread fibers in the octopus cell area were consistently labeled ( Fig . 3B , B′ ) , indicating that some individual SGN fibers managed to innervate both regions in spite of the axon bifurcation defect . Since there are no molecular markers to distinguish bifurcations from interstitial branches , we instead relied on morphological criteria to recognize bifurcations . Bifurcations are usually the first branch points within the cochlear nuclei , are found in a predictable location , and exhibit a characteristic Y-shaped morphology . No branch point in any of the 34 mutant cochlear nuclei in which biocytin was injected into the aVCN or into the cut end of the nerve displayed the morphology of normal bifurcations . Whereas in controls , the parent axon gave rise to two equally thick branches at roughly 120° angles ( Fig . 3C ) , branches in the vicinity of the nerve in Npr2 mutants exhibited more varied angles and one branch was often abnormally thin ( Fig . 3D , arrows ) . Additionally , Npr2 mutant SGN axons extended branches ( Fig . 3G , arrowheads ) that resembled the interstitial branches found in controls ( Fig . 3E , arrowheads ) . Thus , it seems likely that the branches that SGN axons form in Npr2 mutant mice are interstitial branches rather than true bifurcations . Nonetheless , axonal branches in the aVCN terminated in endbulbs of Held with the usual range of sizes and shapes in both controls ( Fig . 3F ) and mutants ( Fig . 3H ) , indicating that despite their defective branching patterns and trajectories , Npr2 mutant SGN axons are still able to find appropriate targets and make specialized synapses with normal morphology . Thus local interactions seem to be able to govern synaptogenesis independent of the changes in axon trajectory . Since the peripheral organization of SGN projections in the cochlea appeared unaffected in Npr2 mutants , SGNs are predicted to receive sharply tuned frequency information from hair cells . However , given the disorganization of SGN central axons in Npr2 mutants , we wondered whether mutant SGN axons preserve the tonotopic order of their projections as they exit the cochlea and find their way into the cochlear nuclei . Crystals of the lipophilic dyes , DiI and DiD , were inserted into the apical and basal turns of the cochlea in fixed E16 . 5 mouse heads and the dye was allowed to diffuse anterogradely through SGN axons to the hindbrain ( Fig . 4A ) . In control animals ( n = 2 wild-type and 2 heterozygote ) , SGN axons were tonotopically segregated within the eighth nerve , and their bifurcation points fanned out in tonotopic order within the developing cochlear nuclei , with axons from more basal SGNs bifurcating more dorsally than apical SGNs ( Fig . 4B ) . The gross tonotopic segregation observed in control embryos was maintained in Npr2 mutants ( n = 4 ) ( Fig . 4C , C′ ) . However , in some Npr2 mutant embryos ( n = 2/4 ) , intermingling of apical and basal projections was observed ( Fig . 4C , C′ , arrowheads ) , suggesting imprecise tonotopy . Additionally , mutant axons appeared to project more strongly towards what will become the pVCN and DCN , quantified by comparing the fluorescence intensity of the branches projecting rostrally vs . caudally ( P<0 . 05 ) ( Fig . 4D ) . Apical SGNs were more strongly affected than basal SGNs and showed a stronger bias towards the developing pVCN and DCN . To determine whether the blurring of tonotopy persists through the onset of hearing , similar dye labeling of SGNs was performed at P14 by placing DiI and DiD crystals in the apical and mid-turns of the cochlea , respectively ( Fig . 4E ) . In controls ( n = 2 wild-type ) , clear segregation of the two dyes was observed in the eighth nerve ( Fig . 4F ) and this segregation was maintained both in the aVCN ( Fig . 4F′ ) and pVCN ( Fig . 4F″ ) , as assessed using confocal imaging . In Npr2 mutants ( n = 4 ) , the axons from apical and mid-turn SGNs were also appropriately segregated within the eighth nerve ( Fig . 4G , H ) , confirming that the dyes labeled distinct populations of neurons in the cochlea . However , the projections overlapped extensively in the aVCN ( Fig . 4G′ , H′ ) and/or pVCN ( Fig . 4G″ , H″ ) . Some overlap was apparent in all of the mutants; variability in precise size and location of the dye crystals prevented quantification of the degree of mixing . Thus , tonotopic segregation appears normal in the auditory nerve , but is degraded within the cochlear nuclei of Npr2 mutants . The abnormal tonotopic organization of SGN projections raised the question of whether intrinsic neuronal circuits within the cochlear nuclei are similarly disrupted . Tuberculoventral ( TV ) cells are glycinergic neurons that reside in the deep layer of the DCN and innervate targets in the aVCN and pVCN , forming a negative feedback circuit . They are tonotopically arranged , receiving input from the same auditory nerve fibers as their targets and therefore exhibit similar tuning [1] , [2] , [34] . The pattern of TV cell connectivity was examined by injecting biocytin into the aVCN , which normally labels TV cell bodies in the DCN as well as SGN afferent fibers that project to that isofrequency band [1] ( Fig . 5A–B ) . Labeling follows the tonotopic organization of the cochlear nuclei: dorsal injections labeled bands of TV cells dorsally in DCN , whereas ventral injections labeled TV cells in a more ventral position in the DCN . In control animals ( n = 16 wild-type and 33 heterozygote ) , a few labeled cells in the DCN were located ventral to the isofrequency band , because their axons crossed the injection site to innervate more ventral regions of the aVCN ( Fig . 5A , arrowhead ) . However , labeled cells dorsal to the band in the DCN were not observed in normal animals , indicative of the sharp tonotopic organization ( Fig . 5A , D ) . In contrast , in Npr2 mutant mice ( n = 28 ) , the labeled cell bodies were found over a large span of the DCN , even when the injections were made ventrally in the aVCN ( Fig . 5B , D ) . To quantify this result , the distribution of labeled cells along the tonotopic axis was measured in reconstructions of 37 slices with injections into the ventral half of aVCN ( Fig . 5C ) . In control mice ( n = 26 cochlear nuclei from 12 heterozygote and 4 wild-type mice ) , the distribution of labeled cells aligned at their peaks showed a sharp peak that tapered ventrally toward the granule cell lamina , with an average half-width of 132±80 µm . In Npr2 mutants ( n = 11 cochlear nuclei from 7 mice ) , the distributions lacked sharp peaks . The average distribution of labeled cells , aligned on the median , was significantly broader , with an average half-width of 276±130 µm ( P<0 . 001 ) , reflecting the more diffuse organization observed within individual cochlear nuclei . These findings indicate that the tonotopic organization of the TV cell projection in mutant cochlear nuclei is less precise than in control animals , consistent with the overall disruption in SGN axon topography shown by genetic and dye labeling . Our anatomical studies show that although the innervation of the cochlea is not altered noticeably , there is a consistent and striking change in SGN central axonal innervation patterns in Npr2 mutant mice . While changes in the periphery are well-known to diminish auditory sensitivity , how a loss of precision in the organization of SGN inputs to the cochlear nuclei might affect hearing is unclear . To address this question , we compared auditory brainstem responses ( ABR ) in six-week old wild-type and Npr2 mutant mice . ABRs are generated by the synchronous firing of groups of aligned axons . In cats , the first large positive and negative waves reflect the firing of axons of SGNs , the second positive wave reflects the firing of neurons in the VCN that lie near the nerve root , the third positive wave reflects activation of the VCN rostral and caudal to the nerve root and the superior olivary complex , and later waves reflect the summation of activity at many stages of the auditory pathway [35] . Similar waveforms are observed in mice; it is broadly accepted that the first two peaks reflect activity in the nerve and cochlear nuclei as in cats [36] , [37] . No significant difference was observed between control ( n = 6 wild-type ) and Npr2 mutant ( n = 15 ) mice in the shape or amplitude of the early peaks in responses to 16 kHz tones , which activate the most sensitive regions of the cochlea in mice [38] ( P>0 . 3 for the amplitudes of peaks one and two at all sound pressure levels ) ( Fig . 6A , B ) . The normal average ABR waveforms confirm the absence of obvious peripheral defects and suggest that the timing of firing of SGNs and of their targets in the VCN is also apparently normal . In addition , ABR thresholds did not differ significantly between wild-type and Npr2 mutant mice ( P = 0 . 38 ) ( Fig . 6C ) . Thus , within the resolution of these measurements , the sensitivity and timing of firing of auditory neurons in the brainstem seem normal in Npr2 mutants . It should be noted that Npr2 mutant mice exhibit additional abnormalities , including dwarfism and cardiac deficits [39] that compromise their health and often cause them to die within the first postnatal month . Thus , it is possible that no obvious ABR phenotype was observed because the animals that survived to the testing stage were the healthiest and least abnormal . However , cochlear nuclear innervation defects were fully penetrant and varied only in severity . Moreover , since it is difficult to establish behavioral baselines in these animals , we were unable to use pre-pulse inhibition of the acoustic startle reflex to test for deficits in specific hearing tasks , such as frequency discrimination , gap detection , and sound localization . Although ABRs did not reveal any significant differences in auditory responsiveness in Npr2 mutant mice , this method assesses the overall activity of the population , leaving open the possibility that individual cells may not transmit signals normally . To determine whether Npr2 mutant axons are indeed able to develop normal synapses despite the change in branching patterns and trajectory , we made intracellular recordings in slices . Cochlear nuclear neuronal responses to sound depend on the pattern of convergence of synaptic inputs , the physiological properties of those inputs , and the electrical properties of target neurons that shape the voltage responses to synaptic currents . Whole-cell patch recordings in slice preparations of the cochlear nuclei confirmed that the three principal cell types of VCN ( bushy , octopus , and T stellate cells ) , recognizable by the differences in their intrinsic electrical properties , are present in Npr2 mutants . In Npr2 mutants as in control animals , bushy and octopus cells fire transiently in response to depolarizing current pulses , whereas T stellate cells respond with trains of action potentials that last for the duration of the depolarization , in both wild-type and mutant animals [10] , [12] , [40] , [41] ( S1 Figure ) . Comparison of wild-type ( n = 3 ) and mutant ( n = 4 ) bushy cell properties revealed no change either in the resting potential ( −65±1 . 7 mV in wild-type vs . −66±2 . 1 mV in Npr2 mutant ) or input resistance ( 92±8 MΩ in wild-type vs . 97±9 MΩ in Npr2 mutant ) ; the properties of T stellate ( n = 2 ) and octopus cells ( n = 2 ) were also within the normal range . The absence of any measurable differences in the intrinsic properties indicates that mutant neurons are capable of signaling as rapidly and precisely as the wild type . Another important determinant of acoustic signal transmission is the number of SGN inputs that contact each target neuron , which ranges from few ( for bushy and T stellate cells ) to many ( for octopus cells ) . Given the abnormal trajectories of SGN axons seen within the cochlear nuclei , we asked whether SGNs would still converge normally on principal cells of the VCN in Npr2 mutants . The number of excitatory inputs that converge on a recorded cell can be estimated by measuring the growth of synaptic responses to shocks of fiber bundles as the shock strength is gradually increased because the synaptic response grows in steps as additional fibers are brought to threshold [16] . The number of steps in the increase in synaptic current is thus an estimate of the number of excitatory inputs . Bushy cells receive converging input from a small number of SGNs . In the mutants , as in control animals , some jumps were small and others were large , reflecting the fact that bushy cells in mice receive input from small bouton endings as well as large endbulbs of Held [7] , [16] , [42] ( Fig . 7A ) . The number of converging inputs to bushy cells in Npr2 mutants fell into the normal range , between 1 and 6 [16] . However , a surprisingly large proportion had only a single input ( 5/7 in Npr2 mutants , compared with 5/21 similar bushy cells in a wild type strain [16] . Interpretation of these findings is complicated by the fact that there are multiple types of bushy cells: globular bushy cells that project to the medial nucleus of the trapezoid body and spherical bushy cells that project to the lateral or medial superior olivary nuclei . The sole electrophysiological distinction between these cell types in slices is the number of converging inputs [7] , [16] . Especially when the aVCN is disorganized , it is impossible to know whether populations of different types of bushy cells were sampled equally . However , our results suggest that bushy cells in the aVCN in Npr2 mutants likely receive input from fewer SGNs than normal . To see whether other target neurons in aVCN also receive fewer inputs , we performed a similar analysis of T stellate cells , which are present both in aVCN and pVCN . Shock-evoked synaptic responses in T stellate cells normally grow with between five and eight steps , each delivering roughly equal steps of current of between 100 and 300 pA [16] , [43] ( Fig . 7B ) . Many of these responses are likely to arise from SGNs but some could also arise from other T stellate cells [43] . In control mice , no differences have been reported between T stellate cells in pVCN , where they are most abundant , and in the aVCN [40] , [44] , [45] . In Npr2 mutants , 6/10 of the T stellate cells we recorded were in the pVCN , near the octopus cell area . Convergence of inputs in these cells was normal , averaging 7 . 5±1 ( n = 6 ) . In contrast , in the 4/10 T stellate cells that were recorded more anteriorly , evoked responses grew in significantly fewer current steps ( 3 . 5±0 . 6 , n = 4 ) ( P<0 . 001 ) ( Fig . 7B ) . Octopus cells reside in pVCN and would therefore not be expected to show a similar change in the number of SGN inputs . However , these cells are so heavily innervated by SGNs that it is not possible to estimate the actual number from the growth of synaptic responses with shock strength [12] , [13] . Instead , the synaptic responses generally grow in steps so small that the growth appears graded . In Npr2 mutant mice , the growth of synaptic responses showed more irregularity than we have observed in CBA or ICR mice [12] , [16] , but this irregularity was also observed in control mice . No differences in convergence between wild-type and mutant mice could be resolved in octopus cells ( Fig . 7C ) . Together , these data suggest that in Npr2 mutants , convergence of SGNs onto bushy and T stellate cell targets in the aVCN is reduced , while in the pVCN , convergence onto T stellate cells and octopus cells is normal . This subtle change in circuit organization is consistent with our finding that SGN projections are biased towards the pVCN and DCN at embryonic stages . To determine whether the observed changes in the pattern of connectivity are accompanied by changes in the nature of transmission between SGNs and their cochlear nuclear targets , we examined the pattern of synaptic responses to trains of shocks . In wild-type mice , repeated stimulation of the auditory nerve consistently evokes synaptic responses , although when driven at high rates , synaptic responses show depression , with a stronger effect in bushy than in T stellate cells [15] , [16] , [46] . Synapses between SGNs and principal neurons in the VCN in Npr2 mutants ( 22 cells from 22 animals ) exhibited the expected synaptic depression observed in wild-type and heterozygous animals ( 21 cells in 10 wild type and 11 heterozygote mice ) ( Fig . 8A , A′ ) . However , Npr2 mutants differed from control animals in that shocks intermittently failed to evoke any responses in some neurons . For instance , in 7/12 bushy cells , some of the shocks in a train failed to evoke a response ( Fig . 8A ) . Failures were sporadic and complete , with no synaptic response at all in the target neuron ( Fig . 8A″ ) . A similar phenotype was also detected in T stellate cells ( Fig . 8B ) , with 3/10 cells sporadically failing to respond to shocks; in contrast , 0 of 10 wild type and heterozygote responses failed . One reason failures may have been detected in relatively fewer T stellate cells than in bushy cells is that failures of small inputs are difficult to detect . Indeed , in T stellate cells failure was often incomplete in that small ( <10% ) synaptic current remained , presumably because the larger of two inputs failed while the smaller one did not . For both bushy and T stellate cells , failures occurred even after the first shock in the train , when depletion of neurotransmitter is not an issue ( Fig . 8C ) . Together with the all-or-none character of the failures , these findings suggest that action potentials sometimes fail to invade the SGN synaptic terminals . To test whether a conduction block could be overcome by making action potentials in the parent axon taller and/or wider , we applied a low concentration of 4-aminopyridine ( 4-AP ) , a non-specific blocker of K+ channels used to relieve conduction block in patients with multiple sclerosis [47] . Indeed , 0 . 1 mM 4-AP eliminated synaptic failures reversibly in both bushy and T stellate cells ( Fig . 8D , E ) . These results support the idea that Npr2 mutant auditory nerve axons suffer from blocks in action potential conduction . Importantly , the responses that did occur showed normal , precise temporal tracking of inputs ( Fig . 7 ) . Thus , our data indicate that Npr2 mutant mice exhibit altered spatial organization of the auditory circuit and less reliable action potential conduction , yet still maintain the overall temporal precision of auditory signal transmission .
Although recent studies have uncovered a number of genes required for cochlear wiring [48] , how SGN central axons navigate to the cochlear nuclei is poorly understood . SGN axons reach the hindbrain and start to bifurcate by E12 in mice [21] . When the aVCN is not present , SGN axons still project to the brainstem and bifurcate [49] , likely because the Npr2 ligand CNP is expressed along the entire rostral-caudal axis of the hindbrain [23] . In addition , since SGN axons enter the hindbrain at the level of rhombomere 4 , which gives rise to the pVCN and DCN [50] , this region may provide attractive cues that are primarily responsible for early SGN guidance decisions . Indeed , we find that Npr2 mutant axons preferentially extend towards the developing pVCN and DCN , indicating that when required to make a directional choice without bifurcating , SGN axons show a caudal bias . Nevertheless , their projections follow aberrant trajectories , suggesting that Npr2 is also required for normal responsiveness to cues in the environment . A direct role for Npr2 in axon guidance has not been clearly shown; although Npr2 mutant DRG axons make occasional guidance errors upon entering the spinal cord , they follow grossly normal paths towards targets in the dorsal and ventral horns [24] . Moreover , expression of CNP is restricted to the dorsal neural tube embryonically [23] , [25] , making it improbable that a CNP-Npr2 interaction directs growth of SGN axons deeper within the developing cochlear nuclei . It therefore seems more likely that the observed guidance defects are secondary to the loss of bifurcation . After bifurcating , developing SGN axon branches must navigate towards distinct regions of the cochlear nuclei while retaining the tonotopic organization established in the cochlea . This fundamental feature of the auditory pathway is established during embryogenesis [33] , [51] and does not require hearing , although it is later refined by activity-dependent mechanisms that need not be driven by sound [52] , [53] . We still know very little about the molecular mechanisms that direct this critical feature of central auditory circuit assembly , and the few examples of mutant mouse strains with disrupted tonotopy are complicated to interpret . For example , tonotopy is abnormal in mice lacking the transcription factor Neurod1 , which acts early during SGN development [54] . However , in these animals , auditory afferents are intermingled with projections from the vestibular endorgans , suggesting that disrupted tonotopic organization is secondary to a general change in neuronal identity . Eph/Ephrin signaling may play a more specific role in central topographic projections , with ephrin-B2 mutants exhibiting abnormally broad frequency bands in the DCN [55] . However , since EphA4/ephrin-B2 signaling is also involved in bundling of peripheral SGN projections extending towards the organ of Corti [56] , the change in frequency responses in these mutants could also arise from peripheral disorganization . Npr2 mutant mice are especially interesting because defects in the tonotopic organization in the cochlear nuclei occur without obvious defects in cochlear organization . SGN axons are topographically ordered in the eighth nerve in Npr2 mutants , but exhibit disorganization in the nerve root and blurred tonotopy in the cochlear nuclei , indicating that trajectories become disarrayed as they enter the auditory brainstem . The phenotype was fully penetrant , with abnormal topography apparent in genetically labeled SGNs , by labeling of the SGN axons with lipophilic dyes , and by biocytin labeling of second order neurons in the cochlear nuclei . Since SGN bifurcation points are tonotopically ordered within the nerve root , with small bundles of SGN axons bifurcating together , it is possible that the abnormal guidance of Npr2 mutant axons exiting the eighth nerve disrupts this bundling , thereby perturbing the local tonotopic order . Indeed , proper fasciculation during axon guidance is known to play a key role in topographic mapping of axons in other systems [57] , [58] . Alternatively , mutant axons may be unable to detect guidance cues in the environment , perhaps because key receptors are not trafficked properly to the branches that do form . In fact , tonotopy worsens over time , with only a few misrouted axons at E16 . 5 but apparent overlap at P14 , as would be expected if the primary problem is defective guidance of the collaterals that sprout from the unbifurcated axons . Higher resolution labeling methods will be needed to discern whether individual fibers extend their primary axons to the tonotopically appropriate location , with blurring due to abnormal guidance of collaterals outside of this area . It is tantalizing to consider whether the disorganization of TV cell projections might also be secondary to the changes in SGN axon branching patterns . However , although expression is initially restricted to sensory neurons [24] , [25] , Npr2 also appears to be transcribed in the cochlear nuclei later in development [59] ( Allen Mouse Brain Atlas ) and could act independently in other populations of neurons . Analysis of cochlear-specific Npr2 conditional knock-outs will be necessary to resolve this issue . Unexpectedly , the loss of tonotopic organization in the central auditory circuits of Npr2 mutant mice results only in subtle changes in auditory function . ABRs , which are generated by the summation of coherent currents and thus largely reflect synchronous firing in bundles of axons [35] are normal in the mutants; intracellular recordings , which assay transmission to individual post-synaptic targets , show that evoked excitatory postsynaptic responses are normal when present but occasionally fail . The absence of any obvious ABR defect in Npr2 mutants is consistent with our anatomical studies , which revealed no abnormalities in peripheral wiring , myelination , axon diameter , or synaptic morphology . It is also consistent with whole-cell patch-clamp recordings from individual neurons , which show that the principal cells of the VCN in Npr2 mutants retain the ability to signal rapidly and with temporal precision . Indeed , the basic features of synaptic transmission were unaffected; EPSC amplitudes , kinetics , depression , and delays in the VCN of Npr2 mutants were in the normal range [16] . The only salient defect observed was an occasional failure in transmission , which would not be expected to alter the timing of signaling in the population of neurons . Moreover , since amplitudes of the first wave did not differ significantly between mutants and controls , either roughly similar numbers of neurons are activated or the smaller heads of mutants compensate for slightly reduced numbers of active neurons . Normal ABRs are also observed in animals in which reorganization of central tonotopic maps is induced by persistent , moderate noise [60] . While this might mean that ABRs are not sensitive enough to detect such changes , it may also reflect the plasticity of central auditory circuits , as has also been described by others [52] . Overall , these physiological studies suggest that Npr2 mutant SGNs are still able to respond to sounds with normal sensitivity and timing , despite the disrupted spatial organization . Thus , in contrast to what has been observed for development of the calyx of Held [17] , functionally normal synapses can form even when SGN axons follow abnormal paths . Although our physiological tests revealed no significant change in auditory responsiveness , it is still unclear whether the blurring of spatial organization represents a functional blurring at the level of frequency discrimination in Npr2 mutants . In wild-type animals , the tuning of bushy and T stellate cells shows similar sharpness to that of auditory nerve fibers [11] , indicating that SGNs that converge onto a single target neuron are similarly tuned . Since frequency coding is likely intact in the cochlea , activity-dependent synapse elimination , not only in the aVCN but also in TV cells of the DCN , could select for appropriate inputs with similar tuning in Npr2 mutant mice even when they are not in the correct spatial location . Thus , the broadening of TV cell isofrequency mapping in Npr2 mutants might reflect appropriate functional connections between cohorts of neurons that transmit similar frequency information , but that are no longer spatially confined to a tight band due to the disorganization of SGN afferents . Although pre-pulse inhibition of the acoustic startle response can be used in mice to test for frequency discrimination [61] , such experiments are not possible with Npr2 mutants , which have dwarfism and cardiac defects , and therefore await generation of Npr2 conditional knockout animals . Although the basic features of synaptic transmission were unaffected by the loss of Npr2 function , auditory signal transmission became less reliable , with some shocks to SGNs failing to produce any response in post-synaptic targets . Since the failures sometimes occurred in the first response of a train , they could not have resulted from the depletion of neurotransmitter . Furthermore , EPSC failures were all-or-none , indicating that some action potentials did not reach the SGN terminals . Additionally , Npr2 mutant axons showed no significant loss of myelin and did not exhibit signs of the increased spike latency or jitter associated with dysmyelinated SGNs [62] . Given the changes in SGN axon branching patterns in Npr2 mutants and our ability to reverse failures with a K+ channel blocker that strengthens action potentials , it is likely that failure occurred at branch points , which have long been recognized as being weak points in conduction [63] . Although a similar functional phenotype has not yet been described in other sensory neurons in Npr2 mutants , DRG neurons do exhibit mildly impaired ability to activate target neurons upon capsaicin treatment [24] , indicating the need for a more detailed analysis of these neurons . Overall , our results suggest that the development of the auditory circuit is robust enough that surprisingly normal synaptic connections can be made even in the face of disorganized topography . It is unlikely that Npr2 mutant mice have completely normal hearing , but more subtle behavioral tests will be required to reveal deficits . Notably , it is estimated that ∼1% of people with normal hearing sensitivity , and therefore normal cochlear function , have defects in their ability to process sound [64] . Unambiguously identifying and characterizing patients with these central auditory processing disorders has been challenging , because multimodal sensory , language , and attention deficits can accompany or mimic central auditory processing disorders , thereby complicating diagnosis [65] . Interestingly , NPR2 mutations cause achondroplasia in humans [30] , suggesting that closer examination of auditory function may be warranted in such patients . The identification of central wiring defects in Npr2 and other mutant mouse strains may lead to more directed clinical analysis of hearing in human patients carrying analogous mutations and therefore improve the diagnosis and classification of these disorders in the future .
The following mouse strains were used: Neurog1-creERT2 mice [33] , AI14-tdTomato mice ( Jackson Laboratories , Stock Number 007908 ) , and Npr2cn mice which carry a missense point mutation ( L885R ) in the guanylyl cyclase domain of the Npr2 gene that prevents the protein from catalyzing cGMP formation [39] ( Jackson Laboratories , Stock Number 003913 ) . Animals were maintained on a mixed genetic background . Mice were genotyped using previously described PCR protocols ( Jackson Laboratories , [33] , [39] ) . For timed pregnancies , embryonic day 0 . 5 ( E0 . 5 ) was defined as noon on the day of a copulatory plug . In most cases , animals were euthanized using C02 exposure followed by cervical dislocation or anesthetized with ketamine/xylazine and then either cervically dislocated or perfused transcardially with fixative . For physiological studies , young animals were decapitated with colostomy scissors . All mice were maintained in accordance with institutional and National Institutes of Health ( NIH ) guidelines approved by the Institutional Animal Care and Use Committees ( IACUC ) at Harvard Medical School ( Protocol 03611 ) and University of Wisconsin ( Protocol M00449-0-12-12 ) . E16 . 5 embryo heads were fixed in 4% paraformaldehyde ( PFA ) in PBS overnight and rinsed in PBS . The cochlea was exposed so that basal and apical turns were visible . In some cases , a small crystal of DiI ( Life Technologies ) was placed in the base of the cochlea , while a crystal of DiD ( Life Technologies ) was placed in the apex . In other cases , a picospritzer was used to inject a small amount of DiI or DiD dissolved in DMSO into the base or apex of the cochlea , respectively . Tissue was incubated at 37°C in PBS for 3–4 days to allow the dye to diffuse along axons . The hindbrain was then dissected out , cleared in ScaleA2 [66] at 37°C for 1 hour , mounted on a slide , and imaged by confocal microscopy to obtain z-projection images . To determine caudal/rostral bias of projections , the bifurcation zone was demarcated with a 100 pixel ( px ) diameter circle , and the intensity of caudal and rostral projections was measured by defining 100 px diameter circles adjacent to this zone . The ratio of caudal to rostral projections was calculated for each image and averaged for controls ( n = 2 wild-type+n = 2 heterozygote embryos ) and Npr2 mutants ( n = 4 embryos ) . Student's t-test was used to assess statistical significance . For P14–P18 animals , mice were perfused transcardially with 4% PFA in PBS ( n = 2 control and 4 Npr2 mutants ) . Their heads were bisected sagittally and fixed overnight in 4% PFA in PBS at 4°C . Tissue was rinsed in PBS and dissected so that the cochlea was exposed , with the brain still attached , then decalcified in 0 . 1 M EDTA in PBS at room temperature for 3 days . The decalcified bone covering the organ of Corti was removed so that mid and apical turns were visible , and small crystals of DiI and DiD were placed in the apical and mid-turns of the cochlea , respectively , using a 30-gauge needle to first create a small slit into which the dye crystal could be inserted . The tissue was incubated at 37°C in 4% PFA in PBS for 1 week , at which point most of the dye had diffused along projections . Since the axons are heavily myelinated at this stage and require a large amount of dye to reach the central projections in the cochlear nucleus , an additional crystal of DiI or DiD was at this time placed in the same slit , and allowed to diffuse for another week . The cochlea and cochlear nuclei were then dissected out , embedded in 5% low melt agarose in PBS , and sectioned by vibratome at 150 µm . For the cochlea , transverse sections of the cochlear nerve were collected , and for the cochlear nucleus , transverse sections of the aVCN and pVCN were collected . These were mounted on a slide and imaged by confocal microscopy ( Leica SP8 X ) . To examine the overall pattern of peripheral projections in the cochlea , wild-type ( n = 2 ) or Npr2 mutant ( n = 2 ) animals were perfused transcardially with 4% PFA in PBS , and the cochleae were dissected out and fixed overnight at 4°C , then subjected to whole-mount immunofluorescence with chick anti-Neurofilament antibody ( 1∶1000 , Abcam ) , using Alexa488-conjugated goat anti-chick secondary antibody ( 1∶1000 , Life Technologies ) . Cochleae were mounted in Vectashield ( Vector Labs ) and imaged on a Leica SP8 X confocal microscope . To visualize individual SGN peripheral processes in the cochlea , Npr2cn/+ mice were crossed with Npr2 heterozygotes also carrying the Neurog1-creERT2 and Ai14: tdTomato alleles . Since leaky Cre expression results in random , sparse recombination of the Ai14: tdTomato allele even without tamoxifen administration , this allowed us to label relatively few SGNs with the red fluorescent protein tdTomato . Cochleae from perfusion-fixed P14 animals ( n = 3 heterozygous control , n = 5 Npr2 mutant ) were collected and further fixed overnight in 4% PFA in PBS , then mounted on a slide in Vectashield for confocal microscopy ( Leica SP8 X ) . To label SGN central axons at E16 . 5 , embryo heads were fixed in 4% PFA in PBS overnight and the cochlea was exposed . A picospritzer was used to inject DiI dissolved in DMSO into the cochlea , and then treated as described above for tonotopic dye labeling . To visualize the overall pattern of SGN projections in the cochlear nuclei at postnatal stages , Npr2cn/+ mice were mated with Npr2 heterozygotes also carrying the Neurog1-creERT2 and Ai14: tdTomato alleles . P14–P18 animals ( n = 5 heterozygous control , n = 6 Npr2 mutant ) were perfused transcardially with 4% PFA in PBS , and their brains were drop fixed overnight in 4% PFA in PBS . Cochlear nuclei were then dissected out and cleared overnight in ScaleA2 . The entire cochlear nucleus was mounted in ScaleA2 on a glass slide and imaged using a Leica SP8 X confocal microscope . Tiled confocal stacks ( ∼300 µm thick ) were obtained at 10× so that the entire cochlear nucleus was covered . These tiled images were stitched together by ImageJ and z-projected to generate a single , large image of the cochlear nucleus including aVCN , pVCN , and DCN . For examination of projections in just the aVCN and pVCN , animals were processed as above , and then cochlear nuclei were embedded in 5% low melt agarose in PBS and cut sagittally at 150 µm using a vibratome . Regular confocal stacks were obtained at 20× and 40× and z-projected . To assess the morphology of auditory nerve fibers and topographic organization of tuberculoventral cell projections , biocytin injections were made into the aVCN in parasagittal slices in mice aged between P14 and P26 . With a single , parasagittal cut , the cochlear nuclei were removed from the brainstem in a single “slice” of up to 400 µm either with a vibratome or with scissors . The slice was maintained in vitro as in electrophysiological experiments . With a picospritzer , normal saline containing 1% biocytin ( Sigma ) was injected into the aVCN through a pipette with a tip diameter of ∼5 µm . Movement of the pipette through the slice disrupted processes that crossed the injection site as pulses of pressure released biocytin . Biocytin was allowed to spread through the tissue for 1 . 5 to 2 hours as slices continued to be superfused with warmed , oxygenated saline . Slices were then fixed in 4% PFA , stored at 4°C , embedded in a gelatin-albumin mixture , and resectioned at 40 to 60 µm in frozen sections . Biocytin in cells and fibers was visualized with horseradish peroxidase ( Vectastain ABC Elite Kit , Vector Laboratories ) [12] . Photomicrographs were taken through a Zeiss Axioskop with a Zeiss Axiocam . After being processed histologically , sections were analyzed with a camera lucida . Each section was reconstructed and marked with the locations of labeled neurons and landmarks . Landmarks were used to reconstruct slices as illustrated in Figure 5D . The distribution of labeled cells in the reconstructed slice was measured by means of a transparent grid that was laid parallel to an isofrequency band . Cells were then counted within parallel rows of squares as illustrated by the histograms in Figure 5D . Comparisons between genotypes were made by lining up peaks in histograms and summing cells in bands . Half widths were statistically compared using a one-way ANOVA test with Origin ( v 7 . 5 ) software . P21 animals were perfused transcardially with 4% PFA in PBS , and bisected heads were fixed overnight at 4°C in fixative ( 2 . 5% PFA , 5% glutaraldehyde , 0 . 06% picric acid in 0 . 2 M sodium cacodylate buffer ) . The cochlear nuclei were dissected out with the eighth nerve attached , and the region where the eighth nerve enters the cochlear nuclei was cut into a 1–2 mm cube in the fixative . The tissue was washed in 0 . 2 M sodium cadocylate buffer three times , followed by incubation in 1% osmium tetroxide/1 . 5% potassium ferrocyanide in water for 1 hour in the dark at room temperature . After three washes in malelate buffer ( pH 5 . 15 ) , the tissue was placed in 1% Uranyl Acetate or maleate buffer for 30 minutes , washed in water three times , and then dehydrated through an ethanol series ( 70% ethanol for 15 min , 90% ethanol for 15 min , and 100% ethanol twice for 15 min ) . Tissue was incubated in propyleneoxide solution for 1 hour , and then infiltrated with Epon resin mixed 1∶1 with propyleneoxide for 2–3 hours at room temperature . Samples were embedded in freshly mixed Epon and polymerized for 24–48 hours at 60°C . Thin sections were cut transverse to the eighth nerve using a Reichert Ultracut-S and were imaged using a Technai G2 Spirit BioTWIN transmission electron microscope with an AMT 2k CCD camera . To calculate the g-ratio , EM images of the eighth nerve were obtained for control ( n = 3 ) and Npr2 mutant ( n = 3 ) mice , and Fiji ( ImageJ ) was used to demarcate the area encompassed by each axon , as well as the area of the entire myelinated fiber . The g-ratio for each axon was calculated for ∼200 axons for each animal by dividing the diameter of the entire myelinated fiber by the diameter of the axon proper , and averages for controls and mutants were calculated . Statistical significance was assessed using Student's t-test . Auditory brainstem responses ( ABRs ) were recorded in 6-week-old mice in a soundproof chamber , as previously described [67] . Average ABR waveforms were plotted using a MATLAB ( MathWorks ) script written by Ann E . Hickox in the laboratory of Dr . Charles Liberman ( Eaton Peabody Laboratories , Massachusetts Eye and Ear Infirmary , Boston , MA ) . Statistical significance was assessed using Student's t-test . Coronal slices of the cochlear nuclei were made from mice between P17 and P25 . Slices ( 220 µm thick ) were cut with a vibrating microtome ( Leica VT 1000S ) in normal physiological saline or in saline with reduced Na+ at 24–27°C , and then transferred to a recording chamber ( ∼0 . 6 ml ) and superfused continually at 5–6 ml/min . Temperature was controlled with a Thermalert thermometer ( Physitemp ) the input of which comes from a small thermistor ( IT-23 , Physitemp , diameter: 0 . 1 mm ) placed between the inflow of the chamber and the tissue . The output of the Thermalert thermometer was fed into a custom-made , feedback-controlled heater that heated the saline in glass tubing ( 1 . 5 mm ) just before it reached the chamber to maintain the temperature at 33°C . Biocytin injections were made under the control of a Wild ( M5 ) dissecting microscope . For electrophysiological recordings , the tissue was visualized through a compound microscope ( Zeiss Axioskop ) with a 63× water immersion objective and CCD Camera ( Hamamatsu ) , with the image displayed on a video screen . Whole-cell patch clamp recordings were made by using an Axopatch 200A amplifier ( Axon Instruments , Burlingame , CA ) . Patch electrodes whose resistances ranged between 3 . 5 and 8 MΩ were made from borosilicate glass . All recordings of eEPSCs were digitized at 40 kHz and low-pass filtered at 10 kHz . The series resistance was compensated by 85–90% in recordings from octopus cells and by 70–80% in recordings from T stellate and bushy cells with a 10-µsec lag [16] . EPSCs were evoked by shocks through a Master-8 stimulator and Iso-flex isolator ( AMPI , Jerusalem , Israel ) , delivered through an extracellular-saline-filled glass pipette ( ∼5 µm tip ) . Analysis of EPSCs was performed by using pClamp ( Clampfit 9 . 0 , Axon Instruments ) . For solutions , all chemicals were from Sigma-Aldrich , unless stated otherwise . Normal saline: The normal extracellular physiological saline comprised ( in mM ) 130 NaCl , 3 KCl , 1 . 2 KH2PO4 , 2 . 4 CaCl2 , 1 . 3 MgSO4 , 20 NaHCO3 , 6 HEPES , 10 glucose , and 0 . 4 ascorbic acid saturated with 95% O2-5% CO2 , pH 7 . 3–7 . 4 , between 24 and 33°C . The osmolality was 306 mOsm/kg ( 3D3 Osmometer , Advanced Instruments Inc , Norwood , MA ) . Cutting solution: Some dissections were performed in a special cutting solution that contained ( in mM ) 99 NaCl , 3 KCl , 1 . 2 KH2PO4 , 1 CaCl2 , 1 . 3 MgSO4 , 20 NaHCO3 , 6 HEPES , 10 glucose , and 72 sucrose . Pipette solution: Recording pipettes were filled with a solution that consisted of ( in mM ) 90 Cs2SO4 , 20 CsCl , 5 EGTA , 10 HEPES , 4 Mg-ATP , 0 . 3 GTP , 5 Na-phosphocreatine , 5 mM QX314 , and was adjusted to pH 7 . 3 with CsOH ( ∼298 mOsm ) . Voltages were corrected for a −10 mV junction potential . | Millions of people suffer from debilitating hearing defects , ranging from a complete inability to detect sound to more subtle changes in how sounds are encoded by the nervous system . Many forms of deafness are due to mutations in genes that impair the development or function of hair cells , which are responsible for changing sound into electrical signals that can be processed by the brain . Both mice and humans carrying these mutations fail standard hearing tests . In contrast , very little is known about the genetic basis of central auditory processing disorders , which are poorly defined and difficult to diagnose , since these patients can still detect sounds . By finding genes that are required for the normal wiring of central auditory circuits in mice , we can investigate how changes at the circuit level affect circuit function and therefore improve our understanding of central auditory processing disorders . Here , we show that the natriuretic peptide receptor Npr2 is required to establish frequency maps in the mouse central auditory system . Surprisingly , despite a dramatic change in circuit organization , Npr2 mutant mice are still able to respond to sounds with normal sensitivity and timing , underscoring the need for better hearing diagnostic methods in mice as in humans . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"auditory",
"system",
"developmental",
"neuroscience",
"biology",
"and",
"life",
"sciences",
"sensory",
"systems",
"neuroscience",
"neural",
"circuit",
"formation"
] | 2014 | Mutation of Npr2 Leads to Blurred Tonotopic Organization of Central Auditory Circuits in Mice |
A first line of defense against pathogen attack for both plants and animals involves the detection of microbe-associated molecular patterns ( MAMPs ) , followed by the induction of a complex immune response . Plants , like animals , encode several receptors that recognize different MAMPs . While these receptors are thought to function largely redundantly , the physiological responses to different MAMPs can differ in detail . Responses to MAMP exposure evolve quantitatively in natural populations of Arabidopsis thaliana , perhaps in response to environment specific differences in microbial threat . Here , we sought to determine the extent to which the detection of two canonical MAMPs were evolving redundantly or distinctly within natural populations . Our results reveal negligible correlation in plant growth responses between the bacterial MAMPs EF-Tu and flagellin . Further investigation of the genetic bases of differences in seedling growth inhibition and validation of 11 candidate genes reveal substantial differences in the genetic loci that underlie variation in response to these two MAMPs . Our results indicate that natural variation in MAMP recognition is largely MAMP-specific , indicating an ability to differentially tailor responses to EF-Tu and flagellin in A . thaliana populations .
Pathogens pose a constant threat to their hosts . While lacking the adaptive immune system present in mammals , plants have evolved a two-tiered immune system of considerable specificity . The first tier of defense involves the recognition of microbe associated molecular patterns ( MAMPs ) that are common to many microbes . Plants recognize MAMPs , such as the elongation factor Tu ( EF-Tu ) and flagellin , or their epitopes elf18 and flg22 , by specialized receptors that allow the plant to discriminate self versus non-self and induce signaling cascades that result in defense responses [1 , 2] . The second tier of the plant immune system involves the recognition of specific microbial strains ( in contrast to semi-universal microbial patterns ) via the activity of resistance proteins . The specificity of the immune system in distinguishing specific microbes was previously thought to lie primarily in the activity of resistance proteins . Recent studies have challenged this paradigm in demonstrating both qualitative and quantitative variation in MAMP perception [3] . Plant hosts have the capacity to recognize multiple MAMPs , which helps ensure that at least one of the many MAMPs is recognized . For example , the order Brassicales evolved the capacity to recognize the MAMP elf18 in addition to ancient flg22 , that is detected by all land plants [4–6] . The perception systems for elf18 and flg22 share common molecular components , such as the co-receptor BAK1 [7 , 8] , and elicit similar changes in gene expression [2 , 9] suggesting that MAMP-triggered signal transduction and the associated signaling cascade converge quickly , regardless of the MAMP trigger . If MAMP-perception systems were completely redundant one would expect identical physiological responses to distinct MAMP triggers . However , some physiological responses differ in detail . For example , elf18 and flg22 induce different Ca2+ signaling and macroscopic growth responses [10] . In particular , flg22 perception induces an equally strong growth reduction in roots and shoots , whereas elf18 perception acts more strongly in leaves than in roots [10 , 11] . Additional evidence hints at a more complex interplay of molecular components after induction with different MAMPs [12–15] . The differentiated responses to separate MAMP classes suggest that the benefit of recognizing multiple classes may extend beyond redundancy . For example , it is possible that plants tailor defense response intensities to different MAMPs , an adaptation that could be especially beneficial when MAMP composition reflects differences in microbial community composition . Our previous work revealed extensive natural variation in the detection of a variant of flg22 and signatures of selection at the genetic loci underlying this variation [3] . This quantitative variation in MAMP perception indicated that these traits are evolving within natural populations . Whether these traits–the perception and response to divergent MAMPs–act and evolve redundantly is an open question . To determine whether detection of flagellin and EF-Tu evolve redundantly in natural populations , we tested a large panel of A . thaliana genotypes for their response to different variants of these two MAMPs . As a read-out for MAMP perception , we utilized the macroscopic phenotype of seedling growth inhibition ( SGI ) , a phenotypic response shown to correlate closely with MAMP perception in several Brassicaceae species [3] . In assaying quantitative variation in the responses to different MAMP variants and different MAMP classes , and determining the genetic loci underlying this variation via genome-wide association mapping , we shed light on the evolution of this first tier of defense .
In order to characterize natural variation in MAMP-induced seedling growth inhibition ( SGI ) in A . thaliana , we challenged 186 natural genotypes with several peptide variants for each of two classes of MAMPs , elf18 and flg22 ( Fig 1 ) . An important feature of the study system is that growth inhibition is triggered by the peptides alone , without any confounding effects of bacterial proliferation and disease . We observed extensive variation in MAMP-induced SGI: Some genotypes exhibited no response while others exhibited a mean reduction in growth of up to 86% ( e . g . , Dra3-1 in response to elf18Pv ) . Different MAMP peptides induced significantly different SGI ( ANOVA F5 , 5828 = 854 , p = < 2e − 16 , see S1 Table for full ANOVA table ) , consistent with previous findings [16–18] . In particular , the highly diverged peptide flg22Pv ( Fig 1b ) induced little to no SGI in most genotypes ( Fig 1a ) . Although this flg22 variant of P . viridiflava induced no response in most A . thaliana genotypes , its elf18 variant induced the strongest SGI of all elf18 variants . In this case , the independent response to P . viridiflava enables the continued detection of pathogenic invader even though recognition of one MAMP fails . Two of the 186 A . thaliana genotypes did not exhibit elf18-induced SGI and 10 did not exhibit flg22-induced SGI ( Fig 2b ) . Genotypes lacking a response to one MAMP class were not impaired in their response to the other class , suggesting that loss-of-function of the elf18 and flg22 perception systems undergo differentiated evolution . In order to investigate the cause of low flg22-induced SGI , we tested FLS2 protein levels of nine genotypes that showed little or no SGI in response to any flg22 peptide variant . We observed compromised protein production in all tested genotypes ( Fig 2d ) . It remains unclear if lower protein levels are caused by mutations within the protein coding sequence of the receptor or trans-acting factors . Analysis of mRNA expression levels in a subset of the flg22-insensitive genotypes revealed lower levels of mRNA expression of FLS2 ( S1 Fig ) in these genotypes . We furthermore obtained FLS2 nucleotide sequences for 9 of the 10 flg22-insensitive genotypes . Six out of nine show putative deletions in the catalytic site of the serine-threonine kinase domain of FLS2 ( S1 Text ) . These results suggest that the absence of SGI in response to flg22 treatment is largely explained by differences in the coding sequence and the abundance of the receptor itself . The comparatively high prevalence of flg22-insensitive genotypes raise the possibility that the elf18 perception system has surpassed the ancient flg22 pathway [4] in importance , at least in A . thaliana . Finally , it is interesting to note that small differences in amino acid sequence of the inducing peptides had the potential to profoundly impact SGI . For example , three amino acid differences between elf18Ps and elf18Pv resulted in an 18% increase in mean SGI ( Welch’s t-test t = −7 . 0 , p = 1 . 24e−11 ) . On the other hand , flg22PsHR− and flg22PsHR+ , which differ by a single amino acid , did not cause statistically significant differences in SGI ( Welch’s t-test t = −0 . 89 , p = 0 . 37 , Fig 1a ) . We asked if genotypes exhibit similar SGI in response to variants of flg22 and elf18 . If these MAMPs act redundantly , one would expect a strong correlation among the induced physiological responses . We observed strong correlations among variants within each MAMP class: the average Pearson correlation coefficient within a MAMP class is 0 . 61 whereas the mean correlation of SGI between MAMP classes is much weaker ( mean R = 0 . 24 , see Fig 2a and S2 Table for significance and confidence intervals of individual comparisons ) . Veluchamy et . al [19] observed a similar disparity between within-class and between-class correlations in a small set of heirloom tomato tested for reactive oxygen production upon treatment with three MAMPs , flg22Pa , flgII-28 and csp22 . Overall , the weak correlation between classes of MAMP peptides suggests some autonomy in the underlying genetic bases . We performed genome-wide association mapping to a ) reveal the genetic basis underlying natural variation in SGI and b ) identify loci that are uniquely important in shaping either flg22- or elf18-induced SGI ( Fig 3 ) . We first calculated marker-based heritability estimates ( h2 = 0 . 29 on average for all perceived peptides , S3 Table ) to test if SGI is amenable to GWA mapping [20] . The MAMP flg22Pv induced little to no SGI in most A . thaliana genotypes , which is reflected in a broad-sense heritability estimate of 0 ( S3 Table ) . We tested the significance of associations between the seven phenotypes and approximately 210 , 000 genome-wide SNP markers [21] in linear mixed models ( EMMAX ) that takes population structure into account [22] . We identified a number of loci that exhibit significant association with natural variation in SGI ( Fig 3 ) . In general , SGI appears to be governed by a complex genetic architecture involving loci with small effect . Only the GWA mapping of elf18DC-induced SGI identified a locus of large putative effect; this locus , which was found on chromosome 5 , explains 28% of the observed variation . The MAMP receptors EFR and FLS2 were considered strong a priori candidate genes for variation in SGI . The top SNPs of the single prominent peak for elf18DC-induced SGI , indeed , co-localize with the EFR locus . Surprisingly , EFR is not strongly associated with variation in SGI induced by the other two elf18 variants , elf18Ps and elf18Pv . To investigate this further , we conducted phylogenetic analysis and revealed two major EFR haplotype groups ( Fig 4 ) that strongly differentiate detection of elf18DC from elf18Ps and elf18Pv ( Figs 3 and S2 ) . Thus , EFR controls natural variation in SGI for only a subset of elf18 variants . Variation in flg22-induced SGI was attributed to many loci with small effect . The most prominent peak ( p ≤ 4 * 10−8 ) in the flg22Pa-induced SGI co-localizes with the a priori candidate gene NADPH/respiratory burst oxidase protein D ( RbohD ) , that is known to fine tune reactive oxygen production and hypersensitive response around pathogen infection sites . None of the SNPs within the 30 kb window comprising the FLS2 gene are significantly associated with flg22-induced SGI ( see close-up of genetic region in S3 Fig . ) Similarly , no significantly associated SNP mapped near the FLS2 co-receptor BAK1 ( Fig 3 ) . The functional importance of BAK1 for flg22-induced immunity is well-established [7]; however , the unusually limited natural genetic variation at the BAK1 locus ( π = 0 . 0008 in 80 genotypes published in [23] versus 0 . 005 for a genome wide estimate by [24] ) does not appear to contribute to natural variation in the SGI . We identified a number of a priori candidate genes that are known to have an effect on SGI ( Table 1 ) . These candidates were found to be enriched in the 0 . 1% tail of elf18DC and elf18Ps but not the other MAMPs ( Table 2 ) . Thus , our results illustrate that genes with strong phenotypic effects in knock-out experiments do not necessarily harbor genetic variation causing natural variation in this plant phenotype [25] . We examined the genetic overlap in peaks identified during GWAS analysis of the responses to each MAMP . While MAMPs within each class share a higher number of peaks than expected by chance ( S4 Table and S4 Fig ) , we did not observe enrichment in the overlap of GWAS peaks for SGI induced by the MAMP classes elf18 and flg22 . This finding is consistent with our observation of high correlations of SGI within , but not between , MAMP classes , and suggests a common genetic architecture within each of two distinct MAMP classes . Our results raise the possibility that flg22 and elf18 induce molecular responses that are more differentiated than is suggested by their similar patterns of gene expression [2 , 9] . The observation that elf18 and flg22 induce different macroscopic changes further supports the differentiation of these responses . Specifically , flg22 acts both on leaf and root tissue , while elf18 is most effective on leaves [10 , 11] . We confirmed this observation in a quantitative experiment demonstrating that application of elf18 significantly alters the shoot/root ratio compared to untreated plants , whereas flg22 treatment does not ( ANOVA F2 , 175 = 77 . 0 , Tukey post hoc elf18Ps vs control p ≤ 2 . 2e−16 , flg22Pa vs control p = 0 . 36 , Fig 2c ) . To further test our hypothesis that different loci underlie natural variation in elf18 and flg22-induced SGI , we selected candidate genes from our GWA for experimental validation using both MAMP classes . For each MAMP , we selected the 0 . 1% tail of p-values ( i . e . , the 203 SNPs most strongly associated with MAMP-induces SGI ) . To account for linkage disequilibrium , we defined peak regions that comprise all SNPs within 15 kb to either side of the SNP with the lowest p-value . This resulted on average in 113 peak regions per MAMP . On average , six genes underlie a peak region , resulting in 4481 genes across all peak regions and six MAMPs ( flg22Pv excluded , see S5 Table for details ) . For experimental confirmation of candidates , we focused on genes that were either mapped in multiple MAMPs , genes with differential expression upon MAMP treatment [2 , 9] or with a known role in defense mechanism or growth . We retrieved one T-DNA insertion line for each of 88 genes from the ABRC seed stock center and successfully confirmed homozygosity of T-DNA insertion for 57 mutant lines that were tested for SGI . We identified 11 candidate genes that significantly alter SGI ( Table 3 ) , some of which are a priori candidates such as EFR and RbohD , while others are genes of unknown function ( e . g . , AT4G21865 and AT5G57345 ) . We consider as confirmed only candidates that pass q ≤ 0 . 05 after correction for false discovery rate . This approach reduces the reporting of false positives but might exclude some true positives ( e . g , RIN4 , involved in modulating flg22-induced SGI [26] . ) Our confirmation of 11 genes reveals a four fold enrichment in significant loci in comparison to the null . Our confirmation rate of 20% ( after correction for multiple testing ) is therefore substantially higher than expected by chance . Of the 11 confirmed candidates , four impact both elf18- and flg22-induced SGI and seven control either elf18- or flg22-induced SGI ( Table 3 ) . This finding suggests differences in the molecular pathways involved in the recognition of and response to different MAMP classes . The discrepancy between the extensive overlap previously identified in the pathways for elf18 and flg22 recognition , and the reduced overlap in genes identified here , may be explicable by either of two hypotheses . One possibility is that loci underlying natural variation in SGI are primarily involved in the initial steps of MAMP perception prior to signal convergence . Functional differences in the receptor alone might be caused by sequence variation in its protein coding region [3] , its transcriptional regulation [15] , its post-translational glycolysation [27 , 28] or modification of obligatory co-receptors [31] . Another possibility is that yet unidentified pathways , unique to perception of one MAMP , lead to observed phenotypic differences . We found some support for this hypothesis by identifying a number of experimentally confirmed loci that alter SGI uniquely upon treatment with elf18 variants but not flg22 variants . Overall , our results reveal a lack of correlation between a plant’s response to elf18 and flg22 . What are the implications of this uncorrelated response ? Natural populations experience fluctuations in both abiotic and biotic conditions , and plants in different environments are colonized by disparate microbial communities and pathogen species . In light of this variability , the decoupling of elf18- and flg22-triggered physiological responses is potentially advantageous . Genotypes that respond more strongly to one MAMP do not in general respond equally strongly to another MAMP i . e . , there are no genotypes hypersensitive to all MAMPs . This uncoupling of the responses allows for responses to different MAMPs to evolve independently in populations . There is also the recent demonstration of epistatic effects of recognizing multiple MAMPs , at least in mammalian systems [32] . The ways in which hosts regulate their microbiomes is a complex issue that is only beginning to be understood , but distinctly tailored MAMP initiated defense is likely one contributing factor .
Plants perceive specific epitopes of EF-Tu and flagellin that are known as elf18 and flg22 , respectively . Flg22 sequences of five natural P . viridiflava strains [33] were identified using primers 5’-GCCATCGCGACGATAACTA-3’ and 5’-GGCGTTTTCGTTGATGTTCT-3’ . Flg22 sequences for P . viridiflava strains LP23 . 1a and RMX3 . 1b as well as both elf18 and flg22 sequences for 20 P . syringae strains isolated from A . thaliana and the surrounding plant Drava verna [34] were derived from genome sequences available in the Bergelson laboratory ( S2–S3 Texts ) . The Pseudomonas strains from which these MAMPs were derived were previously isolated from natural populations of A . thaliana . Infection analyses with the P . syringae strains determined that a subset of strains did not induce HR on A . thaliana [34 , 35] . We tested MAMP variants derived from strains that did not induce HR ( HR- ) and strains that successfully induce HR ( HR+ ) . The identified elf18 and flg22 variants were synthesized by EZBiolab , Carmel , IN . Seedlings growth inhibition ( SGI ) was estimated for 186 genotypes of A . thaliana that were part of the panel in Atwell et al . [21] . Plants grown in the absence ( control ) or presence of 100nM MAMP ( treatment ) were grown in pairs and cultivated in sterile conditions as described in Vetter et al . [3] . Seedling growth inhibition was calculated as relative reduction of fresh mass in percent by [ ( CFM—TFM ) / CFM] * 100 , where CFM stands for control fresh mass and TFM for treatment fresh mass . Phenotypic values of SGI were calculated by averaging at least three pairs of control and MAMP treatments per genotype; these pairs were obtained in five independent biological trials for elf18 and seven independent biological trials for flg22 ( resulting in a maximum number of 15 or 21 biologically independent values of SGI per genotype ) . All data analysis was conducted in R [36] . Particular packages are indicated where appropriate . Plant material was grown and processed as described in Vetter et al . [3] . In short , nitrogen frozen plant material was homogenized , weighed and dissolved in equal amounts of extraction buffer ( 25mM MES pH 6 , 3mM MgCl2 , 10mM NaCl and Sigma protease inhibitor , 2% SDS ) . Proteins were separated in a Novex NuPage 3–8% Tris-acetate gel and blotted onto a PVDF membrane ( Immobilon , Millipore ) . The α-FLS2 antibody was incubated in 1:5 , 000 dilution , overnight at 4° . A horseradish peroxidase , coupled to secondary anti-rabbit IgG in 1:2 , 000 dilution was used to detect FLS2 protein bound to the membrane . This antibody was previously demonstrated to be specific to FLS2 [37] . Expression data for the FLS2 locus were obtained from [38] and the correlation between SGI and mRNA expression level were determined for those genotypes for which we had both SGI and RNA-sequencing information ( 49 accessions ) . The FLS2nucleotide sequence data for the genotypes that failed to respond to flg22 was obtained from http://tools . 1001genomes . org/pseudogenomes/ . We investigated SGI induced by three elf18 and four flg22 peptide variants in 186 genotypes of A . thaliana . Twelve genotypes exhibited a mean SGI < 15% in response to either elf18 or flg22 variants . We considered these genotypes natural SGI mutants and excluded them prior to correlation analysis . We calculated Pearson correlation coefficients and confidence intervals for the remaining 174 genotypes . Note that no genotype failed to recognize both MAMPs . Marker-based heritability and confidence intervals were calculated using the marker_h2 function of the package ‘heritability’ in R [39] . This model incorporates a genetic relatedness matrix and generates REML-estimates of the additive genetic variance ( σ A 2 ) , residual variance ( σ E 2 ) and their standard errors , which allows calculation of heritability according to the equation , h 2 = σ A 2 / ( σ A 2 + σ E 2 ) . The relatedness matrix was generated using plink1 . 07 [40 , 41] and SNP data by Kim et al . [42] . Genome-wide association mapping was conducted on average SGI for each genotype for which high density genotype data was available [42] . Analysis and restructuring of genotype data was conducted in plink1 . 07 [40 , 41] . The mixed linear model EMMAX takes population structure into account by incorporating a K matrix of genetic relatedness [22] . EMMAX positively biases association signals for alleles with low frequency in the population . We therefore disregarded SNPs with a minor allele frequency ( MAF ) smaller than 5% . Our mapping panel contained 214 , 051 discriminative SNPs , of which 203 , 498 passed the MAF filter . In order to assess genetic similarities within and between MAMP classes , we identified shared peak regions . To account for linkage disequilibrium [42] and the fact that the highest associated SNP might not be the causal one [43] , we defined peak regions encompassin 15 kb on either side of the highest associated SNP . A literature search identified 77 a priori candidate genes with published evidence of their participation in the SGI response to either elf18 or flg22 . We tested for enrichment of these a priori candidates among our identified peak regions by calculating their frequency among the genes underlying the mapped peak regions of each trait . In order to assign an empirical p-value , we kept the number of peaks and their genetic distance intact but slid them across the genetic position of each chromosome . We then counted the number of genes underlying these shifted peaks and calculated the frequency of a priori candidates within the shifted peak regions . We repeated this procedure 100 times and determined the number of times we achieved an equal or greater frequency of a priori candidate genes . All custom scripts are available at https://bitbucket . org/mvetter/geneticbasissgi/ . We considered SNPs that fell in the 0 . 1% tail of ranked GWA p-values; these correspond to the 203 most strongly associated SNPs . To account for linkage disequilibrium , we defined peak regions 15 kb to either side of the highest associated SNP and identified genes co-localizing with these peak regions based on TAIR 9 annotation . From this list of 4481 genes , we selected candidate genes for experimental confirmation if they fulfilled at least one of the following criteria: ( 1 ) detected by multiple MAMP variants , ( 2 ) differentially expressed upon MAMP perception ( using data of [2 , 9] ) , or ( 3 ) a priori candidate for growth or defense related processes . We selected 88 loci for which T-DNA insertion mutants were ordered from ABRC seed stock center ( list available in the data folder of the repository bitbucket . org/mvetter/geneticbasissgi . ) We analyzed a single T-DNA insertion line for each candidate gene ( but see [44] ) and could confirm homozygosity of T-DNA insertion for 57 mutant lines using primers according to http://signal . salk . edu/tdnaprimers . 2 . html . Silke Robatzek , The Sainsbury Laboratory , Norwich , kindly provided fls2-24 and efr-0 mutants for control purposes . We grew each SALK mutant genotype in a minimum of 12 replicates with an equal number of wild-type ( WT ) Col-0 or ( WT ) Ler plants ( according to respective mutant background ) . Significant differences in SGI between mutants and WT were determined using a non-parametric Wilcoxon rank-sum test in combination with multiple testing correction using the fdrtool package [45] . To identify MAMP-specific SGI , we first tested the MAMP class and variant that led in the GWA to the candidate gene selection . If a significant result was observed , we tested at least one more variant within the same class as well as a variant in the other class . | Specialized receptors encoded by plants detect different components of bacterial machinery , and initiate an immune response . These recognition events are thought to induce largely redundant defense signaling , the magnitude of which varies quantitatively among populations , perhaps in response to environment specific differences in microbial threat . Here , we sought to determine whether plants evolve distinct or shared responses to two canonical MAMPs within natural populations . We comprehensively tested the extent of functional redundancy in the response of 186 genotypes of Arabidopsis thaliana to variants of each of two classes of bacterial signals , flagellin and EF-Tu . Although plants respond similarly to recognition of different variants of the same MAMP , we found the response to one MAMP class to be largely uncorrelated with the response to the other class . We further investigated the genetic bases underlying growth changes to determine whether similar genes contribute to variation in the response to EF-Tu and flagellin bacterial signals . We find limited genetic similarity , revealing novel MAMP-specific signaling components . The differentiation of these responses reveals MAMP-specific fine tuning of the immune response . | [
"Abstract",
"Introduction",
"Results",
"and",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"bacteriology",
"genome-wide",
"association",
"studies",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"variant",
"genotypes",
"brassica",
"microbiology",
"plant",
"physiology",
"genetic",
"mapping",
"plant",
"science",
"model",
"organisms",
"genome",
"analysis",
"plant",
"pathology",
"seedlings",
"plants",
"bacteria",
"bacterial",
"pathogens",
"research",
"and",
"analysis",
"methods",
"arabidopsis",
"thaliana",
"pseudomonas",
"pseudomonas",
"syringae",
"genomics",
"microbial",
"physiology",
"proteins",
"medical",
"microbiology",
"microbial",
"pathogens",
"genetic",
"loci",
"biochemistry",
"bacterial",
"physiology",
"plant",
"defenses",
"plant",
"and",
"algal",
"models",
"plant",
"disease",
"resistance",
"heredity",
"flagellin",
"genetics",
"biology",
"and",
"life",
"sciences",
"mamp-triggered",
"immunity",
"computational",
"biology",
"organisms",
"human",
"genetics"
] | 2016 | Differentiation between MAMP Triggered Defenses in Arabidopsis thaliana |
To assess the prevalence of HTLV-1 and HTLV-2 infections in a cohort of immigrants living in southern Italy . We screened for antibody to HTLV-1/2 infection 1 , 498 consecutive immigrants born in endemic areas ( sub-Saharan Africa or southern-Asia ) by a commercial chemiluminescent microparticle immunoassay . If confirmed in a Western blot assay , which differentiates anti-HTLV-1 from anti-HTLV-2 , the positive sera were tested for specific HTLV RNA by a home-made PCR . The immigrants investigated were more frequently males ( 89 . 05% ) , young ( median age 26 years ) , with a low level of education ( median schooling 6 years ) , born in sub-Saharan Africa ( 79 . 70% ) . They had been living in Italy for a median period of 5 months . Only one ( 0 . 07% ) subject was anti-HTLV-1 -positive/HTLV-1 RNA-negative; he was an asymptomatic 27-year-old male from Nigeria with 6 years’ schooling who stated unsafe sexual habits and unsafe injection therapy . The data suggest screening for HTLV1 and HTLV-2 infections all blood donors to Italy from endemic countries at least on their first donation; however , a cost-effectiveness study is needed to clarify this topic .
Human T cell leukemia virus type 1 ( HTLV-1 ) or 2 ( HTLV-2 ) infection has a worldwide distribution , with an estimate of up to 15–20 million people affected [1] . The prevalence changes substantially according to the geographical area , and is higher in specific risk groups such as intravenous drug users and sex workers [2 , 3] . Endemic areas for HTLV-1 infection have been reported in Japan , Melanesia , Iran , Central and West Africa , the Caribbean and South America [2–4] . In Europe , North America and Australia , HTLV-1 infection is rare and mainly found in immigrants from endemic areas , and in their sexual partners [2 , 3] . However , the HTLV-1 prevalence has been poorly investigated in several areas of sub-Saharan Africa and in most parts of Asia [4 , 5] . HTLV-2 has a more restricted distribution than HTLV-1 and occurs primarily in the Americas and among pygmy tribes in Africa; Amerindians residing in North , Central , and South America show various rates of positivity for HTLV-2 ( 5 to 30% ) [6 , 7] . Once acquired , HTLV-1 infections persist life-long , most patients remaining asymptomatic viral reservoirs ensuring the transmission chain , but about 4% develop adult T-cell leukemia/lymphoma ( ATLL ) , a highly aggressive CD4+ T-cell malignancy . Type 1 virus has been associated with other diseases , ranging from a mild non-specific dermatitis and uveitis to a disabling HTLV-1-associated myelopathy/tropical spastic paraparesis ( HAM/TSP ) affecting 2–3% of the infected people [8] . HTLV-2 has been associated to hairy cell leukemia , erythrodermatitis and a growing number of neurologic disorders [9] . HTLV transmission is similar to that of human immunodeficiency virus ( HIV ) , hepatitis C virus ( HCV ) and hepatitis B virus ( HBV ) , but is less effective . The most important routes of HTLV-1/2 transmission are vertical transmission , unsafe sexual habits , transfusion of blood or blood products and needle sharing [10] . In particular , several factors have been associated with the sexual transmission of HTLV-1 infection , namely a presence of genital sores or ulcers , unprotected sexual intercourse , having multiple sexual partners , and a lifetime contact with an HTLV-1-infected partner [4 , 5] . Due to socio-economic and political crises in several countries in Africa , Eastern Europe and Central and Eastern Asia in recent decades , Italy has become a land of immigration from areas with an intermediate or high HTLV endemicity . In literature there are few data about the seroprevalence of HTLV-1 and HTLV-2 infection in immigrants from endemic areas for this infection [11 , 12]; moreover , HTLV screening is currently not performed in blood donors in many European countries ( Italy , Spain , Germany ) . Because of the poverty of data , we investigated for the presence of HTLV-1 and HTLV-2 infection a cohort of 1 , 498 immigrants from endemic areas ( sub-Saharan Africa and Southern Asia ) living in southern Italy ( Naples , Caserta or Foggia ) and consecutively observed at one of the five first-level clinical centers from January 2012 to July 2017 .
All procedures used were in accordance with the international guidelines and with the Helsinki Declaration of 1975 and revised in 1983 . The Ethics Committee of the Azienda Ospedaliera Universitaria of the University of Campania “Luigi Vanvitelli” approved the study ( 214/2012 ) . All patients signed an informed consent for the collection and storage of biological samples and for the anonymous use of their data for research purposes . Estimating a prevalence about 3% [2–4] for HTLV infection with an accuracy estimate of 1% , the sample size had to be at least 1 , 118 subjects . Thus , to achieve the aim of the study , all 1 , 498 consecutive immigrants from HTLV-endemic areas ( sub-Saharan Africa and Southern Asia ) , seeking care at one of the five first-level clinical centers between January 2012 and July 2017 , were enrolled . The first-level clinical centers involved in the present study are clinical centers of international charity organizations empowered by the Italian National Healthcare System , with proven experience in the clinical , psychological and legal management of vulnerable groups , as previously specified [13–15] Each first-level clinical center is an out-patients clinic of general medicine . They are located in Naples , Caserta and Foggia , an area that gives hospitality to a large population of refugees and undocumented immigrants from Africa , Central and Eastern Asia and Eastern Europe . The study population consists of all the undocumented or refugee immigrants from sub-Saharan Africa and the India-Pakistan sub-continent consecutively seen for a clinical consultation . During the consultation , a physician from the clinical center and a cultural mediator explained to the immigrants the importance of testing for HBV , HCV , HIV and HTLV serum markers and offered them screening free of charge , in anonymity and in full accordance with the law on privacy . Informed consent to participate in a surveillance and monitoring program covering different viral infections , including HTLV-1 and HTLV-2 , was obtained on a voluntary basis . Of the interviewed immigrants 94 . 7% agreed to participate and were included in the present study . Patients who did not consent to participation in the study received the same treatment if they had any medical problem . The participants were interviewed and relevant information was collected , including socio-demographic data and risk factors . More precisely , we recorded age , sex , geographical origin , time of immigration , level of education , religion , cohabitation details , sexual habits , a history of previous vaccination , surgery , dental care , tattooing , piercing , drug addiction , blood transfusion , tribal rituals ( body scarification and/or infibulation ) and abortion . In each case the clinical history was obtained with the help of a physician and a cultural mediator in the course of a prolonged , in-depth clinical consultation and counseling . The serum samples obtained were transported to the laboratory of infectious diseases of the University of Campania Luigi Vanvitelli—Naples , where they were tested for HBsAg , anti-HBc , anti-HCV , anti-HIV and anti- HTLV-I/II . All undocumented immigrants and refugees received the results of their serological screening and full instructions on the prevention and transmission of the infections . Serum samples were tested for HTLV by a commercial chemiluminescent microparticle immunoassay ( Architect HTLV-I/II assay; Abbott , Wiesbaden-Delkenheim , Germany ) and those positive by Western blot to confirm HTLV-1 or -2 infections ( MP diagnostics HTLV- I/II blot 2 . 4 ) . The Q6138 region was amplified by polymerase chain reaction ( PCR ) from whole blood sample [16] to confirm infection by the presence of proviral HTLV DNA in individuals who were positive for the Western blot test . Moreover , serum samples were tested for HBsAg , anti-HCV , anti-HIV and total anti-HBc by commercial immunoenzymatic assays ( Abbott Laboratories , North Chicago , IL , USA: AxSYM HBsAg ( V . 2 ) M/S for HBsAg , AXSYM HCV 3 . 0 for anti-HCV , AXSYM HIV 1/2 COMBO for HIV , AXSYM core for total anti-HBc ) . Anti-HIV reactivity was always confirmed by a Western blot assay ( Genelabs Diagnostics , Science Park Drive , Singapore ) , which identifies both HIV-1 and HIV-2 strains .
The demographic and serological data of the 1 , 498 immigrants obtained at the time of enrolment are shown in ( Table 1 . ) These subjects , prevalently young males and with a low level of education , had been living in Italy for a median period of 5 months; 48 . 26% of them were undocumented immigrants , 45 . 86% low income refugees and 5 . 87% did not state their legal status ( Table 1 ) . Most of them ( 79 . 7% ) came from sub-Saharan Africa and 20 . 3% from southern Asia . More detailed information on the country of origin is shown in ( Table 2 ) . ( Table 1 ) shows the risk factors for parenteral and/or sexual acquisition of viral infections . Nearly 33% of immigrants stated invasive medical procedures such as surgery and dental care , 65% unsafe injective therapy , 8% tattooing and/or piercing , 57% tribal practices and nearly 47% unsafe sexual habits . Drug addiction was infrequently stated . Of the 1 , 498 immigrants enrolled , only one ( 0 . 07% ) was anti-HTLV-1 positive , 153 ( 10 . 2% ) HBsAg-positive , 605 ( 40 . 39% ) HBsAg-negative/anti-HBc-positive , 74 ( 4 . 9% ) anti-HCV-positive , 24 ( 1 . 6% ) anti-HIV-positive and 13 ( 0 . 87% ) had a multiple infection ( 4 HBsAg/anti-HIV-positive , 2 anti-HCV/anti-HIV-positive and 7 HBsAg/anti-HCV-positive ) ( Table 1 ) . No patient was anti-HTLV-2 positive . The immigrant found to be anti-HTLV-1-positive was a 27-year-old Nigerian Christian male with a low level of education ( 6 years ) who had been living in Italy for 3 months and was asymptomatic and HTLV-1-RNA-negative . This subject was HBsAg , anti-HBc , anti-HCV and anti-HIV-1/2-negative and stated unsafe sexual habits and unsafe injection therapy .
In the present study , enrolling 1 , 498 African and Asian immigrants living in Italy , the prevalence of positive immunoezymatic tests for HTLV-1 , confirmed by Western-blot test , was 0 . 07% . The only HTLV-1 subject infected was asymptomatic and HTLV-1-RNA-negative , in accordance with the literature data reporting a low prevalence of HTLV-1-RNA-positive patients among the anti-HTLV-1 positive [5 , 6] . In literature there are few data on the prevalence of HTLV-1/2 in immigrants [17] and , to our best knowledge , this is the first prospective study performed in Europe on HTLV infection in a population of undocumented and refugee immigrants from sub-Saharan Africa and the India-Pakistan subcontinent . In this study 1 , 194 , subjects from several states of sub-Saharan Africa and 304 from the India-Pakistan area ( India , Pakistan , Bangladesh , Sri-Lanka and Afghanistan ) were screened for HTLV 1–2 antibodies . The subjects were enrolled at one of the five first-level medical centers with years of experience dealing with immigrants . Thus , the present study is , to our best knowledge , the first prospective study performed in Italy on an undocumented and refugee immigrant population providing valuable information on HTLV infections in relation to the demographics and geographical areas of origin . The subjects enrolled were young ( median age 26 years ) , prevalently males and asymptomatic , and were representative of the immigrant population from developing countries living in Italy and Europe [13 , 14 , 18] . The data on the prevalence of HTLV 1–2 infection in this population are useful to devise proper healthcare strategies of screening for immigrant populations from different countries and to evaluate the need for screening of blood donors , today not indicated for this infection in Italy . In fact , in high-income countries , considering the low prevalence of HTLV-1 infection and the low rate of disease onset ( below 10% ) , cost-effectiveness of universal HTLV-1 screening of blood donors is debated . However , the identification of HTLV-1 carriers is nevertheless important to avoid transmission through blood transfusion . The World Health Organization global database on blood safety does not include HTLV testing as one of its data collected from some countries around the world . The following countries test all blood donations for HTLV-1 and -2 antibody: Argentina , Australia , Brazil , USA , Canada , China ( some regions ) ( 13 ) , Colombia , French West Indies , Iran , Israel , Jamaica , Japan , New Zealand , Saudi Arabia , Peru , Sweden , Taiwan , Uruguay and Venezuela . In Europe , HTLV-1/2 antibody testing is currently performed on all blood donations in France , Greece , Ireland , Netherlands , Portugal , Romania , United Kingdom and on first time blood donations in Denmark , Finland , Norway and Sweden [19] . Another important question is the high prevalence of HBV and HCV infection among immigrants living in Western countries [13 , 14 , 20 , 21] , but fortunately in Italy screening for both of these viruses is planned at the time of blood donation . Worthy of mention , in Italy the immigrant population constitutes one tenth of the whole population and its contribution to blood donation is strongly desired . Although HTLV-1 and HTLV-2 infections were rarely detected in the immigrant population investigated in this study , we believe that every effort must be made to make all blood donations as safe as possible . It may be advisable for the Italian Healthcare Authorities to start screening for HTLV-1 and HTLV-2 infections all donors from endemic countries at least on their first donation , a practice useful for a correct cost-effectiveness analysis and for a conclusive decision on this topic . In fact , even if the prevalence of HTLV-1 and HTLV-2 is low , it is not negligible and , considering the transmissibility through blood and sexual route , could represent a significant problem in Western countries both in terms economic and public health . However , a cost-effectiveness study is needed to clarify this point . | As in Italy the immigrants constitute one tenth of the whole population and its contribution to blood donation is strongly desired we believe that to start screening for HTLV-1 and HTLV-2 infections in all donors from endemic countries at least on their first donation is a practice useful for a correct cost-effectiveness analysis and for a conclusive decision on this topic . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"body",
"fluids",
"pathology",
"and",
"laboratory",
"medicine",
"european",
"union",
"cost-effectiveness",
"analysis",
"economic",
"analysis",
"pathogens",
"blood",
"counts",
"geographical",
"locations",
"microbiology",
"social",
"sciences",
"habits",
"retroviruses",
"viruses",
"rna",
"viruses",
"africa",
"medical",
"microbiology",
"htlv-1",
"behavior",
"microbial",
"pathogens",
"economics",
"italy",
"people",
"and",
"places",
"blood",
"anatomy",
"blood",
"donation",
"viral",
"pathogens",
"physiology",
"biology",
"and",
"life",
"sciences",
"europe",
"vascular",
"medicine",
"organisms"
] | 2018 | Low prevalence of HTLV1/2 infection in a population of immigrants living in southern Italy |
Dengue has propagated widely through the Americas . Most countries have not been able to maintain permanent larval mosquito control programs , and the long-term effects of control actions have rarely been documented . The study design was based on a before-and-after citywide assessment of Aedes aegypti larval indices and the reported incidence of dengue in Clorinda , northeastern Argentina , over 2003–2007 . Interventions were mainly based on focal treatment with larvicides of every mosquito developmental site every four months ( 14 cycles ) , combined with limited source reduction efforts and ultra-low-volume insecticide spraying during emergency operations . The program conducted 120 , 000 house searches for mosquito developmental sites and 37 , 000 larvicide applications . Random-effects regression models showed that Breteau indices declined significantly in nearly all focal cycles compared to pre-intervention indices clustered by neighborhood , after allowing for lagged effects of temperature and rainfall , baseline Breteau index , and surveillance coverage . Significant heterogeneity between neighborhoods was revealed . Larval indices seldom fell to 0 shortly after interventions at the same blocks . Large water-storage containers were the most abundant and likely to be infested . The reported incidence of dengue cases declined from 10 . 4 per 10 , 000 in 2000 ( by DEN-1 ) to 0 from 2001 to 2006 , and then rose to 4 . 5 cases per 10 , 000 in 2007 ( by DEN-3 ) . In neighboring Paraguay , the reported incidence of dengue in 2007 was 30 . 6 times higher than that in Clorinda . Control interventions exerted significant impacts on larval indices but failed to keep them below target levels during every summer , achieved sustained community acceptance , most likely prevented new dengue outbreaks over 2003–2006 , and limited to a large degree the 2007 outbreak . For further improvement , a shift is needed towards a multifaceted program with intensified coverage and source reduction efforts , lids or insecticide-treated covers to water-storage containers , and a broad social participation aiming at long-term sustainability .
The global incidence of dengue has increased exponentially since 1955 to reach 1–50 million infections per year in 2000–2005 [1] . Dengue has become the most important arboviral disease of humans , and an increasing urban health and economic problem in tropical and subtropical regions worldwide . Classic dengue fever ( DF ) and its more severe forms , dengue hemorrhagic fever ( DHF ) and dengue shock syndrome ( DSS ) , have expanded worldwide from Southeast Asia since 1946 [2] . The four existing serotypes of dengue virus ( DEN-1 , DEN-2 , DEN-3 and DEN-4 ) only confer life-long immunity against reinfection by the same serotype , and subsequent infections with a different serotype enhance the likelihood of DHF/DSS [3] , [4] . There are currently no anti-dengue drugs available , and promising vaccines still await efficacy trials [5] . Aedes aegypti , a domestic mosquito species that develops in water-holding containers , is the main vector of dengue and urban yellow fever . Current dengue control strategies seek to reduce Ae . aegypti abundance; optimize diagnosis and treatment of dengue cases , and decrease the frequency , magnitude and severity of dengue epidemics through integrated control strategies [6] . Dengue became a recognized public health threat in the western hemisphere after the 1981 Cuban epidemic , the first major DHF/DSS outbreak in the region [7] . Eradication programs initiated in 1915 achieved the apparent elimination of Ae . aegypti from most of the region by 1970 utilizing vertically-structured programs that included full-coverage source reduction supplemented with perifocal insecticide spraying with DDT [7] , [8] . During the 1970s , however , Ae . aegypti reinfested most of the countries from where it had been eliminated and is no longer considered a target for elimination [7] . Despite the transition from eradication to control , very little information on the long-term effects of vector control actions are available [9] , [10] and success stories are limited . Moreover , “Experience of vector control programs at various program levels and of successful and unsuccessful disease and vector surveillance systems needs to be recorded to allow adoption of best practices in other places” [1] . After a contained dengue outbreak by DEN-1 and DEN-4 in Roraima ( northern Brazil ) in 1981–1982 , the Southern Cone countries of South America began to suffer major DF outbreaks in 1986 ( Brazil ) , 1987–1988 ( Bolivia ) , 1988–1989 ( Paraguay ) , 1998 ( Argentina ) and 2002 ( the island of Pascua , Chile ) [7] , [11] . Uruguay and continental Chile have remained free of local dengue outbreaks , though they notified imported cases in 2006–2007 . Northern Peru has had dengue activity since 1990 , but the southern departments have remained free of Ae . aegypti and dengue so far . In Brazil , the number of reported dengue cases between 1995 and 2007 ranged from approximately 100 , 000 to 800 , 000 per year [11]–[14] . The northeastern and southeastern Brazilian states were invaded by DEN-1 in 1986–1987; by DEN-2 in 1990–1991 , and by DEN-3 in December 2000 . DHF/DSS cases have occurred every year in Brazil since 1994 [14] and in Bolivia since 2002 . In Argentina , the last epidemic outbreak of DF before the eradication era occurred in eastern and northeastern provinces in 1926 [15] . Declared eradicated in 1963 , the presence of Ae . aegypti was again detected in two northeastearn provinces in 1986 , and most provinces north to 35°S were found to be reinfested by 1999 [16] . The number of localities infested currently exceeds those recorded before the eradication era [17] . The first local outbreak of DF ( by DEN-2 ) occurred in the northwestern Province of Salta in 1998 [18] , [19] , and was most likely linked to an underreported outbreak of DEN-2 emerging in Bolivia in 1996 and 1997 [20] , [21] . The second outbreak of dengue ( by DEN-1 ) occurred in the northeastern provinces of Misiones and Formosa in 2000 , on the border with Paraguay where a massive DEN-1 outbreak occurred in 1999–2000 [22] . The emergence of DEN-3 in Salta in early 2003 ( co-circulating with DEN-1 and DEN-2 ) was shortly followed in 2004 by significant outbreaks of DEN-3 there and in the neighboring provinces of Jujuy and western Formosa [23] , and in 2006 in Salta and Misiones ( by DEN-2 and DEN-3 ) . The historical pattern of dengue case notifications in the Southern Cone between 1995 and 2007 shows increasing hyperendemic transmission of several serotypes with multiannual fluctuations in Brazil , Bolivia and Paraguay , with epidemic behavior in northern Argentina . Clorinda , the most affected city in Formosa during the 2000 outbreak , is located in one of the high-risk zones of potential dengue transmission in Argentina [24] . In collaboration with the local Municipality and other health and research institutions , Fundación Mundo Sano ( FMS ) launched a citywide control program in late 2002 with the objective of “reducing the risk of occurrence of autochthonous cases of dengue in Clorinda and its area of influence through the application of strategies that reduce the population abundance of Ae . aegypti” . The intervention program had a health promotion and education component but relied primarily on the systematic application of larvicides combined with source reduction efforts . Between 2003 and 2007 , the program conducted and monitored 14 cycles of focal treatment , with only the first previously reported [25] . As part of a cooperative agreement between FMS and the University of Buenos Aires established in late 2006 , we herein describe the implemented intervention program and assess the long-term effects of vector suppressive actions on Ae . aegypti larval indices and on the reported incidence of dengue during the five-year period . We also sought to diagnose major limitations in the control strategy conducted , and prescribe possible solutions for improved vector control .
The city of Clorinda ( 25°17′S , 57°43′W , Formosa , Argentina ) is located on the southern margin of Pilcomayo River at 45 km from the city of Asunción ( population , 519 , 000 people in 2005 ) in Paraguay ( Figure S1 ) . Clorinda had 38 identified neighborhoods with 10 , 752 houses and 47 , 240 inhabitants in 2001 , and approximately 16 , 000 houses and 49 , 000 people in late 2007; individual neighborhood size differed greatly from <10 to 133 blocks [26] . Nearly 25% of the population had unsatisfied basic needs ( i . e . , an index of poverty that combines lack of adequate housing and of tap water , crowding , and low income ) , and 15% of the houses were of mud-and-thatch construction with earthen floors . Approximately 2 , 000–7 , 000 people were estimated to commute daily between Clorinda and Paraguay through two bridges . Although the water service covers most of the city , the distribution of potable water has traditionally been rather discontinuous and unreliable . This fact determines that many residents store water in tanks , barrels and drums or construct wells . The Municipality supplies potable water in community tanks in some peripheral neighborhoods , but this does not fully satisfy the demand . Solid garbage disposal is discontinuous , inappropriate and mainly covers the most affluent neighborhoods in downtown Clorinda . Meteorological data were collected from January 1 , 2003 to December 31 , 2007 by a local weather station run by Cooperativa de Provisión de Obras y Servicios Públicos Clorinda Limitada . Mean temperature was 23 . 1°C , with mean monthly maxima ( 32 . 6–34 . 8°C ) in January–February and minima ( 8 . 9–12 . 3°C ) in May–August ( Figure 1 ) . Annual cumulative rainfall ranged from 1 , 160 to 1 , 559 mm/year of which 997–1 , 160 mm occurred between October and April . Mean relative humidity was 75% ( mean monthly maximum , 93% , and minimum , 52% ) . Dengue is a reportable disease in Argentina . Prior to 2000 , dengue virus activity was limited in Clorinda ( i . e . , 36 imported cases in 1999 ) . In February–May 2000 Clorinda experienced a classic-dengue outbreak with 529 suspected and 74 serologically confirmed cases [19] , [27] . Most cases were 15–49 years of age , and one-third of them reported travel history to Paraguay before the onset of symptoms [27] . In immediate response to this outbreak , febrile syndromic surveillance commenced at the public hospitals and health care centers , and emergency vector control operations ( focal treatment with 1% temephos , indoor space spraying with pyrethroid insecticides , citywide vehicle-mounted ULV insecticide spraying , and elimination of discarded containers in the least affluent neighborhoods ) were initiated . Vector suppressive actions only covered 5% of Clorinda during the second semester of 2001 , when the overall prevalence of IgG antibodies to dengue virus was 11 . 9% ( Elena Pedroni , unpublished data , December 2001 ) . Estimates of house and Breteau indices were 22% and 54 , respectively . Emergency vector control operations over three weeks reduced house index only by ∼45% shortly after interventions . In collaboration with the Municipality of Clorinda , the Ministry of Health of Argentina and researchers from Centro de Investigaciones de Plagas e Insecticidas ( CIPEIN ) , FMS launched a citywide control program in late 2002 following general guidelines for dengue control programs [7] . Originated in the yellow fever tradition , the desired control targets were a house index <1% and a Breteau index <5 . The preparation phase started in November 2002 with a media campaign in radio , television and newspapers , supplemented by the distribution of 5 , 000 posters and 20 , 000 leaflets to inform the community and to request that access to houses was given to vector control personnel [25] . Field activities were carried out by 40 community members affiliated to a welfare support plan run by the Municipality of Clorinda . This task force had previously been trained by vector control personnel and researchers on the basics of dengue and Ae . aegypti biology , geographic reconnaissance , collection and identification of immature stages of Ae . aegypti , application of larvicides , treatment and disposal of discarded containers , how to interview householders and to record data in household visit forms . The data collected over all cycles of focal treatment , monitoring and indoor house sprays were entered in an AccessR database to calculate the outcome parameters: the house index ( i . e . , the percentage of inspected houses that were positive for Ae . aegypti larvae or pupae ) ; the Breteau index ( i . e . the number of water-holding containers that were positive for Ae . aegypti larvae or pupae per 100 houses inspected ) ; the percentage of all visited houses that were closed or vacant or whose occupants were unavailable on first visit; the percentage of all visited households who denied entry to premises , and of all visited houses that were treated with larvicides ( temephos or BTI ) in each cycle . The first and third quartiles of the citywide house and Breteau index were calculated with package Hmisc in R software [28]; quartiles were based on the appropriate larval index value at each neighborhood j weighted by the number of houses inspected at each j and focal cycle i ( i . e . , time post-intervention , range 1–14 ) . Breteau and house indices were checked for normality and then transformed to logarithms to the base 10 of count+1 . Two-tailed paired t tests were used to examine differences between larval indices recorded in the preliminary survey and the first focal treatment cycle at each neighborhood . Odds ratios ( OR ) and incidence rate ratios ( IRR ) were estimated to assess the effect size of control interventions at each neighborhood j and cycle i compared with pre-intervention ( baseline ) values at each j and i = 1 by fitting random-effects regression models clustered by neighborhood to time-dependent larval indices using Stata 9 . 0 [29] . The use of random-effects models responds to the fact that households within a neighborhood roughly share the same environment and other undetermined characteristics that can create dependencies between responses ( positive containers or houses ) within the same geographic unit . The outcome measures were the Breteau and house indices at each j and cycle i . The main model included as explanatory variables the effects of focal cycle number ( a categorical variable that compares each cycle 2–14 to the pre-intervention cycle 1 ) ; temperature and rainfall at different time lags before the exact onset of cycle i at j ( see selection procedures below ) ; the intensity of surveillance coverage at j and i−1 ( i . e . , a continuous variable measured by the proportion of total houses at j that were inspected at the immediately preceding focal cycle ) , and the corresponding baseline larval indices at j and i = 1 ( pre-intervention ) . Pre-intervention larval indices were included as predictors based on the notion that the most infested neighborhoods before the intervention program may continue having more foci after interventions . The entire set of predictors was selected a priori based on existing knowledge . Variations in house and Breteau indices were tested with the commands glm ( family binomial ) and xtmixed using maximum likelihood procedures in Stata 9 . 1 . The short-term effects of interventions on larval indices ( as determined by the early post-intervention surveys at cycles 1–7 ) at fixed blocks within infested neighborhoods were assessed using the same procedures described before for each cycle separately . A dummy variable measured the effects of interventions at each cycle , and a time lag of 7 days was used for temperature and rainfall in all cycles . Selection of the most adequate time lag and variables for representing weather effects was based on Akaike Information Criterion ( AIC ) [30] . Four sets of candidate variables were identified: mean daily temperature; minimum daily temperature; maximum daily temperature , and daily rainfall averaged for each neighborhood at every focal cycle . Each of these variables was computed for several weekly time lags ( range , 0–7 weeks ) . The set of temperature variables was considered linearly or by adding a quadratic term to account for putative non-linear effects . For each set of weather-related variables , models including non-weather variables were fitted to Breteau indices for each time lag . The best lag was identified by the model with the smallest AIC . Models with every combination of the four selected lagged variables were computed in order to select the best model including all the variables considered .
Mean house ( 19 . 5% ) and Breteau indices ( 22 . 5 ) at the preliminary survey in late 2002 were consistent with those recorded before interventions in late 2001 ( 22% and 54 , respectively ) . Neighborhood-specific Breteau indices decreased between the preliminary survey and the first focal cycle before larviciding operations but the difference was not statistically significant ( t = 1 . 59 , df = 33 , P = 0 . 122 ) . In contrast , the observed decrease in neighborhood-specific house indices was significant ( t = 2 . 42 , df = 33 , P = 0 . 021 ) . Mosquito larvae or pupae collected during surveys were predominantly Ae . aegypti; only Culex sp . ( but no Aedes albopictus ) were detected in a few containers . House and Breteau indices recorded at the first focal cycle varied widely between neighborhoods ( coefficient of variation , 70% and 89% , respectively ) . The distribution of Breteau indices was bimodal and highly skewed , with 10 neighborhoods exceeding 30 . Between late 2002 and 2007 , ≈170 , 000 households were visited and of these , 120 , 000 were surveyed for Ae . aegypti and 37 , 000 were treated with larvicides ( Table 1 ) . Household inspections per cycle averaged 8 , 511 ( SD , 2 , 127; range , 5 , 587 to 12 , 366 ) ( Table 1 and Figure S2 ) . On average , 25% ( SD = 2% ) of the houses visited in each cycle were closed or vacant or with householders temporarily absent on first visit , and 3% ( SD = 1% ) denied entry for inspection of premises ( Figure S2 ) . The denial rate declined from 7% at baseline to 3% at 21 months post-initial intervention ( MPI ) ( focal cycle 6 , November 2004 ) but increased again to 5% at 29–33 MPI ( cycles 8–9 , June–October 2005 ) , only to decline again to 1–2% at 41–47 MPI ( cycles 12–14 , August 2006–June 2007 ) when a dengue outbreak emerged in Paraguay in early 2006 . The percentage of visited houses that were treated with larvicides averaged 22 . 2% ( SD = 2 . 8% ) . The ratio between larvicide-treated houses and Ae . aegypti positive-houses averaged 5 . 8 ( SD = 3 . 1 ) but varied widely from 2 . 8 to 13 . 5 over time . The total number of positive containers detected ( mean = 738 , SD = 418 ) and the kilograms of temephos applied ( mean = 193 kg , SD = 45 ) at each focal cycle were not significantly correlated ( r = 0 . 350 , n = 14 , P>0 . 1 ) . Citywide house indices declined sharply from 13 . 7% at baseline to 3 . 7% at the second focal cycle conducted mostly through spring of 2003 whereas Breteau indices declined from 19 . 0 to 4 . 8 ( Figure 2 ) . Larval indices then fluctuated seasonally and peaked every year between summer and early fall , with large variations between neighborhoods within anyone focal cycle as expressed by interquartile ranges . Monthly house and Breteau indices at a citywide scale over the five years were highly positively correlated ( r = 0 . 966 , n = 60 , P<0 . 001 ) . Weather-related variables exerted highly significant effects on larval indices , especially when time lags were allowed for ( Table 2 ) . Based on AIC scores , the best lags were 1 week for mean daily rainfall and mean minimum daily temperature , and 4 weeks for mean daily temperature and mean maximum daily temperature . In every case the best lag identified clearly surpassed other candidate lags ( range of ΔAIC of weather variables , 5 . 6–11 . 0 ) . Because substantial model selection uncertainty was found when comparing models including temperatures with and without quadratic terms , the linear variables were selected for the remaining analyses . Using the selected time lags for each variable , the best model found included mean rainfall , mean temperature and mean minimum temperature . All of the selected variables had a positive association with larval indices ( Table 3 ) . Control actions exerted significant impacts on larval indices , with exception of a limited upsurge at 36 MPI ( cycle 10 , February 2006 ) ( Figure 2 ) . When post-intervention larval indices at focal cycles 2–14 were compared to pre-intervention indices at cycle 1 by random-effects multiple regression , log-transformed Breteau indices declined significantly ( P<0 . 05 ) or highly significantly ( P<0 . 001 ) in all cycles except cycle 10 and 12 after allowing for lagged effects of temperature and rainfall ( P<0 . 001 ) , baseline Breteau index ( P<0 . 001 ) , and surveillance coverage ( P = 0 . 512 ) at the preceding focal cycle ( Wald χ2 test , df = 18 , P<0 . 001 ) ( Table 3 ) . Significant heterogeneity between neighborhoods was indicated by the standard deviation of the μ parameter representing between-cluster variations ( Likelihood ratio test , χ2 = 89 . 5 , df = 1 , P<0 . 001 ) . The residual intraclass correlation ( ρ ) was 0 . 30 , thus indicating that 30% of the variance in post-intervention log Breteau indices that is not explained by the covariates is due to time-invariant neighborhood-specific characteristics . House indices were significantly reduced by 13–54% relative to pre-intervention indices in cycles 2–3 , 5–7 , 9 , and 14 . Post-intervention and pre-intervention Breteau indices were positively and significantly associated , and post-intervention house indices were negatively and significantly associated with surveillance coverage at the preceding cycle . The early post-intervention surveys after focal cycles 1–7 showed that larval indices seldom fell to 0 at the same infested spatial units shortly after treatment ( Figure 3 ) . Random-effects multiple regression showed that pre-intervention house indices were reduced significantly at focal cycles 1 ( from 12 . 2% to 3 . 5% , P = 0 . 001 ) and 4 ( from 9 . 7% to 3 . 0% , P = 0 . 047 ) , with borderline significant effects at cycle 5 ( P = 0 . 057 , from 5 . 5% to 2 . 7% ) . Average pre-intervention house indices were barely reduced at other cycles ( P>0 . 3 ) . Breteau indices showed similar patterns . The abundance and infestation of water-holding container types and distribution of larval indices differed largely among types of container . For example , at focal cycle 12 ( spring–summer 2006–2007 ) , tanks , barrels and drums for water storage ( B type ) were the most abundant containers ( 4 , 380 ) and the most likely to be infested ( 15 . 2% ) , accounting for 49% of all infested containers found ( Figure 4 ) . The second most important container class was disposable bottles , cans and plastics ( E type ) , which were abundant ( 1 , 476 ) and as frequently infested ( 15 . 0% ) as B containers , but accounted for only 16% of all infested containers . The frequency distribution of containers per type in cycle 12 was highly overdispersed between neighborhoods , with coefficient of variations that increased from 36% ( type B ) to 288% ( D ) . The reported incidence of classic DF cases in Clorinda declined from 10 . 4 per 10 , 000 inhabitants in 2000 ( 46 confirmed cases including 20 autochthonous cases by DEN-1 , and about 500 suspect cases ) to 0 from 2001 to 2006 [22] , [27] , and then increased up to 4 . 5 cases ( by DEN-3 ) per 10 , 000 in January–April 2007 [31] ( Figure 5A ) . Of 267 suspect DF cases in 2007 , 21 were confirmed ( including only 5 autochthonous cases ) ; 86 were excluded as dengue , and 160 remained without confirmation [31] . Meanwhile in Paraguay , following the 1999 outbreak with 1 , 164 reported cases , 24 , 282 cases by DEN-1 were reported in 2000 ( incidence , 49 . 3 per 10 , 000 ) though estimates ranged up to 300 , 000 [32] ( Figure 5B ) . This outbreak was followed by a low-level transmission period with decreasing number of cases from 1 , 871 in 2002 ( by DEN-1 , DEN-2 and DEN-3 ) , to 137 and 164 cases in 2003 and 2004 ( by DEN-3 ) , and another upward trend with 405 cases in 2005 ( by DEN-2 ) and 4 , 271 cases ( by DEN-3 ) in 2006 [11] . In Asunción , with high infestation levels , 1 , 700 DF cases were reported in 2006 [12] , and the incidence of DF was 135 . 7 per 10 , 000 inhabitants in January–April 2007 [13] . The 2007 outbreak included 28 , 130 reported cases by DEN-3 ( up to week 21 , several with unusually severe manifestations ) , 55 confirmed cases of DHF/DSS , and 17 deaths probably related to dengue [33] .
Gubler [2] noted that “… The sporadic nature of dengue epidemics and the misguided reliance on using insecticidal space sprays to kill adult mosquitoes prevented most countries from developing and implementing programs that focused on larval mosquito control , which were much more difficult to implement and maintain” . Here we advocate that such programs are still needed; they should be run permanently with high coverage , especially in a hyperendemic context with regional expansion of DHF/DSS cases , and may be mostly maintained with locally available resources , duly coordinated and supervised . This does not preempt the fact that governments must invest much more resources and efforts on scientifically-based vector control programs run by qualified personnel than they have done so far . Although the Clorinda intervention program did not reach target levels , it had a positive impact on public health because it prevented the serious dengue outbreaks that occurred in neighboring countries during the study period . Without the interventions , at the reported incidence rates in Paraguay the situation in Clorinda would have been much worse . The Clorinda experience was successful at some program levels and left some lessons for further improvement . The use of controlled-release insecticide formulations or sachets that are retrievable during cleaning and washing of water-storage containers would extend the residual activity of temephos [40] . There is a great need of new , more effective larviciding products that last longer , but they will also face water management and coverage issues . The upward trend observed in temephos resistance indicates that resistance management schemes should be developed and considered in the near future , because RRs around 3 do not revert spontaneously to pre-intervention susceptibility levels . Resistance to temephos with potential cross-resistance to other insecticides has expanded greatly and caused repeated control failures in Brazil [60] , [61] , and has been detected elsewhere in Argentina [62] . Pyriproxyphen is a relevant candidate for larvicide replacement [10] . Identification of neighborhoods at increased risk of infestation and transmission are needed for developing more cost-effective , targeted control strategies . Several sources of heterogeneity pose major challenges to the control of Ae . aegypti . The implemented program was born as a community-based intervention by some definitions [58] and gradually turned into a top-down vector control program suffering from excessive reliance on insecticides . In Clorinda , current interventions should evolve towards a multifaceted integrated program with intensified coverage , source reduction and environmental management measures , such as providing lids or insecticide-treated covers to water storage containers [10] , [63] . Such integrated program needs also strengthened communication and health education components [64] and a broad social participation aiming at long-term sustainability . Other strategic solutions include development of the infrastructure for providing potable water and improved disposal of solid waste . Vector control and disease management must remain a regional effort to prevent “spillovers” such as those from Asunción to Clorinda or from Brazil to Paraguay , within the frame of sustainable development rather than being viewed exclusively as a matter of health [59] . These issues are common to other neglected tropical diseases affecting vulnerable populations in the Gran Chaco region over Argentina , Bolivia and Paraguay [65] . The price of not establishing regional control efforts of a more permanent nature has lead to the predicted [3] and observed expansion of DHF/DSS in the Americas , which reached Paraguay in 2007 . | Dengue has become the most important viral disease of humans transmitted by arthropods in tropical and subtropical urban regions . Most countries have not been able to maintain permanent larval mosquito control programs against Aedes aegypti , the main vector of dengue and yellow fever , partly because of misguided reliance on using insecticidal space sprays to kill adult mosquitoes . Control actions targeting larvae are based on regular treatment with insecticides of every mosquito developmental site , usually artificial containers found inside or around human dwellings . The reasons why sustained larval control programs cannot reduce infestations to the desired levels have rarely been investigated . A five-year intervention program conducted in Clorinda , northeastern Argentina , reduced significantly larval infestations compared to pre-intervention levels . Infestations depended on weather variations , pre-intervention larval infestation , and the percentage of houses that were visited and treated with larvicides . Although the program did not reach the low levels of infestation desired , it most likely prevented or limited new local dengue outbreaks . Large containers for permanent water storage were the most important mosquito development site . For further improvements , a multifaceted intervention program is needed . It should include intensified surveillance and treatment coverage with larvicides that last longer , more efforts on reducing the potential number of mosquito development sites , and a broad social participation aiming at long-term sustainability . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"public",
"health",
"and",
"epidemiology/epidemiology",
"infectious",
"diseases/viral",
"infections",
"public",
"health",
"and",
"epidemiology/infectious",
"diseases",
"ecology/population",
"ecology",
"infectious",
"diseases/epidemiology",
"and",
"control",
"of",
"infectious",
"diseases"
] | 2009 | Effects of a Five-Year Citywide Intervention Program To Control Aedes aegypti and Prevent Dengue Outbreaks in Northern Argentina |
When Drosophila melanogaster embryos initiate zygotic transcription around mitotic cycle 10 , the dose-sensitive expression of specialized genes on the X chromosome triggers a sex-determination cascade that , among other things , compensates for differences in sex chromosome dose by hypertranscribing the single X chromosome in males . However , there is an approximately 1 hour delay between the onset of zygotic transcription and the establishment of canonical dosage compensation near the end of mitotic cycle 14 . During this time , zygotic transcription drives segmentation , cellularization , and other important developmental events . Since many of the genes involved in these processes are on the X chromosome , we wondered whether they are transcribed at higher levels in females and whether this might lead to sex-specific early embryonic patterning . To investigate this possibility , we developed methods to precisely stage , sex , and characterize the transcriptomes of individual embryos . We measured genome-wide mRNA abundance in male and female embryos at eight timepoints , spanning mitotic cycle 10 through late cycle 14 , using polymorphisms between parental lines to distinguish maternal and zygotic transcription . We found limited sex-specific zygotic transcription , with a weak tendency for genes on the X to be expressed at higher levels in females . However , transcripts derived from the single X chromosome in males were more abundant that those derived from either X chromosome in females , demonstrating that there is widespread dosage compensation prior to the activation of the canonical MSL-mediated dosage compensation system . Crucially , this new system of early zygotic dosage compensation results in nearly identical transcript levels for key X-linked developmental regulators , including giant ( gt ) , brinker ( brk ) , buttonhead ( btd ) , and short gastrulation ( sog ) , in male and female embryos .
The earliest stages of animal development are under maternal control until mRNAs deposited prior to fertilization degrade and zygotic transcription is initiated during a period known as the maternal to zygotic transition ( MZT ) . In Drosophila melanogster , the MZT occurs amidst the 14 rapid and synchronous mitotic divisions that mark the first several hours of development , with zygotic transcripts appearing as early as mitotic cycle 8 [1] . By cycle 14 , when cellularization of the previously syncytial blastoderm occurs , most processes are under the control of zygotic transcripts . As zygotic transcription begins , the different numbers of X chromosomes ( two in females , one in males ) results in different transcript levels for a small number of genes on the X chromosome ( the X chromosome signal elements , or XSEs ) , which lead to female-specific expression of the master sex control gene Sex lethal ( Sxl ) [2]–[4] . The low levels of SXL in males lead to the male-specific formation of a dosage compensation complex composed of five proteins ( MSL-1 , MSL-2 , MSL-3 , MOF , MLE ) and two non-coding RNAs ( rox1 and rox2 ) that bind to the X chromosome , hyperacetylate histone H4K16 , and induce hypertranscription of the male X chromosome [5]–[8] . However , there is a lag between the onset of zygotic transcription and the establishment of MSL-mediated dosage compensation: the complex is not localized on DNA , and H4K16 acetylation is not detectable , until after the blastoderm stage [9] , [10] , 60 to 90 min after the onset of zygotic transcription . During this gap , zygotic transcription drives a host of important developmental processes , including segmentation along the anterior-posterior axis , the establishment of tissue layers along the dorsal-ventral axis , and cellularization . These events often require the precise spatial localization and concentration of transcription factors and other proteins . It is therefore interesting that many important blastoderm regulators are on the X chromosome , and thus present in varying dosage in males and females , including the A–P factors giant ( gt ) , buttonhead ( btd ) , orthodenticle ( otd ) and runt ( run ) , D–V factors brinker ( brk ) , short gastrulation ( sog ) and neijire ( nej ) , and the cellularization factor nullo . We were intrigued by the possibility that the absence of MSL-mediated dosage compensation during the MZT might lead to higher levels of mRNAs derived from genes on the X chromosome in females , and sex-specific differences in patterning or cellularization that have not been detected because systematic studies of early developmental transcription have never differentiated male and female embryos . A variety of approaches have been used to profile zygotic transcription during the MZT , including genome-wide expression profiling with microarrays [11]–[15] and in situ hybridization [16] . However the genomic studies pooled mixed-sex embryos based only on developmental time , and generally have not had sufficient temporal resolution to distinguish events during the rapid mitotic cycles of early development . Embryos produced to lack entire chromosomes or chromosome arms have been used to distinguish maternal and zygotic transcription [12] , but the effects of these significant aberrations are unknown . Imaging studies have intrinsically higher temporal resolution , and have used differences in RNA localization to begin to unravel the maternal and zygotic contributions to mRNA pools . But doing such experiments on a genomic scale requires considerable time and resources , and current imaging projects do not distinguish male and female embryos . To address these limitations , we developed methods to characterize , by sequencing , the mRNA content of individual D . melanogaster embryos , which we combined with methods to precisely stage and sex single embryos to generate sex-specific time courses of maternal and zygotic transcript abundance spanning the first wave of early zygotic transcription through the MZT to the end of the blastoderm stage when MSL-mediated dosage compensation is thought to begin [9] , [10] .
We chose to focus on the period of development bounded by cycle 10 ( when early zygotic transcription is detectable ) and the completion of cellularization in mitotic cycle 14 ( when widespread zygotic transcription has been established , right before MSL-mediated dosage compensation is thought to begin ) . To determine developmental stage , we took advantage of two characteristics of early embryos: the tightly controlled synchronous mitotic cycles and the process of cellularization as the embryo transitions from a syncytium to a cellular blastoderm ( Figure 1 ) . We examined live embryos from a maternal line carrying an RFP-labeled histone under a fluorescent microscope and used a combination of direct observation of mitotic cycles and quantification of nuclear density to select embryos during interphases of mitotic cycles 10 , 11 , 12 , 13 , and 14 . Stage assignments were based on examination of the entire embryo to avoid cases where different portions were in different mitotic cycles [17] . We further refined the staging within cycle 14 by examining embryos under a light microscope and quantifying the extent of membrane invagination during cellularization , assigning embryos to stages 14A ( 0%–25% invagination ) , 14B ( 25%–50% ) , 14C ( 50%–75% ) , and 14D ( 75%–100% ) . Selected embryos were immediately immersed in TRIzol , ruptured , and frozen for subsequent extraction . We selected at least four embryos each for cycles 10 , 11 , 12 , 13 , 14A , 14B , 14C , and 14D , and extracted DNA and RNA from each embryo independently . We carried out whole-genome amplification on the DNA from each embryo and genotyped it for Y chromosomal markers to determine the sex of the embryo , and selected at least one male and female embryo from every stage for transcriptome analysis . Figure 1 shows the embryos we selected immediately before DNA and RNA were extracted . We obtained 75 to 100 ng of total RNA from each embryo . As this was less starting material than required for standard mRNA sequencing protocols , we modified the Illumina mRNA-Seq protocol to obtain reliable data from such small quantities of input mRNA without amplification by performing all purification and size selection steps using magnetic beads , and reducing the volume of some reactions ( a complete protocol is available in Protocol S1 ) . These relatively minor alterations were sufficient to lower the amount of starting material required by more than an order of magnitude . We sequenced a total of 24 mRNA samples on an Illumina GAIIx Genome Analyzer . We aligned reads to the D . melanogaster reference sequence ( version 5 . 23 ) using Bowtie [18] and inferred transcript levels using TopHat [19] and Cufflinks [20] . We normalized expression levels between samples so that the total inferred expression levels of autosomal transcripts were identical . Statistics on the sequencing and mapping are reported in Table 1 . The single embryo mRNA-Seq method was highly reproducible and has a wide dynamic range ( Figure 1C ) . Transcript levels over all genes from individuals of the same sex and stage had correlation coefficients from 0 . 95–0 . 97 ( Spearman's rank correlation ) ; transcript levels from individuals of the same stage but different sex were correlated to a similar degree . In contrast , transcript levels from embryos of the same sex but different stage had correlations ranging from 0 . 80–0 . 97 . In order to distinguish zygotic transcripts from those deposited by the mother , we analyzed embryos produced by a cross of two genetically distinct D . melanogaster lines: a w1 derived maternal line ( which contained the His2Av-RFP marker ) and a Canton-S ( CaS ) paternal line . We sequenced both lines to roughly 35× coverage ( see Table 2 ) , mapped reads to the reference genome using maq ( maq . sourceforge . net ) , and identified 285 , 927 sites that differed between the strains . The vast majority of these differences were biallelic single-nucleotide polymorphisms ( SNPs ) known from resequencing projects to be polymorphic in the North American D . melanogaster population ( dpgp . org and DGRP ) . This is consistent with these strains representing independent samples drawn from resident populations in the United States . Although both lines have been in laboratories for decades , we found that each harbored a significant amount of residual polymorphism , especially CaS . We therefore restricted our subsequent analyses to a set of 122 , 672 SNPs that were fixed between strains . Exactly 10 , 492 of 14 , 833 annotated genes ( over 70% ) contained at least one fixed polymorphism , allowing us to assign RNA-Seq reads spanning the polymorphism to either w1 or CaS ( Figure 2A ) . Since maternally deposited mRNAs should all be w1 , any CaS ( paternal ) reads must have been the result of zygotic transcription . We were thus able to partition the overall expression of any mRNA containing w1-CaS differences into its maternal and zygotic component ( Figure 2B ) . As expected , transcripts at cycle 10 were almost entirely maternal ( Figure 2C ) . We observed widespread zygotic transcription beginning in the middle of cycle 14 , and by the end of cycle 14 , we find a mix of persistent maternal and zygotic transcripts , in varying proportions , depending on the gene ( Figure 2C ) . We used the strain-specific time series to classify genes as maternal , zygotic , or maternal and zygotic . Briefly , we clustered ( k-medians ) the 5 , 226 genes with at least 10 reads spanning a w1-CaS polymorphism into 20 groups based on similarity of their inferred abundance of maternally and paternally derived transcripts . We classified each cluster as maternal ( only w1 mRNAs detected with levels declining over time ) , zygotic ( no mRNA at cycle 10 , with both w1 and CaS alleles detected over time ) , or maternal and zygotic ( only w1 mRNAs detected at cycle 10 , with CaS mRNAs appearing over time ) . Because of the absence of paternal alleles for genes on the X chromosome in males , all assignments were based on data from females only . We classified genes lacking polymorphisms distinguishing the strains by comparing their mRNA abundances from the eight female samples to the average patterns from each of the previously assigned groups . We assigned genes to the group with which their expression pattern was best correlated ( if the correlation coefficient was greater than 0 . 8 ) . Overall , 5 , 598 genes were classified as maternal , 2 , 210 as zygotic , and 1 , 195 and maternal+zygotic ( the classification for each gene is listed in Dataset S1 ) . Previous studies of sex determination and dosage compensation have described the expression sex-specific patterns of expression in a number of zygotically transcribed genes [2]–[5] . We examined the expression patterns of these genes in our data to confirm that we could effectively detect transcript differences in zygotically transcribed genes between male and female embryos ( Figure 3 ) . As expected , we observed that the numerator genes sisA , sisB ( also known as sc ) , and run are expressed at higher levels in females ( twice as high during cycles 11–12 , Figure 3A ) , that early Sxl expression is substantially higher in females ( Figure 3B ) , and that msl-2 is more abundant in males ( Figure 3C ) . We did not observe msl-2 transcript until the middle of cycle 14 , consistent with earlier studies demonstrating that MSL-mediated dosage compensation is not established until after cellularization [9] , [10] . Collectively , these data establish that we can reliably detect sex-specific differences in expression where they exist . We next compared transcript levels of all 2 , 210 purely zygotic genes in male and female embryos . Zygotically derived transcripts from autosomal genes were observed at the same levels in females and males ( Figure 4A ) . In contrast , zygotically derived transcripts from the X chromosome were consistently observed at higher levels in females than in males ( Figure 4A ) , with a female to male ratio ranging from 1 . 0 to 2 . 0 . The female to male ratio , and thus the level of compensation , did not correlate with expression level of the gene , or the position of the gene on the X chromosome . The difference between the X chromosome and autosomes can be seen clearly when total abundance of zygotically expressed genes in males and females is compared between the X chromosome and autosomes ( Figure 4B ) . Autosomal transcript levels were effectively identical in females and males at all time points , and X chromosome transcript levels were higher in females , yet not twice as high as in males . The ratio of transcript levels of zygotic genes from the female to male X chromosomes was approximately 1 . 45 over cycle 14 ( mean of 1 . 5 , median of 1 . 4 , over all X chromosomal zygotic genes; for zygotic genes on chromosome 2L , mean and median female to male ratios are 1 . 1 and 1 . 0 , respectively ) . We observed no difference in the levels of transcripts derived from maternal or paternal chromosomes for either autosomes or ( in females ) the X chromosome . Total expression of zygotic genes from the paternal and maternal X chromosomes of females was very similar ( average Spearman's rank correlation ρ = 0 . 97 across stages , some as high as ρ = 0 . 999; Figure 4C ) . The total abundance of zygotic genes from the single male X chromosome was consistently higher ( Figure 4C ) than from either female X chromosome—demonstrating that transcript abundance in the early embryo transcription is subject to some form of dosage compensation . The sex ratio of transcript abundance for individual genes varied somewhat over cycle 14 , especially earlier in cycle 14 where there are not many zygotic genes expressed . But across cycle 14 , the X chromosome consistently had an excess of genes with higher transcript level in females ( Figure 5A ) . Yet by the end of cycle 14 , there were few genes on the X chromosome that had a 2-fold excess of transcript levels in females , and only approximately 30% had more than a 1 . 75-fold enrichment . Approximately half of the factors on the X chromosome had less than a 1 . 5-fold excess of transcripts in females . The expression patterns of the patterning genes whose presence on the X chromosome motivated us to examine sex-specific expression in the early embryo were particularly striking . For example , giant , a textbook example of an important early embryonic regulator on X for which differences in levels would likely impact development [21] , was almost perfectly dosage compensated ( Figure 5B ) , with equal transcript levels in males and females corresponding to roughly 2-fold greater abundance of mRNAs derived from the paternal X chromosome compared to the X chromosomes in females . Other key X-linked developmental regulators , including vnd , nullo , btd , tsg , and sog , were also present at roughly equal levels in males and females . As is often the case with dosage compensation mechanisms , early zygotic dosage compensation is not universal , and several genes showed no evidence of compensation at the transcript level ( Figure 5C ) . We assigned every zygotic gene with appreciable expression levels ( maximum normalized RPKM greater than 3 . 0 ) a compensation score equal to the slope of the line fit ( by least squares ) to the male and female transcript levels for that gene . The distribution of these values for autosomal genes were centered around 1 . 0 and rarely showed a greater than 1 . 5-fold difference ( Figure S1A ) . In contrast , 77 of the 85 zygotic genes on the X chromosome were greater than 1 . 0 and 36 were greater than 1 . 5 ( Figure S1A ) . These compensation scores are given in Dataset S1 , and plots of all zygotic genes on X sorted by this score are shown in Figure S1B .
Our development of methods to examine sex-specific gene expression in early D . melanogaster embryos was motivated by the expectation that the earliest stages of zygotic transcription are not dosage compensated and that resultant sex differences in the levels of crucial patterning genes might have interesting phenotypic consequences . Instead , our genome-wide time course of transcript levels in individual male and female embryos has revealed extensive dosage compensation of X chromosomal transcript levels before the canonical MSL-mediated dosage compensation process is thought to be engaged . Crucially , mRNAs for key X-linked developmental regulators , including gt , brk , btd , and sog , are present at essentially identical levels in male and female embryos . Although there is clearly early zygotic dosage compensation ( EZDC ) , our data speak only indirectly to the mechanism by which it occurs . Assuming that , in an uncompensated system , we would expect transcription to produce twice as many zygotically derived copies of X chromosomal genes in females than in males , the generally lower levels we observe in females must arise through sex and X-chromosome-specific transcriptional or post-transcriptional regulation . The simplest explanation is that the MSL-based dosage compensation system is active before and during cycle 14 , leading to hypertranscription of the male X . However , several imaging studies of the male-specific localization of MSL proteins to , and the subsequent acetylation of histones on , the male X chromosome describe an at least 1 h lag between the onset of zygotic transcription and these hallmarks of MSL-mediated dosage compensation [9] , [10] . While it is possible that these studies missed earlier low-level or highly targeted MSL-binding and compensation that escaped detection in the microscope , independent evidence exists for MSL-independent dosage compensation in the early embryo [22] . Through an analysis of larval cuticle patterns of male and female embryos carrying various combinations of run hypomorphic alleles , Gergen demonstrated that the X-linked gene run , which is involved in both sex-determination and segmentation , is functionally dosage compensated [22] . Although run is expressed throughout embryogenesis , the effects on larval cuticle patterns these studies examined arise during the blastoderm stage and are thus an example of EZDC . We also find that run is dosage compensated during cycle 14 . Gergen [22] , and later Bernstein and Cline [23] , showed that dosage compensation of run is MSL independent but requires the early female-specific form of Sxl . Since SXL is an RNA-binding protein known to modulate splicing and translation , it was proposed that dosage compensation of run might result from direct SXL-mediated reduction of the translation or stability of run in females [24] . Consistent with this possibility , the 3′UTR of run mRNA contains several matches to the SXL consensus sequence [24] . However , a direct role for SXL in run dosage compensation has not been confirmed . The two best-characterized targets of SXL are msl-2 mRNA , which it regulates by translational repression , and its own mRNA , which it regulates by controlling how it is spliced . However , a total of 88 genes ( including run ) have transcripts whose 3′UTRs contain three or more SXL target sites ( AUUUUUUU or UUUUUUUU ) . And of these an astonishing 76 are on the X chromosome . This striking enrichment , originally noted by Kelley et al . [24] and expanded by Cline [25] , suggests a broad role for SXL in specifically regulating the stability or activity of mRNAs derived from the X chromosome . If the female-specific SXL is controlling EZDC directly , it would have to do so by reducing the levels of X chromosomal RNAs in females , as SXL is not present in males . While such an activity has not been established for SXL , many other RNA binding proteins are known to affect transcript levels [26]–[29] . There is , however , imperfect agreement between predicted SXL targets and genes we observe to be dosage compensated . Many genes with high degrees of EZDC are not predicted SXL targets ( Figure S1; gt , for example , is not a predicted target ) and many predicted SXL targets are not or are poorly dosage compensated ( Figure S1B ) . Furthermore , many predicted SXL targets on X are maternally deposited , with no early zygotic transcription . These genes are not expected to be affected by chromosomal dosage differences . Indeed SXL acting to reduce the levels of these genes in females would produce , rather than eliminate , dosage differences . To resolve whether SXL plays a role in EZDC , we are currently determining whether EZDC is present in Sxl mutants , and whether SXL interacts specifically with EZDC targets . If it turns out that neither the MSL complex or Sxl are required , it is possible that dosage compensation arises from gene-specific feedback . Many developmental regulators regulate their own transcription [30] , [31] , and such interactions could lead to full or partial compensation of initially higher transcript levels in females than males . However , this kind of feedback would also likely have a significant time lag between the emergence of differences in transcript levels and their compensation . There is evidence that the early embryo is generally robust to environmental factors such as temperature and some forms of genetic variation [32]–[35] . Systems conferring such robustness might also sense and compensate for deviations arising from differences in X chromosomal dose . Each of the models discussed above assume that , without intervention , 2-fold differences in DNA dose inherently produce 2-fold differences in transcription and transcript abundance , which need , at least for some subset of genes , to be compensated . However , this is not necessarily the case . Studies on autosomal regions with altered dosage in Drosophila suggest an average 1 . 3–1 . 5-fold increase in transcript level per copy [36]–[38] . Dosage compensation of the X chromosome in Drosophila results in a ∼2-fold increase in transcription in males , relative to the autosomes [36]–[38] . A recent study [39] estimates that the MSL-complex has a 1 . 35-fold effect on expression of the X chromosome in males , and suggests that X chromosome dosage compensation could simply be the interaction of this 1 . 35× effect with the baseline 1 . 5× dosage effect . However , the effects of these altered gene dosages in these experiments , which measure precise differences in expression , are unknown . It is unclear whether these dosage differences are comparable to the wild-type differences in X chromosome dosage , and how to interpret the quantitative effects as characterized . Regardless of what the baseline threshold for compensated versus uncompensated transcription is with a 2-fold dosage difference , we see many factors on the X chromosome with no difference in transcription rates in males and females . Additionally , the expectations of the interactions of gene dosage and expression may not be the same in the unique transcription environment of the early embryo . A recent study by Lu et al . [40] compared gene expression during early development in diploid and haploid embryos and found that transcript levels for a large class of zygotically transcribed genes ( those whose transcription is dependent on developmental time , rather than nucleocytoplasmic ratio ) were dosage independent . To explain this observation , Lu et al . [40] proposed a model in which transcription is limited by an unknown , maternally deposited , factor . Since both haploid and diploid embryos would have the same amount of this limiting factor , and since individual genes would be present in the same proportion to each other , rates of transcription across the genome would be the same . However , the limiting factor hypothesis cannot explain X chromosomal dosage compensation , as halving the dosage of X chromosomal genes relative to autosomal genes in males would lower the relative rate of transcription of X chromosomal genes ( compared to autosomes ) at any concentration of the limiting factor . There is a related alternative to the limiting factor hypothesis that could explain both dosage compensation and insensitivity to ploidy , concerning the accessibility of DNA templates . Homologous chromosomes are known to be paired throughout Drosophila development [41] , [42] , and imaging of nascent transcripts in the early embryo consistently shows the close proximity of transcribed alleles . Given that transcription involves localization to specific subnuclear regions and attachment to large protein machines , it seems possible that the transcription of one allele could make it difficult or even impossible to transcribe the other allele . If such an effect occurred , then the embryo will be inherently dosage compensated . If only one copy of a gene is present ( for the whole genome in haploid embryos or the X chromosome in males ) , it is transcribed at whatever rate the various regulatory systems active dictate . If two copies of the same gene are present ( as in diploids and females ) , the gene would be expressed at the haploid level , with expression divided across the two alleles . While no such mechanism has been described , the rapid mitotic cycles of early development place constraints on transcription [43] and might make the early embryo particularly sensitive to such effects . It has also long been observed in Diptera , that homologous chromosomes pair during mitosis , as well as meiosis [41] , [42] . Expression can be affected by the pairing of homologs , through phenomena such as transvection [44]–[46] , the control of genes by regulatory interactions with their homologs in trans . Pairing of some homologous loci is observed as early as cycle 13 and increases through cycle 14 [47]–[50] , precisely at the times EZDC is observed . As pairing of homologous loci also seems to occur at particular sites rather than “zippering” along a chromosome [48] , this could also explain why some sites seem compensated and others do not . Yet , contrary to this , near synchronous appearance of two adjacent dots in many nuclei in RNA in situ hybridization of intronic probes from autosomal genes demonstrates that paired alleles can both be transcribed at roughly the same time [4] , [51] , [52] . But it leaves open the possibility that the transcription of one allele could affect the rate at which the other is transcribed . Whatever the mechanism turns out to be , our data provide an unprecedented window on the temporal dynamics of transcript levels in male and female embryos , and establish that some mechanism exists that ensures that differences in sex chromosome dose do not translate into differences in mRNA abundance during a crucial period of D . melanogaster development . While our focus here was on dosage compensation , our data represent a significant advance over earlier methods to monitor gene expression in the early D . melanogaster embryo by providing higher temporal resolution and precision , sex specificity , and unambiguous discrimination of maternally deposited and zygotically transcribed mRNAs . Our use of individual embryos also provides a window onto embryo-to-embryo variability in transcript levels , which we found to be surprisingly low . We hope that our data , which are being made available in full here , will help address a number of other open questions about transcription during early D . melanogaster embryogenesis . And we suspect that the methods we developed for analyzing mRNA from individual Drosophila embryos and other aspects of our experimental design will be of interest to researchers interested in the analysis of small RNA samples . Although our experiments worked exceptionally well , in carrying them out , we made several observations that should be of interest and use to other investigators . First , we routinely obtained at least 10 times more material from processing the RNA from a single embryo than was needed for a single Illumina sequencing lane . This suggests that the RNA content of even smaller samples could be routinely analyzed without RNA amplification . Second , for a variety of reasons , mostly involving cost , we carried out 36 base pair single-end sequencing runs . In retrospect , we would have been able to assign many more reads to distinct parental chromosomes , and perhaps detected sex-specific splicing , had we carried out longer , paired-end runs . Finally , analyzing embryos from a cross of divergent strains was very useful . But we were surprised at how polymorphic the supposedly inbred strains we used in our crosses were . We suspect this is a general phenomenon , and suggest that all researchers doing experiments that require highly inbred lines specifically inbreed the lines they are using and resequence them to characterize residual polymorphism prior to use .
Flies were raised on standard fly media in uncrowded conditions , at room temperature . 2–3-d-old virgin females of the His2Av-mRFP1 III . 1 line ( Bloomington stock center , stock no . 23650 ) [53] were crossed to Canton-S males , and eggs were collected from many 3–6-d-old females , thus minimizing chances that multiple embryos sampled would come from the same mother . After collection , eggs were dechorionated , placed on a slide in halocarbon oil , and visualized using a Nikon Eclipse 80i microscope , with a Nikon DS-UI camera , and the NIS Elements F 2 . 20 software . Embryos were photographed both for fluorescence with an RFP filter to visualize nuclei and under white light with a DIC filter to visualize the extent of cellularization in mitotic cycle 14 embryos ( Figure 1B ) . Embryos were then moved from the slide , cleaned from excess oil , and placed in a drop of TRIzol reagent ( Invitrogen ) within a minute or less of imaging . Embryos are then ruptured with a needle , allowed to dissolve , and moved to a tube with more TRIzol reagent , which was then frozen at −80°C . For determining the age ( mitotic cycle ) of each embryo , images of nuclei were analyzed in ImageJ ( 1 . 42q ) , where nuclear numbers were counted in a 200×200 pixel box , to confirm the nuclear division cycle of each embryo . For those embryos within mitotic cycle 14 , the DIC photographs showing the extent of membrane invagination were used to create subclasses within cycle 14 ( Figure 1B ) . RNA and DNA extraction from single embryos was done with TRIzol ( Invitrogen ) reagent according to the manufacturer's protocol , except with a higher volume of reagent relative to the amount of material ( i . e . starting with 1 mL of TRIzol despite expecting very small amounts of DNA and RNA ) . Extracted DNA was amplified using the Illustra GenomiPhi V2 DNA Amplification Kit ( GE Healthcare ) , and embryos were sexed by detecting the presence of a Y chromosome , using PCR with primers to a region of the male fertility factor kl5 on the Y chromosome ( forward primer GCTGCCGAGCGACAGAAAATAATGACT , reverse primer CAACGATCTGTGAGTGGCGTGATTACA ) , and a region on chromosome 2R ( forward primer AAAAGGTACCCGCAATATAACCCAATAATTT , reverse primer GTCCCAGTTACGGTTCGGGTTCCATTGT ) as a control . Total RNA was made into libraries for sequencing using the mRNA-Seq Sample Preparation Kit from Illumina , following an altered mRNA-Seq library making protocol developed at Illumina ( see complete protocol in Protocol S1 ) . Libraries were quantified using the Kapa Library Quantification Kit for the Illumina Genome Analyzer platform ( Kapa Biosystems ) , on a Roche LC480 RT-PCR machine , according to the manufacturer's instructions . An alternate flow cell loading protocol for small concentration sequencing libraries was developed for this study and used here , despite the libraries created largely being concentrated enough not to necessitate use of this method ( see Protocol S2 ) . For each sample , 40 pM of library ( relative to final concentration loaded on to flow cell ) was diluted in 4 uL , and 1 uL of 0 . 5 M sodium hydroxide was added . Samples were left 5 min to denature , then placed on ice , and 1 uL 0 . 5 M hydrochloric acid added , then diluted to final loading concentration ( of at least 20 uL ) with Illumina hybridization buffer . To load sample on an Illumina flow cell , an air gap was created , the entire sample drawn into the hybridization manifold , an air gap left after the sample , and hybridization buffer used to push the sample until it is centered on the flow cell ( see complete protocol in Protocol S2 ) . The rest of the cluster generation and sequencing were according to normal protocols , for 40 cycle sequencing with the Illumina Genome Analyzer ( GAIIx ) . We prepared genomic DNA from 10 females from our CaS and w1 stocks . We prepared Illumina paired-end sequencing libraries using standard protocols and sequenced two 101 bp paired-end lanes for each strain on an Illumina GAIIx Genome Analyzer . Reads from each RNA-Seq sample were mapped to the reference D . melanogaster genome ( FlyBase release 5 . 27 [54] , [55] ) using Bowtie [18] and TopHat [19] , and transcript abundances for annotated RNAs were called by Cufflinks [20] . Data from each sample were normalized so that the total expression ( reads per kb of sequence , per million mapped reads; RPKM ) of autosomal genes was constant . Genomic reads were mapped to the D . melanogaster genome ( FlyBase release 5 . 27 ) using maq ( maq . sourceforge . net ) . We found that consensus base and SNP calling algorithm was adversely affected by the high level of polymorphism , especially in the CaS sample , so we exported the base-by-base pileup from maq and developed our own SNP calling algorithm . We designated a position as a CaS-w1 SNP if there were at least 13 reads covering the base in each strain , if the frequency of the most common base in each strain was at least 95% , and if these most frequent bases differed . We also generated a w1-CaS consensus sequence consisting of the reference D . melanogaster bases , except where the sequences of the two strains agreed but differed from the reference . We identified all RNA reads expected to differ between the strains , counted their frequencies in each sample , and partitioned the RPKM values for individual genes into their w1 and CaS components in proportion to the fraction of reads in that sample that mapped to the maternal or paternal chromosome . Upon examination of the data , we became concerned that absence of reads from the paternal X chromosome and the low levels of Sxl in embryo F13 arose from a genotyping error . So for graphs showing single genes , we use an average between F12 and F14A for this time point . We used the strain-specific time series to classify genes as maternal , zygotic , or maternal and zygotic . We clustered ( k-medians ) the 5 , 226 genes with at least 10 reads spanning a w1-CaS polymorphism into 20 groups based on similarity of their inferred abundance of maternally and paternally derived transcripts using Cluster 3 . 0 [56] . We classified each cluster as maternal ( only w1 mRNAs detected with levels declining over time ) , zygotic ( no mRNA at cycle 10 , with both w1 and CaS alleles detected over time ) , or maternal and zygotic ( only w1 mRNAs detected at cycle 10 , with CaS mRNAs appearing over time ) . Because of the absence of paternal alleles for genes on the X chromosome , all assignments were based on data from females only . We classified genes lacking polymorphisms distinguishing the strains by comparing their mRNA abundances from the eight female samples to the average patterns from each of the previously assigned groups . We assigned genes to the group with which their expression pattern was best correlated ( if the correlation coefficient was greater than 0 . 8 ) . All reads have been deposited in the NCBI GEO under the accession number GSE25180 and will be made available at the time of publication . The processed data are available at the journal website ( Dataset S1 ) and at eisenlab . org/dosage . | Variation in gene dose can have profound effects on animal development . Yet every generation , animals must cope with differences in sex chromosome numbers . Drosophila compensate for the difference in X chromosome dosage ( two in females , one in males ) with a mechanism that allows for more transcription of the single X chromosome in males . But this mechanism is not established until over an hour after the embryo begins transcription , during which time a number of important events in development occur such as cellularization and segmentation . Here we use an mRNA sequencing method to characterize gene expression in individual female and male embryos before the onset of the previously characterized dosage compensation system . While we find more transcripts from X chromosomal genes in females , we also find many genes with equal transcript levels in males and females . These results indicate that there is an alternate mechanism to compensate for dosage acting earlier in development , prior to the onset of the previously characterized dosage compensation system . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genetics",
"and",
"genomics/genomics",
"developmental",
"biology/embryology",
"genetics",
"and",
"genomics/gene",
"expression",
"developmental",
"biology/pattern",
"formation",
"genetics",
"and",
"genomics/chromosome",
"biology",
"developmental",
"biology/developmental",
"molecular",
"mechanisms"
] | 2011 | Noncanonical Compensation of Zygotic X Transcription in Early Drosophila melanogaster Development Revealed through Single-Embryo RNA-Seq |
Cone-rod homeobox ( CRX ) protein is a “paired-like” homeodomain transcription factor that is essential for regulating rod and cone photoreceptor transcription . Mutations in human CRX are associated with the dominant retinopathies Retinitis Pigmentosa ( RP ) , Cone-Rod Dystrophy ( CoRD ) and Leber Congenital Amaurosis ( LCA ) , with variable severity . Heterozygous Crx Knock-Out ( KO ) mice ( “+/−” ) have normal vision as adults and fail to model the dominant human disease . To investigate how different mutant CRX proteins produce distinct disease pathologies , we generated two Crx Knock-IN ( K-IN ) mouse models: CrxE168d2 ( “E168d2” ) and CrxR90W ( “R90W” ) . E168d2 mice carry a frameshift mutation in the CRX activation domain , Glu168del2 , which is associated with severe dominant CoRD or LCA in humans . R90W mice carry a substitution mutation in the CRX homeodomain , Arg90Trp , which is associated with dominant mild late-onset CoRD and recessive LCA . As seen in human patients , heterozygous E168d2 ( “E168d2/+” ) but not R90W ( “R90W/+” ) mice show severely impaired retinal function , while mice homozygous for either mutation are blind and undergo rapid photoreceptor degeneration . E168d2/+ mice also display abnormal rod/cone morphology , greater impairment of CRX target gene expression than R90W/+ or +/− mice , and undergo progressive photoreceptor degeneration . Surprisingly , E168d2/+ mice express more mutant CRX protein than wild-type CRX . E168d2neo/+ , a subline of E168d2 with reduced mutant allele expression , displays a much milder retinal phenotype , demonstrating the impact of Crx expression level on disease severity . Both CRX[E168d2] and CRX[R90W] proteins fail to activate transcription in vitro , but CRX[E168d2] interferes more strongly with the function of wild type ( WT ) CRX , supporting an antimorphic mechanism . E168d2 and R90W are mechanistically distinct mouse models for CRX-associated disease that will allow the elucidation of molecular mechanisms and testing of novel therapeutic approaches for different forms of CRX-associated disease .
CRX ( Accession: AAH53672 . 1 ) is an Otd/OTX-like ‘paired’ homeodomain transcription factor that is preferentially expressed in vertebrate rod and cone photoreceptor cells in the retina and pinealocytes in the brain [1] , [2] . CRX plays an essential role in the development and maintenance of functional mammalian rod and cone photoreceptors [3] . Previous studies show that CRX acts as a transcriptional activator [1][4]–[6] by interacting with co-activators , promoting histone acetylation at target gene promoters [7][8] and mediating enhancer/promoter intrachromosomal looping interactions [9] of target photoreceptor genes . Crx encodes a 299 amino acid protein that contains a homeodomain ( HD ) near its N-terminus that is responsible for DNA binding ( Figure 1A ) [1][10] . The HD is followed by glutamine rich ( Gln ) , basic , WSP and OTX-tail motifs . The C terminal region of CRX ( from the basic to the OTX-tail domains ) is required for transactivation activity [4] . CRX interacts with transcription co-regulators including the rod-specific transcription factors NRL ( Accession: NP_006168 . 1 ) [11][12] , NR2E3 ( Accession: AAH41421 . 1 ) [13][14] , and general co-activator proteins GCN5 , CBP and p300 ( Accessions: AAC50641 . 1 , AAC17736 . 1 , NP_001420 . 2 , respectively ) [7] to coordinately control photoreceptor gene expression . In the homozygous Crx Knock-Out mouse ( “−/−” ) , photoreceptors fail to form outer segments ( OS ) , a highly specialized photoreceptor organelle which contains visual pigment opsins and other proteins required for phototransduction [15][16] . As a result , −/− photoreceptors do not function [3] , form abnormal synapses [17] , and undergo progressive degeneration [3] . Gene expression profile studies showed that −/− mice have severely reduced expression of many photoreceptor specific genes [18]–[20] . Most of these genes are direct CRX targets as detected by ChIP-seq analyses of the genomic CRX binding profile in the mouse retina [21] . Mutations in human CRX ( NCBI Reference Sequence: NG_008605 . 1 ) have been associated with autosomal dominant forms of the retinal degenerative diseases Retinitis Pigmentosa ( adRP ) , Cone-Rod Dystrophy ( adCoRD ) and Leber Congenital Amaurosis ( adLCA ) , with different ages of onset and severity [12][22]–[45] . CRX is the only gene associated with all three diseases [22][23][26][43] , demonstrating its central role in rod and cone integrity . However , null mutations in CRX may not be responsible for severe dominant disease . A null mutation in CRX , P9ins1 , was associated with LCA in a heterozygous patient but the patient's father , a carrier of P9ins1 , had a normal ocular phenotype suggesting either recessive or multigenic inheritance [44] . The heterozygous Knock-Out mouse ( “+/−” ) , also shows only a slight delay in photoreceptor development and fails to model severe forms of dominant human disease [3] . The phenotypes of the human heterozygous null mutation and the +/− mouse phenotype suggest that haploinsufficiency is unlikely to underlie the severe forms of dominant CRX-associated disease . Dominant disease-causing human CRX mutations primarily fall into two classes ( Figure 1A ) : frameshift mutations ( blue text ) mostly in the transactivation domains and amino acid substitution mutations ( black text ) mostly within the DNA binding homeodomain . Both classes are expected to produce mutant forms of CRX protein that are pathogenic . Truncated CRX proteins resulting from the frameshift mutations E168d1 , E168d2 , A196d4 and G217d1 lost the ability to transactivate the promoter of Rhodopsin ( Rho ) in HEK293 cell transient transfection assays , but are expected to bind DNA normally since CRX 1–107 , a complete activation domain truncation mutant , retained CRX target binding activity [4] . It was predicted that these truncated mutant proteins could interfere with the function of WT CRX by an antimorphic mechanism and cause a severe dominant retinal phenotype . Supporting this hypothesis , E168d1 , E168d2 , G217d1 and several other truncation mutations were linked to early onset ( 0–20 years ) severe adCoRD/adLCA [22]–[33][36]–[40] and A196d4 was associated with adult onset adCoRD [42] . Furthermore , rescue experiments of the otduvi phenotype in Drosophila demonstrate the CRX truncation mutation I138fs48 possessed dominant-negative activity on target gene expression [46] , providing experimental evidence for an antimorphic mechanism for this class of CRX mutations . Four substitution mutations in the homeodomain: R41W , R41Q , R90W [11][45][47] , and K88N [12] , also reduced the ability of CRX to bind to and transactivate the Rhodopsin promoter . R41Q and R90W both reduced CRX:NRL protein interaction [11] , while K88N additionally interfered with basal NRL-mediated transcription [12] . R41W , R41Q , and R90W were predicted to represent hypomorphic alleles associated with either recessive or less severe dominant forms of disease , while K88N was predicted to possess antimorphic activity on NRL function causing a stronger phenotype . Supporting this hypothesis , R41W , R41Q , R90W and several other substitution mutations were associated with late onset ( ∼40–60 years old ) adCoRD [22][23][33][36][42][45] , while K88N was associated with adLCA [12] . A patient homozygous for R90W was also diagnosed with autosomal recessive LCA [45] . In contrast , four other substitution mutations E80A [22][23][33][39] , A56T [31] , A158T and V242M [42] did not lose DNA binding or transactivating activity [47] and were associated with early onset adCoRD/LCA . In vivo rescue experiments in Drosophila also demonstrate that E80A but not R90W or K88N possesses some dominant-negative activity on Rh5 expression [46] . Collectively these experiments support our hypothesis that substitution mutations may cause disease through several distinct mechanisms . Currently , there is no treatment strategy for CRX-associated diseases . Establishing animal models that accurately recapitulate different disease mechanisms is critical for developing and testing novel therapeutic approaches . Here we report the generation of two mechanistically distinct Knock-IN ( K-IN ) mouse models , each carrying a different class of CRX mutation , and present a detailed morphological , functional and biochemical characterization of these mouse models . The frameshift mutation E168d2 produces a severe dominant phenotype through an antimorphic mechanism , while the substitution mutation R90W produces a very mild late-onset ‘CoRD-like’ phenotype in heterozygotes and ‘LCA’-like disease in homozygotes . Furthermore , the expression level of a mutant allele can dramatically affect the disease phenotype , providing insight into potential treatment strategies .
In this study , we have generated two Crx K-IN mouse lines , each carrying a human disease-causing mutation in the mouse allele ( Accession: NM_007770 . 4 ) . CrxE168d2 ( “E168d2” ) mice carry a 2-bp deletion mutation , Glu168del2 , which resulted in a codon frameshift and early truncation of the transactivation domains of CRX protein ( Figure 1A–C ) . CrxR90W ( “R90W” ) mice carry Arg90Trp , an amino acid substitution mutation in the homeodomain of CRX ( Figure 1A–C ) . An intermediate subline of each ( “E168d2neo” and “R90Wneo” ) carrying a neomycin ( neo ) cassette in intron 3–4 was also maintained ( Figure 1B ) , since the neo cassette specifically reduced the expression of the mutant allele ( Figure 2 ) . The neo was removed from the germline by crossing E168d2neo and R90Wneo mice to the Sox2-Cre mouse [48] to generate the final E168d2 and R90W mouse lines ( Figure 1B ) . Successful K-IN was confirmed by PCR amplification of neo ( Primer set: Neo F/R ) and the respective Crx allele ( Table S1 , Figure S1 ) and Sanger sequencing of homozygous mice ( Figure 1C ) . To determine if E168d2 and R90W K-IN mice properly express their respective CRX proteins , immunofluorescence ( IF ) staining for CRX was performed on paraffin-embedded retinal sagittal sections of P10 mice ( Figure 2 ) . The mouse monoclonal CRX antibody M02 ( Abnova ) used recognizes WT ( Accession: NP_031796 . 1 ) and both mutant forms of CRX . Slides were immunostained in the same batch and imaged using a common exposure . As reported previously [13][19][49] , CRX staining in WT retina ( Figure 2A ) was predominantly localized to the outer nuclear layer ( ONL ) , comprised of the rod and cone photoreceptor cell bodies . Less intense CRX staining was also seen in the outer portion of the inner nuclear layer ( INL ) , which is comprised of bipolar and horizontal cell bodies . E168d2 homozygous ( “E168d2/d2” ) and heterozygous ( “E168d2/+” ) mouse retinas showed higher intensity CRX staining than WT , especially in the ONL ( Figure 2B&C ) . The heterozygous E168d2neo ( “E168d2neo/+” ) retina on the other hand showed similar intensity CRX staining as WT retina ( Figure 2D vs 2A ) . In contrast , CRX staining in the ONL of R90W homozygous ( “R90W/W” ) and heterozygous ( “R90W/+” ) mouse retinas was reduced compared to WT retinas , although a few cells expressing high levels of CRX are scattered across the ONL ( Figure 2E&F ) . This mosaic pattern of variable CRX expression was not seen in WT retinas . Crx Knock-Out ( “−/−” ) retinas didn't show CRX reactivity in the ONL and served as negative controls ( Figure 2G ) . The positive CRX staining in E168d2/d2 and R90W/W retinas suggests that the CRX[E168d2] and CRX[R90W] mutant proteins were expressed in the appropriate cell layers . The expression levels of WT CRX and mutant CRX[E168d2] , CRX[R90W] proteins were compared and quantified in P10 E168d2 and R90W K-IN retinas using quantitative Western blots assayed with the polyclonal CRX 119b-1 antibody [7] , which also recognized all forms of CRX proteins assayed . WT retina extracts showed a ∼37 kD band ( Figure 2H , Lane 1 ) . In contrast , a ∼27 kD dublet CRX band was detected in E168d2/d2 ( Lane 2 ) and homozygous E168d2neo ( “E168d2neo/d2neo” ) ( Lane 3 ) retinas , suggesting that the CRX[E168d2] protein was a truncated CRX protein as predicted by Sanger sequencing and genomic alignment ( Figure 1C ) . Furthermore , the band intensities suggest that the amount of CRX[E168d2] protein in mutant retinas is higher than that of the full-length CRX in WT retinas ( Figure 2H , Lanes 2&3 vs . Lane 1 ) . Quantification of CRX protein levels ( Figure 2I ) revealed a significant genotype difference ( p = 0 . 0002 ) overall . E168d2/d2 retinas made twice as much total CRX protein as WT retinas , while E168d2neo/d2neo retinas produce similar amounts of CRX protein as WT retinas . Heterozygous E168d2/+ ( Figure 2H , Lane 4 ) and E168d2neo/+ ( Lane 5 ) mice expressed both full-length WT CRX and truncated CRX[E168d2] protein but in different ratios . Quantification of CRX protein in E168d2/+ retinal extracts ( Figure 2I ) revealed that the full-length WT CRX protein was present at approximately half of the level in WT retinas , but the level of CRX[E168d2] protein was more than twice that of the WT CRX . As a result , the total CRX protein level in these retinas was significantly increased by 2-fold compared to normal retinas . E168d2neo/+ retinal extracts also expressed WT CRX at approximately half WT levels but expressed less CRX[E168d2] protein than E168d2/+ retinas ( Figure 2H , Lane 5 vs . 1&4 , Figure 2I ) . As a result , the total CRX level in E168d2neo/+ was comparable to the WT control levels . These results are consistent with immunostaining results shown in Figure 2B–D and suggest that the E168d2 allele overproduces mutant protein , which was prevented by the presence of the neo cassette in E168d2neo . CRX expression patterns in R90W mice differed from E168d2 . In P10 R90W/W retinal extracts ( Figure 2H , Lane 6; Figure 2I ) , CRX[R90W] was not significantly different from CRX in WT retinal extracts ( Figure 2H , Lane 1; Figure 2I ) , while levels were reduced in R90Wneo/Wneo retinas ( Figure 2H , Lane 7; Figure 2I ) . R90W/+ retinas ( Figure 2H , Lane 8; Figure 2I ) had normal total CRX protein levels compared to WT mice , although it was not possible to distinguish the quantity of WT CRX vs . CRX[R90W] . As seen with the E168d2 allele , the presence of the neo cassette reduced total CRX protein levels in R90Wneo/Wneo and R90Wneo/+ retinas , compared to corresponding R90W retinas ( Figure 2H , Lane 7 vs . 6 , Lane 9 vs . 8; Figure 2I ) . Thus , the presence of the neo cassette similarly affected the expression of both K-IN alleles . To investigate whether the changes observed in CRX protein levels correlate with altered Crx mRNA transcription , Crx mRNA levels were determined by quantitative real-time reverse transcriptase PCR ( qRT-PCR ) ( Figure 2J ) . Specific PCR primer pairs were used that selectively amplified sequences from either WT or total ( WT+mutant ) Crx cDNA ( Primer sets: Crx E168WT F/R and Crx R90WT F/R; Table S1 ) . Primer specificity was validated by amplification of WT , E168d2/d2 and R90W/W retinal cDNA preparations . The results show that E168d2/d2 retinas made twice as much total Crx mRNA as WT retinas , consistent with the elevated CRX protein levels in E168d2/d2 . Total Crx mRNA levels in E168d2neo/d2neo retinas were lower than E168d2/d2 levels ( FDR p = 0 . 07 ) but remained elevated relative to the WT ( p<0 . 05 ) retinas , in contrast to the normal total CRX protein levels observed in these retinas . E168d2/+ mice also showed moderately elevated total Crx mRNA levels ( Figure 2J ) . Similar to protein levels , E168d2 mRNA levels ( deduced from Total - WT ) were much higher than WT levels ( ∼2∶1 ratio ) . By comparison , E168d2neo/+ mice expressed slightly elevated levels of total Crx mRNA that were lower than E168d2/+ . WT and E168d2 alleles were evenly expressed in these retinas . These results are consistent with the differences in CRX protein levels , supporting an RNA-based mechanism for CRX[E168d2] overexpression , which was partially reversed in E168d2neo/+ mice . R90W mice showed a distinct pattern of mRNA expression compared to E168d2 . R90W/W retinas had normal Crx mRNA levels ( Figure 2I ) , in contrast to their reduced CRX protein levels . This suggests a post-transcriptional mechanism either in the production or degradation of CRX[R90W] protein is likely responsible . Crx mRNA levels in R90Wneo/R90Wneo mice were substantially reduced in comparison to WT ( p<0 . 05 ) and R90W/R90W mice ( FDR p = 0 . 07 ) . The R90W/+ and R90Wneo/+ mice showed essentially normal levels of total Crx mRNA , contributed either by both alleles equally ( in R90W/+ ) or the WT allele predominantly ( in R90Wneo/+ ) . Together , our results suggest that E168d2 and R90W mRNA and corresponding proteins are produced in K-IN mouse retinas , but expression levels are differentially regulated . The mechanism of differential expression appears to be determined by features intrinsic to each mutant allele . To determine the effect of E168d2 and R90W mutations on retinal morphology , paraffin embedded retinal sections from E168d2/d2 and R90W/W mice at P14 , 1 month ( mo ) and 3 mo were stained with hematoxylin and eosin ( H&E ) , imaged by light microscopy and compared to sections from WT and −/− mice [3][17] . Cell specification in WT retina is complete by P14 and three distinct neuronal layers are present: the ONL , INL and the ganglion cell layer ( GCL ) ( Figure 3A ) . At P14 E168d2/d2 , R90W/W and −/− retinas all had established normal cellular lamination ( Figure 3B–D ) . Quantitative morphometric measures across the sagittal plane of the retina presented by ‘spider graphs’ ( Figure 3M ) did not show a genotype*distance interaction ( the statistical threshold required to make individual comparisons when analyzing data with two-way ANOVA ) ( p = 0 . 15 ) at P14 . These results support previous finding that CRX is not required for retinal cell fate specification [3] , including rod photoreceptors , which constitute the majority of cells in the ONL . However , unlike WT retinas none of the mutant ONL cells had begun to form OS's at this age ( Figure 3B , C , D vs . A ) . This OS defect persisted through 1–3 mo when OS's were fully formed in WT retina ( Figure 3F , G , H vs . E; J , K , L vs . I ) . By 1 mo , loss of ONL nuclei was evident in all mutant retinas ( Figure 3F–H ) . In comparison to the ∼12 rows of ONL nuclei seen in WT retinas , E168d2/d2 had only ∼3–4 rows , and R90W/W and −/− had ∼7–9 rows ( Figure 3F , G , H vs . E ) . Quantification of ONL thickness shows photoreceptor degeneration occurred evenly across the sagittal plane of all mutant retinas ( Figure 3N , red , green & blue lines vs . black ) . While R90W/W and −/− mice had similarly reduced ONL thickness ( green and blue line , respectively ) , E168d2/d2 retinas showed greater ONL thinning at 1 mo ( red line vs . green & blue ) , suggesting that degeneration was accelerated in these retinas . At 3 mo , all models exhibited greatly reduced ONL thickness ( Figure 3O ) with only ∼2–3 rows of ONL cells remaining ( Figure 3J , K , L vs . I ) , suggesting ONL degeneration is progressive and extensive in all homozygous mutant mice . To determine if ONL thinning is mediated by programmed cell death , “terminal deoxynucleotidyl transferase dUTP nick end labeling” ( TUNEL ) analysis was performed on P21 and P35 sagittal retinal sections ( Figure S2 ) . At P21 ( Figure S2A–E ) , E168d2/d2 , R90W/W and −/− mice all had significantly increased TUNEL+ cells present , almost exclusively in the ONL , E168d2/d2 exhibited the highest number of TUNEL+ cells ( ∼34 fold over WT ) . At P35 ( Figure S2F–J ) , TUNEL+ cells remained elevated in the ONL of all mutant models but E168d2/d2 mice showed fewer TUNEL+ cells compared to R90W/W and −/− mice . There was no increase in TUNEL+ cells in other retinal layers of any of the mutant mice . These timecourse analyses suggest that the peak of ONL degeneration is earlier in E168d2/d2 mice compared to R90W/W and −/− mice , corresponding with the earlier ONL thinning observed in morphometric analyses . To assess the consequence of these morphological changes on retinal function , electroretinograms ( ERG ) were performed under various light intensities on WT , E168d2/d2 and R90W/W mice at 1 month of age [50] . E168d2/d2 and R90W/W mice did not show any detectable dark-adapted or light-adapted responses ( Figure S3 ) . These results suggest E168d2/d2 and R90W/W mice are blind at young ages , similar to the phenotype reported for −/− mice [3] . The functional deficits of rod and cone photoreceptors in E168d2/d2 and R90W/W mice are consistent with the necessity of photoreceptor OS's for phototransduction [15][16] and suggest defective development of photoreceptor function in the homozygous mutant mice , similar to deficits in retinal function in LCA patients . In spite of reduced Crx expression levels , homozygous mice from the sublines of each strain that carry a neo cassette ( E168d2neo/d2neo , R90Wneo/Wneo ) displayed retinal morphology and function ( data not shown ) that was indistinguishable from the respective neo-deleted line . Thus , in homozygous mice lacking WT alleles , the onset and rate of photoreceptor degeneration was not greatly affected by mutant protein expression level . To determine the inheritance of E168d2 and R90W-associated phenotypes , retinal morphology of heterozygous E168d2/+ , E168d2neo/+ and R90W/+ mice was assessed by histology and morphometry . Paraffin embedded sagittal retina sections of heterozygous mutant mice at P14 , 1 mo , 3 mo and 6 mo were stained with H&E , imaged by light microscopy and compared to WT sections ( Figure 4A–P ) . At P14 , all retinas of heterozygous mutant mice displayed normal cellular lamination ( Figure 4B–D vs . A ) . However , morphometric measurements of the ONL thickness showed that E168d2/+ had increased thickness at the two points most proximal to the optic nerve head ( ON ) ( Figure 4Q , colored lines vs . black ) . E168d2/+ retinas also showed shortened rod OS's compared to WT ( Figure 4B vs . A ) . The OS defect in E168d2/+ retinas remained at 1 mo ( Figure 4F vs . E ) , 3 mo ( Figure 4J vs . I ) and 6 mo ( Figure 4N vs . M ) . At 1 mo and 3 mo ( Figure 4E–L , R–S ) , morphometric measurements of ONL thickness did not identify a significant genotype*distance interaction overall , therefore differences at each distance were not tested . However , at 3 mo , E168d2/+ had fewer rows of ONL cells ∼6–8 and had reduced mean ONL thickness at each distance . By 6 mo , most of E168d2/+ ONL cells had degenerated with only ∼2–3 rows of nuclei remaining ( Figure 4N vs . M; Figure 4T , red vs . black line ) . By morphometric analyses , E168d2/+ exhibited reduced ONL thickness at all distances . These results suggest that E168d2/+ retinas undergo progressive rod photoreceptor degeneration through 6 mo of age . Consistent with this observation , TUNEL analysis showed at P35 E168d2/+ mice had 15-fold more TUNEL+ cells than WT all of which were located in the ONL ( Figure S2L vs . K; Figure S2O ) , consistent with the observed photoreceptor degeneration phenotype . These results suggest that the E168d2 mutation causes dominant rod photoreceptor morphological defects and degeneration . To determine if mice expressing lower levels of CRX[E168d2] protein have a less severe retinal phenotype , the morphology of E168d2neo/+ retinas was compared with that of E168d2/+ retinas . At P14 , similar to E168d2/+ ( Figure 4B ) , the OS's of E168d2neo/+ mice appeared shorter than in WT mice ( Figure 4C vs . A ) . However , unlike E168d2/+ , E168d2neo/+ formed fully elongated outer segments by 1 mo ( Figure 4G vs . F ) , which were well maintained at 3 mo ( Figure 4K vs . J ) and 6 mo ( Figure 4O vs . M ) . These results suggest that , despite a delay in maturation , E168d2neo/+ mice had less disrupted rod photoreceptor structure than E168d2/+ . Furthermore , E168d2neo/+ did not show significant thinning of the ONL through 6 mo ( Figure 4S&T , blue vs . black line ) or elevated TUNEL+ cells compared to WT ( Figure S2M vs . K; Figure S2O ) . Overall , the rod photoreceptor phenotype of E168d2neo/+ mice is mild compared to E168d2/+ mice , suggesting that E168d2 disease severity was influenced by the expression level of the mutant allele in heterozygous mice , consistent with E168d2 being an antimorphic mutation . To further reveal morphological defects in E168d2 photoreceptors at the ultra-structural level , transmission electron microscopy ( TEM ) imaging analyses were performed on the retinas of P21 E168d2/+ , E168d2neo/+ and WT mice ( Figure 4U–W; Figure S4 ) . Images were randomly coded for blinded data analysis . Compared to the morphology of WT OS's ( Figure 4U ) , E168d2/+ mice ( Figure 4V ) exhibited severely shortened and disordered OS's including the presence of ‘wave-like’ disc patterns ( white ‘*’s ) , ectopic vesicle formation ( white ‘+’s ) , and improper stacking of OS discs including vertically oriented discs ( white triangles ) . OS morphology was largely normal in E168d2neo/+ mice ( Figure 4W ) ; although minor ‘wave-like’ disc patterns and ectopic vesicle formation were occasionally seen . Rod nuclei in P21 WT retina adopt a characteristic nuclear architecture with large areas of highly electron dense heterochromatin in the center and smaller regions of translucent euchromatin in the nuclear periphery [51] ( Figure S4A&D ) . The chromatin pattern of E168d2/+ rods , however , appeared less condensed than WT ( Figure S4B&E vs . A&D ) . This did not occur in E168d2neo/+ mice ( Figure S4C&F vs . A&D ) . To quantify these changes , the percentage of the nuclear area comprised of condensed heterochromatin was measured in randomly selected WT , E168d2/+ and E168d2neo/+ rod nuclei . Figure S4G shows that the mean area of heterochromatin in E168d2/+ rods was significantly reduced by 8% compared to WT . This reduction in rod heterochromatin territory was not seen in E168d2neo/+ mice , suggesting more normal rod nuclear architecture . In addition , photoreceptor degeneration in E168d2/+ and E168d2neo/+ mice was evidenced by the presence of highly electron dense nuclei corresponding to pyknotic photoreceptor cells undergoing cell death , which were not observed in WT retinas ( Figure S4E&F vs . G , white pentagon ) . Unlike E168d2/+ , R90W/+ mice had normal retinal morphology at all ages ( Figure 4D , H , L&P ) , comparable to +/− mice [3] . They formed and maintained full-length OS's and normal ONL thickness ( Figure 4H , L&P ) through 6 mo of age . No increase in TUNEL+ cells over WT was detected ( Figure S2N&O ) . These results suggest rod photoreceptor development and maintenance are normal in R90W/+ mice . This is consistent with clinical evaluations for heterozygous R90W carriers in human cases [22][23][33][42][45] . To determine how mutant forms of CRX protein affect target gene transcription , we assessed their ability to bind to DNA and transactivate transcription . First , electrophoretic mobility shift assays ( EMSA ) were used to measure DNA binding activity of CRX WT , CRX[E168d2] and CRX[R90W] protein expressed in HEK293 cells on the rhodopsin promoter target site BAT-1 [1] ( Figure 10A ) . To compare relative binding affinity , the amount of CRX in each nuclear extract was quantified using Western blots and equalized between transfections ( Figure 10B ) . EMSA was then performed on a 2-fold dilution series of nuclear extracts of each CRX protein . Following incubation with BAT-1 probe , WT CRX extract produced a single species of specific band shift ( marked as ‘WT’ ) with a concentration-dependent intensity . This shifted band represented specific binding of the indicated CRX protein to BAT-1 CRX sites , as it is absent in the lane receiving the GFP control extract and when the probe contains mutated CRX binding sites ( BAT-1 Mut AB ) . CRX[E168d2] nuclear extract also produced a specific band shift ( marked ‘E168d2’ ) , which migrated much faster than the full-length CRX band as expected for a truncated protein . The intensity of the E168d2 band was comparable to the WT full-length band at each corresponding concentration , suggesting that CRX[E168d2] binds target sites with similar efficiency as WT CRX , providing a basis for competition binding to common targets . In contrast , CRX[R90W] nuclear extract produced a faint band with the same mobility as WT ( Figure 10A ) , but significantly reduced intensity ( ∼69% lower than WT ) . Reduced but not abolished DNA binding activity was also reported for bacterially expressed CRX homeodomain peptides carrying the R90W mutation [47] . These results support the hypothesis that CRX[E168d2] protein maintains normal DNA binding ability , while CRX[R90W] protein has reduced DNA binding ability . To determine if in vitro DNA binding activity of each mutant reflected ability to associate with target chromatin in vivo , the association of WT CRX , CRX[E168d2] and CRX[R90W] protein with target gene promoter regions was examined using chromatin immunoprecipitation ( ChIP ) assays . ChIP was performed on P10 mouse retinas of WT , E168d2/d2 , R90W/W and −/− mice using the CRX 119b-1 antibody [7] . As expected , enrichment of CRX[E168d2] protein was detected on the promoter of genes expressed in rods ( Rho , Gnat1 ) , cones ( Arr3 , Opn1mw , Opn1sw ) and both rods/cones ( Crx , Rbp3 ( Accession: AJ294749 . 1 ) ) ( Figure 10C , red bars ) . Despite reduced DNA binding activity in vitro , CRX[R90W] protein was found on the promoter of all candidate genes tested ( Figure 10C , green bars ) . The mechanism by which CRX[R90W] , which has reduced DNA-binding ability , is recruited to target gene chromatin in vivo remains to be determined . However , these results are consistent with R90W's hypomorphic effect on target gene expression in the retina ( Figure 8 , Figure 9 ) . The ability of CRX[E168d2] and CRX[R90W] proteins to transactivate target promoters , either alone or in combination with WT CRX , was assessed by dual-luciferase reporter assays in transiently transfected HEK293 cells . Consistent with a previous report [47] , WT CRX was able to cooperate with NRL to activate a Rhodopsin promoter-driven luciferase reporter , BR130 ( Figure 10D ) . However , CRX[E168d2] failed to increase transactivation above NRL alone , suggesting that CRX[E168d2] was unable to form functional interactions with transcription co-activators despite its normal DNA binding ability . In contrast , CRX[R90W] weakly promoted NRL-mediated transactivation , consistent with CRX[R90W]'s weak ability to bind target DNA ( Figure 10A ) and interact with NRL [11] in vitro to promote low levels of gene expression in the retinas of homozygous R90W mice ( Figure 8 ) . To test the effect of mutant protein on WT CRX function , E168d2 and R90W expression vectors were each co-transfected at increasing concentrations with WT CRX . CRX[E168d2] protein significantly impaired WT CRX function when the ratio of E168d2:WT vector reached 2∶1 or higher , suggesting CRX[E168d2] actively interfered with WT CRX via an antimorphic mechanism , consistent with the dose-dependent toxicity observed in E168d2/+ and E168d2neo/+ mice . In contrast , at the same mutant:WT vector ratios , CRX[R90W] protein did not disrupt WT protein function , consistent with the hypomorphic effect of R90W in mice . The Crx promoter is another known CRX direct target . It contains two CRX consensus binding sites within a 500-bp upstream region that is required for CRX auto-activation [57] . However , unlike Rhodopsin , which is downregulated , Crx was overexpressed in E168d2 mice ( Figure 2 ) . To determine if Crx overexpression resulted from the direct action of CRX[E168d2] protein on the Crx promoter , dual-luciferase reporter assays using the 0 . 5K Crx promoter were performed ( Figure 10E ) . As expected , WT CRX protein transactivated this in a concentration-dependent manner ( Figure 10E ) , while CRX[E168d2] and CRX[R90W] at the highest concentration did not transactivate . When both WT and mutant proteins were present , CRX[E168d2] interfered with the transactivation activity of WT CRX , even at a 1∶2 mutant:WT vector ratio . CRX[R90W] protein also reduced WT CRX transactivation activity , though less strongly , at the 1∶1 and 2∶1 mutant:WT vector ratios . These results suggest that both CRX[E168d2] and CRX[R90W] proteins 1 ) are less effective than WT CRX at activating target promoters , and 2 ) interfere with WT CRX autoactivation . Taken together , functional analyses of CRX[E168d2] and CRX[R90W] proteins revealed that they affected target gene transcription via distinct mechanisms . While CRX[E168d2] bind DNA equally well as WT CRX , it fails to activate transcription and interferes with WT CRX function , resulting in a dose-dependent antimorphic effect . In contrast , CRX[R90W] has reduced ability to bind target DNA and regulate transcription , qualifying CRX[R90W] as a hypomorphic protein .
Several pieces of evidence support that CRX[E168d2] and CRX[R90W] protein cause disease via different mechanisms , as illustrated in Figure 11 . CRX[E168d2] protein bound to DNA , interfered with the function of CRX WT and impaired the expression of CRX target genes , classifying it as an antimorphic protein with dominant-negative activity ( Figure 11B ) . All of our results suggest that CRX[E168d2]'s activity was largely restricted to CRX target genes . Of the 82 uniquely downregulated genes identified in homozygous E168d2neo mice , most ( 76 . 8% ) also exhibited direct CRX binding . The average fold change of these distinct genes was less dramatic than genes shared between E168d2 and −/− , suggesting they were likely to be similarly affected in −/− but failed to pass the significance threshold . Many shared genes including: Rho , Arr3 , Ramp3 , Drd4 , Cpm , and Pde6c were more strongly downregulated in homozygous E168d2neo than −/− mice ( Figure 8D ) . This suggests CRX[E168d2] protein had an antimorphic effect on the expression of these genes even in the complete absence of WT CRX , possibly by interfering with other co-factors like the homeodomain transcription factor OTX2 . Supporting this hypothesis , removal of one allele of Otx2 from the −/− mouse produced a severe phenotype similar to homozygous E168d2 mice [63] . Since OTX2 and CRX have overlapping spatial and temporal roles in retinal development and share DNA binding domain homology , it is possible that CRX[E168d2] interfered with OTX2 activity , resulting in a stronger phenotype than −/− . This antimorphic effect is unlikely to involve interference with NRL function , since NRL expression was comparable in all homozygous models ( Table S5 ) , CRX[E168d2] did not interfere with NRL transactivation ( Figure 10D ) and a similar truncation mutation in bovine CRX C160 ( 1–160 ) maintained interaction with NRL [11] . qRT-PCR analysis of CRX target gene expression showed downregulation that correlated with mutant CRX expression level ( Figure 9 ) , supporting the conclusion that CRX[E168d2] is an antimorphic mutant protein with dominant negative activity . The E168d2 mouse model thus demonstrates the effects of an antimorphic truncated CRX protein associated with human disease . The CRX[R90W] protein had reduced DNA binding and weakly promoted transcription in vitro , classifying CRX[R90W] as a hypomorphic protein ( Figure 11C&F ) . Although binding of CRX[R90W] to the BAT-1 oligo in vitro was reduced ( Figure 10A ) , CRX[R90W] associated with CRX target DNA in vivo ( Figure 10C ) , suggesting co-factors may anchor CRX[R90W] to target DNA . CRX[R90W] weakly promoted NRL-mediated transactivation of the Rho promoter in vitro ( Figure 10D ) , consistent with early findings that CRX[R90W] protein reduced the physical interaction with NRL [11] . Thus , even though CRX[R90W] was associated with target promoters in vivo , it may have lost specific interactions with co-factors , therefore reducing its function . Indeed , despite being present on target promoters , CRX[R90W] only weakly promoted target gene expression in vivo , as shown by reduced expression of many CRX target genes in homozygous R90W retinas as detected by microarray ( Figure 8A–E ) and qRT-PCR ( Figure 8J–M ) . However , target gene expression in R90W retinas was less reduced compared to −/− retinas ( Figure 8D , J–M ) , suggesting CRX[R90W] possessed some residual transcriptional activation activity . In Drosophila , human R90W was able to partially rescue the otduvi phenotype , consistent with a hypomorphic mechanism [46] . Taken together , our results show that CRX[R90W] is a predominantly hypomorphic mutant CRX protein , representative of substitution mutations associated with mild forms of CRX disease . The molecular functions of several CRX mutations associated with human retinopathy have been investigated in vitro [12][45][47] and in vivo in Drosophila [46] . Such studies indicate that mutant CRX proteins have distinct molecular functions , which could in part explain the variation in CRX-disease phenotypes . The distinct phenotypes of mice carrying E168d2 , an antimorphic frameshift mutation , and R90W , a hypomorphic substitution mutation , further expand our understanding of the impact of mutation type on disease pathology and closely match the functions and associated phenotypes of other similar type mutations . This suggests that E168d2 and R90W K-IN mice are representative animal models for two larger groups of disease causing mutations , increasing their utility as research tools for studying pathology and developing therapies . There are likely additional mechanisms of CRX-associated disease yet to be modeled in vivo , such as substitution mutations that do not affect DNA-binding but are nonetheless associated with dominant disease [12][45][47] . Collectively , these studies demonstrate the diversity of molecular defects mediating CRX-associated disease and highlight the value of having multiple small-animal models to understand them . Currently , there are no treatment strategies for CRX-associated diseases . Since CRX influences many cellular processes , designing targeted therapy is exceptionally difficult . The availability of phenotypically and mechanistically distinct models for CRX-associated disease will greatly improve our ability to develop novel therapies . E168d2 , E168d2neo and R90W present unique mechanistic challenges for therapy to address . Stem cell based therapies have previously been shown to restore function in the −/− mouse [64] . Like −/− mice , E168d2/+ , E168d2/d2 and R90W/W mice all have highly abnormal photoreceptor morphology and undergo rapid degeneration , which may restrict the time course and effectiveness of treatment . The improved phenotype of E168d2neo/+ mice , compared to E168d2/+ , provides evidence that gene replacement strategies that shift the ratio of WT to mutant CRX could be effective at improving vision and promoting rod and cone survival in cases were a mutant protein is toxic and/or overexpressed . Previous studies have shown this strategy to be effective in treating a dominant-negative adRP RHO animal model [65][66] . Lastly , the similarity of the E168d2/+ mouse and the Rdy/+ cat provide excellently matched small and large animal models . Therapies that are proven to be effective in the E168d2/+ mouse can immediately be tested in the Rdy/+ cat , which improves our ability to develop translational therapies . In summary , Crx E168d2 and R90W are mechanistically distinct mouse models for CRX-associated disease , demonstrating how different classes of CRX mutations yield drastically different retinal phenotypes . E168d2 and R90W accurately recapitulate human diseases caused by distinct classes of human mutations and have greatly improved our understanding of disease pathobiology . The availability of these stratified mouse models for CRX-associated disease is an invaluable resource for developing effective mechanism based therapies .
All procedures involving mice were approved by the Animal Studies Committee of Washington University in St . Louis , and performed under Protocols # 20090359 and 20120246 ( to SC ) . Experiments were carried out in strict accordance with recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health ( Bethesda , MD ) , the Washington University Policy on the Use of Animals in Research; and the Guidelines for the Use of Animals in Visual Research of the Association for Research in Ophthalmology and Visual Science ( http://www . arvo . org/animals/ ) . Every effort was made to minimize the animals' suffering , anxiety , and discomfort . Mice were housed in a barrier facility operated and maintained by the Division of Comparative Medicine of Washington University School of Medicine . All mice used for experiments were backcrossed to C57BL6/J mice obtained from Jackson Laboratories ( Bar Harbor , ME , Stock number 000664 ) for at least 5 generations . Knock-IN of E168d2neo and R90Wneo were generated by the Mouse Genetics Core , Department of Ophthalmology and Visual Sciences , Washington University ( Saint Louis , MO ) . E168d2neo and R90Wneo constructs were transfected into 129Sv/J SCC#10 ( ATCC SCRC-1020 ) embryonic stem cells and Knock-IN was achieved by homologous recombination into the endogenous mCrx locus and selected by neomycin . The targeted ES cells were injected into C57BL6/J blastocysts to form chimeric Knock-IN E168d2neo and R90Wneo mice . Germline transmission of E168d2neo and R90Wneo was identified by PCR genotyping and Sanger sequencing of genomic DNA from F1 mice ( Figure 1 , Figure S1 , Table S1 ) . Crx−/− mice were provided by Dr . Constance Cepko , Harvard University ( Boston , MA ) . Genomic DNA was prepared from mouse tail tissue using the Gentra Puregene Tissue Kit ( Qiagen ) . PCR amplification was performed using Jumpstart RedTaq ( Sigma-Aldrich ) . Primer sets ( Table S1 ) are as follows: For all mice: neo ( Neo-F/R ) and Crx ( Total Crx-F/R ) ; for E168d2 colony: WT Crx allele ( E168d2 WT-F , E168d2-R ) , E168d2 allele ( E168d2 Mut-F , E168d2-R ) ; for R90W colony: WT Crx allele ( R90W WT-R , R90W-R ) , R90W allele ( R90W Mut-F , R90W-R ) . Genomic DNA was prepared from mouse tail tissue using the Gentra Puregene Tissue Kit ( Qiagen ) . mCrx DNA was amplified by PCR using the Genomic mCrx Int/Ex4-F/R primer pair ( Table S1 ) . Sanger sequencing was performed by the Protein and Nucleic Acid Chemistry Laboratory ( Washington University , Saint Louis , MO ) using the Sequencing primers E168 and R90W ( Table S1 ) and Big Dye V3 . 1 ( Advanced Biotechnologies ) . At least 5 mice of each genotype were tested for ERG at 1 mo , 3 mo , or 6 mo of age . Bilateral flash ERG measurements were performed using a UTAS-E3000 Visual Electrodiagnostic System running EM for Windows ( LKC Technologies , Inc . , Gaithersburg , MD ) and recordings from the higher amplitude eye were used for analysis . Mice were dark-adapted overnight , then anesthetized with 80 mg/kg ketamine and 15 mg/kg xylazine under dim red illumination for electrode placement and testing . Body temperature was maintained at 37±0 . 5°C with a heating pad controlled by a rectal temperature probe ( FHC Inc . , Bowdoin , ME ) . The mouse's head was positioned just inside the opening of the Ganzfeld dome and pupils were dilated with 1 . 0% atropine sulfate ( Bausch & Lomb , Tampa , FL ) . The recording electrode was a platinum loop 2 . 0 mm in diameter , positioned in a drop of 1 . 25% hydroxypropyl methylcellulose ( GONAK; Akorn Inc . , Buffalo Grove , IL ) on the corneal surface of each eye . The reference needle electrode was inserted under the skin at the vertex of the skull . The ground electrode was inserted under the skin of the mouse's back or tail . The stimulus ( trial ) consisted of a brief , full-field flash ( 10 µs ) either in darkness , or in the presence of dim ( 29 . 2 cd/mm ) background illumination after 10 minutes adaptation time to the background light . The initiation of the flash was taken as time zero . The response was recorded over 250 ms plus 25 ms of pre-trial baseline . Responses from several trials were averaged . For complete test parameters see Table S7 . The log light intensity ( log [cd*s/m2] ) was calculated based on the manufacturer's calibrations . The mean amplitudes ( in microvolts ) of the averaged dark-adapted A and B-waves and light-adapted B-waves were measured and quantified for comparison . The between-group differences in peak amplitude were determined by testing genotype*flash intensity interactions ( p<0 . 05 , n≥5 ) at each age were compared using two-way ANOVA for repeated measurement data to account for potential correlations among readings from the same mice . If the overall genotype*flash intensity interaction was significant , post-hoc multiple comparisons for differences between each genotype and the control group at each light intensity level were performed . All the tests were two-tailed , significance: p<0 . 05 . The statistical analysis was performed using SAS 9 . 3 ( SAS Institutes , Cary , NC ) . p-values were adjusted for multiple comparisons by a permutation test using the default parameters provided in the LSMestimate statement in Proc Mixed . Average percent reductions for each wave form were calculated by normalizing the peak amplitude of the mutant to WT and results were averaged for the flashes listed in Table 1; ±STDEV . For retinal sections: eyes were enucleated by removing the cornea and lens and fixed in 4% paraformaldehyde for 24 hrs at 4°C . A small corneal tag on the superior portion of the eye was used for orientation . Eyes were embedded in paraffin and 5 µM sagittal retinal sections were cut using a Leica RM 2255 microtome as previously described [67] . Hemotoxylin and eosin immunohistochemistry was performed on sections for histology . Fluorescent antibody immunostaining was performed using as previously described using 1% BSA/0 . 1% Triton X in 1× PBS for blocking and antigen retrieval for all samples [13][67] . For whole flat-mounted retinas: eyes were enucleated by removing the cornea and lens and fixed in 4% paraformaldehyde for 1 hr at 4°C . Retinas were then dissected from the eye cup and 4 evenly spaced relief lines were cut ( Figure 6A ) . A scleral tag was left on the superior retina for orientation . Retinas were mounted on poly-D lysine coated slides ( Thermo Scientific ) , blocked with 1% BSA/0 . 1% Triton X in 1× PBS and immunostained as previous . Primary antibodies and dilutions used as follows: Mouse monoclonal anti-CRX M02 ( 1∶200 , Abova ) , rabbit anti-CRX 261 ( 1∶200 ) , rabbit anti-cone arrestin ( CARR ) ( 1∶1000 , Millipore ) , Rabbit anti-Opsin Red/Green ( MOP ) ( 1∶1000 , Millipore ) , Goat anti-OPN1SW ( N-20 ) ( SOP ) ( 1∶500 , Santa Cruz ) , Mouse anti-Rhodopsin RET-P1 ( RHO ) ( 1∶400 , Chemicon ) , Peanut Agglutanin conjugated to Rhodamine ( PNA ) ( 1∶500 , Vector Labs ) . Secondary Antibodies ( 1∶400 ) : Goat anti-rabbit or mouse IgG antibodies coupled to Alexa Fluor A488 , Rhodamine 568 or Cy2 647 ( Molecular Probes ) and Chicken anti-goat IgG ( Molecular Probes ) . All slides were counterstained with hard set DAPI ( Vectashield ) , except when using Cy2 secondary , which were counterstained with Slow Fade Gold DAPI ( Invitrogen ) . All brightfield and fluorescent imaging was performed using an Olympus BX51 microscope and Spot RT3 Cooled Color Digital camera ( Diagnostic instruments inc . ) . TUNEL analysis was performed using the Apoptag Fluorescein in situ Apoptosis Detection Kit ( Millipore ) per kit instructions . TUNEL+ cells were counted in retinal sagittal sections of P21 and P35 mice . Significant differences from WT control ( p<0 . 05 ) were determined by the Kruskal-Wallis rank order test , which was used to protect against departures from the normal distribution assumption . For ONL morphometry , 20× retinal composites of hematoxylin and eosin ( H&E ) stained sagittal sections were analyzed using Image J software ( http://rsb . info . nih . gov/ij/ ) . The distance from the Optic Nerve ( ON ) was determined by drawing a curved line along the outer limiting membrane . The ONL thickness was measured at 100 µM , 500 µM , 1000 µM , and 1500 µM from the ON and 200 µM from the peripheral edge on both the superior and inferior retina . Results are presented by ‘spider graph’ . The between-group differences in ONL thickness were determined by testing overall genotype*distance interactions ( p<0 . 05 , n≥3 ) at each age were tested using two-way ANOVA for repeated measurement data , followed by a post-hoc test to adjust p-value for multiple comparisons between each genotype and the WT control group at each distance using SAS 9 . 3 ( SAS Institutes , Cary , NC ) , as above . Cone nuclear localization was determined by immunostaining retinal sections with CARR . The ONL was divided into 3 equally sized zones ( OONL , MONL , IONL; Figure 5A ) on 20× retinal composite images using Image J software ( http://rsb . info . nih . gov/ij/ ) and the cone nuclei within in each zone from three sections for each mouse were counted . Significant differences from WT for each zone were determined by Kruskal-Wallis rank order test ( p<0 . 05 , n≥3 ) For cone density and opsin expression assessment , 10 images at 40× magnification of whole flat-mounted retinas were taken in the zones specified in Figure 6A . All peripheral images were taken ∼400 µM from the edge of the retina and the central image was taken ∼250 µM from the ON along the lateral axis . Cones were counted within a 200×200 µM square grid for each image using Image J software and the density of cones/ ( mm2*1000 ) was calculated . The between-group differences in cone density were determined by testing overall genotype*retinal region interactions ( p<0 . 05 , n≥3 ) at each age were tested using two-way ANOVA for repeated measurement data , followed by a post-hoc test to adjust p-value for multiple comparisons between each genotype and the WT control group in each retinal region using SAS 9 . 3 ( SAS Institutes , Cary , NC ) , as above . For regional cone opsin expression analysis ( Figure 6E–P ) , differences in the fraction of cones expressing SOP , MOP , SOP/MOP or no opsin was tested in each region using a Kruskal-Wallis rank order test ( p<0 . 05 ) . For TEM studies , eyes were enucleated by removing the cornea and lens and fixed in 2% paraformaldehyde/3% gluteraldehyde in 0 . 1 M phosphate buffer ( pH 7 . 35 ) for 24 hrs , post-fixed in 1% osmium tetroxide for 1 hr and stained en bloc with 1% uranyl acetate in 0 . 1 M acetate buffer for 1 hr . Blocks were then dehydrated in a graded series of acetones and embedded in Araldite 6005/EMbed 812 resin ( Electron Microscopy Sciences ) . Semi-thin sections ( 0 . 5–1 µm ) were cut through the entire retina at the level of the optic nerve and stained with toluidine blue , post-stained with uranyl acetate and lead citrate , viewed on a Hitachi H7500 electron microscope and documented in digital images . Three retinas for each genotype were sampled at P21 at 800–1200 µM from the optic nerve . ≥10 images of four key features were collected by random sampling: OS-RPE ( 10000× ) , OS-IS ( 12000× ) , ONL ( 5000× ) , OPL ( 10000× ) . Images were analyzed in a blinded manner using Image J software . The nuclear percent area of heterochromatin was measured using Image J software in a randomized blinded analysis . For each genotype , 10 5000× images of the ONL were taken for three mouse retinas . For each image , 10 rod nuclei were randomly selected for analysis . The rod nucleus was outlined using the segmented polygon tool , electron dense regions of the nuclei associated with heterochromatin were thresholded and the percentage of the area above the threshold was measured . Thresholding was variably adjusted to accommodate for differences in brightness and contrast . The between-group differences were compared using one-way ANOVA for repeated measurement data , to account for potential correlations among photos from the same mouse . All the tests were two-tailed , significance: p<0 . 05 ( n = 3 ) . The statistical analysis was performed using SAS 9 . 3 ( SAS Institutes , Cary , NC ) . The overall test for genotype difference was statistically significant ( p = 0 . 02 ) , therefore E168d2/+ and E168d2neo/+ were compared to WT ( Figure S2 ) . HEK293 cells ( ATCC CRL-11268 ) were cultured on 60 mm plates in Dulbucco's minimum essential media ( DMEM ) with 10% fetal bovine serum and Penicillin/Streptomycin . Cells in 60% confluence were transfected with pCAGIG-NRL and pCAGIG-hCRX WT , E168d2 and R90W either alone or in combination using CaCl ( 0 . 25 M ) and Boric Acid Buffered Saline ( 1× ) pH 6 . 75 as previously described [13] . Cells were harvested 48 hours post transfection for either RNA ( PerfectPure RNA tissue kit , 5Prime ) , protein ( NePER nuclear and cytoplasmic extractions reagents , Thermo Scientific ) , or Dual-luciferase assays . Dual-luciferase assays were performed as previously described [13] . Significant differences from pcDNA3 . 1hisc control were determined by Kruskall-Wallis rank order test ( p<0 . 05; n = 3 ) . Post-hoc comparisons ( Figure 10 D&E; indicated by brackets ) were tested using a less conservative FDR p-value method for multiple comparisons using PROC Multtest of SAS ( V9 . 3 ) . FDR p<0 . 09 was considered marginally significant . Whole retina protein lysates were prepared by homogenization of four genotype-matched isolated whole retinas from P10 mice and lysis in 1× RIPA buffer ( Sigma ) for 10 min with protease inhibitors ( Aprotinin , Leupeptin , peptistatin , 0 . 1 mM Phenylmethaneslfonyl fluoride ) . Nuclear lysates were prepared using NE-PER Nuclear and Cytoplasmic Extraction Reagants ( Thermo Scientific ) with protease inhibitors . Either 30 µg of whole protein lysate or 5 µg of nuclear protein lysate was boiled for 10 min . Samples were run on a 4–11% SDS-PAGE gel and transferred onto Transblot Turbo nitrocellulose membranes ( Bio-Rad ) using the Transblot Turbo system ( Bio-Rad ) . Membranes were probed with Rabbit anti-CRX 119b1 ( 1∶750 ) and Mouse anti-β-Actin ( Sigma ) ( 1∶1000 ) . Goat anti-Mouse IRDye 680LT and Goat anti-Rabbit IRDye 800CW ( LI-COR ) were used as secondary antibodies . Signal was detected and quantified using the Odyssey Infrared Imager ( LI-COR ) and associated manufactory software . Kruskal-Wallis rank order test ( Proc Npar1way of SAS , V9 . 3 ) was used to test for an overall difference among genotypes ( p = 0 . 0002 ) , then each genotype was compared to WT control ( p<0 . 05 ) . Post-hoc analyses were performeded using FDR p methods for multiple comparisons using PROC Multtest of SAS ( V9 . 3 ) ( FDR p<0 . 09 ) ( n≥3 ) . RNA was extracted from whole retinas of one male and one female mouse at either P10 or P21 for each biological replicate using the PerfectPure RNA tissue kit ( 5Prime ) . RNA was quantified using a NanoDrop ND-1000 spectrophotometer ( NanoDrop Technologies , Wilmington , DE ) . cDNA was synthesized from 1 µg of RNA using the Transcriptor First Strand cDNA Synthesis kit ( Roche Applied Science ) . A 10 µl QRT-PCR reaction mixture containing 1× EvaGreen with Low Rox reaction mix ( BioRad ) , 1 µM primer mix , and diluted cDNA was prepared and run on a two-step 40 cycle amplification protocol with melt curve determination on a BioRad CFX thermocycler in triplicate . The Cq's of technical replicates were averaged and the results were analyzed using the Delta Cq method in QBase software ( Biogazelle ) . Primer sets ( Table S1 ) were designed using MacVector software and synthesized by IDT DNA technologies . For mCrx allele specific amplification the following primers were used: for E168d2 and E168d2neo mice: WT allele specific- Crx E168d2 WT RTF/R , total- Crx R90W WT-RTF/R; for R90W and R90Wneo mice: WT allele specific- Crx R90W WT-RTF/R , total Crx E168d2 WT RTF/R ( Figure 2J ) , Relative gene expression was normalized to Ubb and Tuba1b . Kruskal-Wallis rank order test ( Proc Npar1way of SAS , V9 . 3 ) was used to test for an overall difference among genotypes ( p<0 . 05; n≥3 ) . Post hoc analyses were adjusted for multiple comparisons using FDR p methods , as above ( FDR p≤0 . 09 ) . Triplicate RNA samples were prepared from 4 pooled retinas from 1 male and 1 female mouse at P10 for WT and homozygous E168d2neo , R90Wneo and −/− mice . The RNA was fluorescent labeled and hybridized to MouseWG-6 v2 . 0 Expression Beadchips ( Illumina ) by Washington University Genome Technology Access Center ( GTAC ) . The raw microarray datasets are available at the NCBI GRO website ( http://www . ncbi . nlm . nih . gov/gds , access number: GSE51184 ) . Microarray data were analyzed using significance analysis of microarrays ( SAM ) following background subtraction and quantile normalization in Illumina Genome Studio platform . Control probes and probes with detection p-value <0 . 05 across all samples were removed prior to any analysis . Candidate probes with 2 . 0-fold disregulation at false discovery rate ≤0 . 05 from each comparison were chosen for further analysis . Cellular processes associated with differentially expressed genes were assigned based on gene ontology provided by Mouse Genome Informatics ( http://www . informatics . jax . org/ ) . BAT-1 and BAT-1 mutated AB probes 5′ end-labeled with 700 IRDye were synthesized by Integrated DNA Technologies ( IDT ) . Nuclear protein extracts from HEK293 cells ( ∼1×108 cells ) transfected with pCAGIG-hCRX , pCAGIG-hCRX E168d2 , or pCAGIG-hCRX R90W were prepared following NE-PER kit instructions ( Thermo Scientific ) . Nuclear extracts were tested for CRX expression by running on a Western Blot as above ( Figure 10B ) . CRX levels were quantified by normalizing to β-Actin ( Sigma ) and a 2-fold dilution series of equivalent amounts of CRX WT , CRX[E168d2] and CRX[R90W] protein were used for binding reactions . Binding reactions were performed using the Odyssey Infrared EMSA kit ( LI-COR ) , per kit instructions using 1 µg of nuclear protein extract and 50 nM IRDye labeled oligo . Samples were run on a native 5% polyacrylamide; 0 . 5× Tris/Borate/Ethylenediaminetetraacetic acid ( EDTA ) buffered gel and imaged on the Odyssey Infrared Imager ( LI-COR ) . ChIP was performed as previously described [7][13][68] . Basically , 6 retinas per sample were dissected and chromatin was cross-linked with 1% formaldehyde in PBS for one minute at room temperature . After cell lysis and chromatin fragmentation by sonication , chromatin fragments were immunoprecipitated with the CRX 119b-1 antibody [7] or normal rabbit IgG ( Santa Cruz ) bound to Protein A beads ( GE Healthcare Life Sciences , Piscataway , NJ ) . After extensive washing , the immunoprecipitated chromatin was eluted with 50 mM NaHCO3 1% SDS , heated to 67°C to reverse the cross-links , the DNA purified by ethanol precipitation and analyzed by PCR with gene-specific primers ( Table S1 ) ( n≥3 ) . Fold enrichment was determined by quantitative ChIP PCR . Critical threshold ( Ct ) values for CRX and IgG immunoprecipitation ( IP ) were normalized to input and mock subtracted . The fold enrichment of CRX:IgG was calculated based on the formula shown below . Significant enrichment was determined by testing overall promoter*genotype interactions by two-way ANOVA for repeated measures using SAS 9 . 3 ( SAS Institutes , Cary , NC ) ( p<0 . 05 , n = 3 ) , as above . ΔCt = ( Ct[CRX or IgG]-Ct[Input] ) ΔΔCt = ΔCt[CRX or IgG]- ΔCt[mock] Fold enrichment = ( ( 2−ΔΔCt CRX ) / ( 2−ΔΔCt IgG ) | The transcription factor Cone-Rod Homeobox ( CRX ) plays a central role in regulating gene expression of rod and cone photoreceptors , the primary light sensing cells of the retina . Mutations in the human CRX gene have been associated with the retinal degeneration diseases Retinitis Pigmentosa ( RP ) , Cone-Rod Dystrophy ( CoRD ) and Leber Congential Amaurosis ( LCA ) . These diseases cause progressive and permanent loss of vision , vary widely in age of onset and severity , and are currently untreatable . To understand how mutations in CRX cause distinct forms of retinal disease , we have genetically engineered mice to carry human disease-causing mutations in their Crx gene . These mouse lines accurately recapitulate distinct forms of CRX-associated disease , demonstrating that different classes of CRX mutations are responsible for phenotype variability in humans . We have characterized the pathology of these mice and identified critical mechanisms of disease . In addition , we have discovered that modifying the level of mutant protein had a dramatic effect on disease pathology in one mutant model , suggesting that targeted therapy against the mutant CRX could be an effective treatment strategy . These mouse models will allow for the testing of novel therapeutic strategies for retinal diseases caused by CRX mutations . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"genome",
"expression",
"analysis",
"neurobiology",
"of",
"disease",
"and",
"regeneration",
"animal",
"genetics",
"genetic",
"mutation",
"gene",
"regulation",
"neuroscience",
"dna",
"transcription",
"gene",
"function",
"animal",
"models",
"retinal",
"disorders",
"model",
"organisms",
"molecular",
"genetics",
"gene",
"expression",
"biology",
"mouse",
"visual",
"system",
"inherited",
"eye",
"disorders",
"ophthalmology",
"genetics",
"sensory",
"systems",
"genomics",
"molecular",
"cell",
"biology",
"genetics",
"of",
"disease"
] | 2014 | Mechanistically Distinct Mouse Models for CRX-Associated Retinopathy |
SARS-coronavirus ( CoV ) is a zoonotic agent derived from rhinolophid bats , in which a plethora of SARS-related , conspecific viral lineages exist . Whereas the variability of virulence among reservoir-borne viruses is unknown , it is generally assumed that the emergence of epidemic viruses from animal reservoirs requires human adaptation . To understand the influence of a viral factor in relation to interspecies spillover , we studied the papain-like protease ( PLP ) of SARS-CoV . This key enzyme drives the early stages of infection as it cleaves the viral polyprotein , deubiquitinates viral and cellular proteins , and antagonizes the interferon ( IFN ) response . We identified a bat SARS-CoV PLP , which shared 86% amino acid identity with SARS-CoV PLP , and used reverse genetics to insert it into the SARS-CoV genome . The resulting virus replicated like SARS-CoV in Vero cells but was suppressed in IFN competent MA-104 ( 3 . 7-fold ) , Calu-3 ( 2 . 6-fold ) and human airway epithelial cells ( 10 . 3-fold ) . Using ectopically-expressed PLP variants as well as full SARS-CoV infectious clones chimerized for PLP , we found that a protease-independent , anti-IFN function exists in SARS-CoV , but not in a SARS-related , bat-borne virus . This PLP-mediated anti-IFN difference was seen in primate , human as well as bat cells , thus independent of the host context . The results of this study revealed that coronavirus PLP confers a variable virulence trait among members of the species SARS-CoV , and that a SARS-CoV lineage with virulent PLPs may have pre-existed in the reservoir before onset of the epidemic .
The coronaviruses ( CoV , family Coronaviridae ) are among the most relevant groups of viruses with zoonotic potential . CoVs are large , positive-sense , single-stranded RNA viruses that cause acute and prolonged infections in a variety of mammals and birds . Pathogenic human CoVs include members of the genus Alphacoronavirus , termed human coronavirus ( HCoV ) -NL63 and HCoV-229E , as well as members of the genus Betacoronavirus , termed HCoV-OC43 and HCoV-HKU1 . These endemic viruses cause upper and lower respiratory tract infections in humans worldwide . Past zoonotic descent can be inferred for HCoV-OC43 and -229E , respectively [1–4] . Actual zoonotic acquisition is known for two betacoronaviruses that both cause severe lung disease in humans . The Middle East respiratory syndrome coronavirus ( MERS-CoV ) is a zoonotic agent that is frequently and repeatedly acquired by humans upon contact with dromedary camels in the Arabian Peninsula and parts of Africa [5 , 6] . This virus seems to cause only limited human-to-human transmission , but is considered a major threat to global public health due to recurring nosocomial outbreaks that may facilitate onward adaptation to humans [7–9] . The severe acute respiratory syndrome ( SARS ) -CoV caused an epidemic with sustained human-to-human transmission during 2002 to 2003 in China and other countries , involving more than 8 , 000 notified infections with a case fatality proportion of about 10% [10 , 11] . Human SARS cases were derived from at least two independent zoonotic transmissions from a putative intermediary reservoir in feral carnivores . The majority of cases were part of one continuous chain of human-to-human transmission [12 , 13] . Bats are now known to harbor SARS-CoV strains that can directly infect primate cells [14–18] . SARS-CoV has become a paradigmatic subject to study pre-pandemic processes , both on an ecological and molecular level [14 , 19–21] . Due to the zoonotic nature of many CoVs , there is an increasing interest to understand functional differences in the relative ways in which viruses deal with host cell defenses across different host species , particularly the innate immune response mediated by type I interferons [22] . The type I interferon ( IFN ) response is an effective antiviral barrier that may limit zoonotic cross-host infection in general terms [23 , 24] . CoV infection is sensed by melanoma differentiation antigen 5 ( MDA5 ) and signaled via mitochondrial antiviral-signaling protein ( MAVS ) , stimulator of IFN genes ( STING ) , and IFN regulatory factor 3 ( IRF-3 ) , eventually leading to type I IFN gene transcription [25 , 26] . Several IFN antagonist functions of zoonotic SARS-CoV are known . Within the viral structural and accessory proteins , proteins 3b , 6 , as well as the nucleocapsid protein have been demonstrated to antagonize type I IFN ( e . g . , [27] ) . However , these viral proteins are expressed only after polyprotein processing , nonstructural gene expression , and subgenomic RNA transcription and , therefore , antagonize the downstream effects of IFN receptor signaling rather than the induction of IFN . Distinct from the above SARS-CoV IFN antagonists , the papain-like protease ( PLP ) is an IFN antagonist that constitutes a domain of the replicase polyprotein and , therefore , may be active at an early stage of the replication cycle to antagonize an upstream step of IFN induction . Additionally , unlike accessory proteins , which can vary greatly between CoV species , maintenance of PLP catalytic activity is critical to viral replication and is therefore conserved across all CoVs [28 , 29] . The coronavirus PLP proteins are multifunctional and encode a catalytic triad domain that catalyzes site-specific peptide cleavage of the viral polyprotein and the removal of both ubiquitin and IFN stimulated gene ( ISG ) 15 post-translational modifications . PLP protease activity catalyzes the processing of the replicase polyprotein at cleavage sites between nsp1/nsp2 , nsp2/nsp3 , and nsp3/nsp4 . PLP deubiquitinating ( DUB ) activity has been demonstrated in several CoV species , and acts directly and indirectly on several signal molecules in the IRF-3-dependent IFN induction pathway including retinoic acid inducible gene-I ( RIG-I ) , tumor necrosis factor receptor-associated factor 3 ( TRAF3 ) , TANK-binding kinase 1 ( TBK1 ) and STING [30 , 31] . K63-linked polyubiquitin chains play a general role in signaling cascades of the proinflammatory and IFN systems . K48-linked polyubiquitins label proteins for degradation by the proteasome and activate proinflammatory and antiviral factors . For instance , NFκB is activated by proteasomal degradation of its inhibiting factor IκB [32] . PLP DUB function also involves deISGylating activity , causing the removal of ISG15 modifications from viral and host proteins [33 , 34] . ISG15 is an IFN-inducible , antiviral protein that structurally resembles K48-linked di-ubiquitins . It can also be deconjugated by PLPs of other RNA viruses , in particular the ovary tumor ( OTU ) domain in the papain-like protease 2 ( PLP2 ) of arteriviruses , and the L-gene-encoded OTU domain of nairoviruses [35 , 36] . The protease recognition sequence LXGG is common to cleavage sites in the viral protein as well as ubiquitin and its derivatives . The DUB- and deISGylating activities in CoV PLPs should therefore be widely conserved . Due to the importance of ubiquitin-based innate immune functions , PLP functions may constitute a relevant predictor of the capability of reservoir-borne CoVs to overcome species barriers . PLP activity profiles may differ between relevant zoonotic CoV species . For instance , SARS-CoV has better ability to deconjugate K48- , as opposed to K63-linked polyubiquitins , whereas these activities are balanced in MERS-CoV [37–39] . The processing of ISG15 and K48-linked di-ubiquitin is more effective for SARS- than MERS-CoV PLP [37] . The blocking of induction of IFN by DUB activity was also confirmed for MERS-CoV , but in contrast to SARS-CoV , this inhibition is not independent of PLP’s protease activity [40 , 41] . Interestingly , MERS-CoV is more sensitive to the effects of IFN than SARS-CoV [42 , 43] . Unfortunately , the PLP activity profile cannot be derived from phylogenetic relatedness . For instance , above-mentioned studies found the distantly related HCoV-NL63 and SARS-CoV to be similar in essential features such as protease-independent , DUB-mediated IFN antagonism , while the MERS-CoV that is much closer related to SARS-CoV only inhibits IFN induction when the protease function is intact ( e . g . , [34 , 37] ) . Direct studies of PLP functions of reservoir-borne viruses are therefore necessary to help us determine if there are differences in intrinsic virulence or virus-host interactions . In view of the complexity of PLPs interactions with innate immunity , functional studies have to take the whole viral replication cycle into account . To date , the DUB functions of SARS-CoV PLP have not been studied in the context of a replicating virus . Moreover , no studies have so far compared PLP functions between members of one same viral species including natural variants existing in the zoonotic reservoir . Differences between reservoir-borne and epidemic viruses may uncover mechanisms that aid viral emergence of potentially pandemic strains . Based on epidemic and reservoir-borne variants of the species SARS-CoV , here we exemplify functional differences in PLP domains . By reverse genetics , we show that the PLP of the epidemic SARS-CoV has an enhanced IFN antagonist function that is independent of PLP protease activity , and that is not present in the PLP of a bat-associated SARS-CoV . Additional mutagenesis studies in replicating virus context associate the functional difference to a more efficient binding of ubiquitin or ubiquitin-like modifiers . Against the general assumption that reservoir-associated viruses are highly adapted to their hosts , we find the PLP of the human epidemic virus to counteract IFN better than the bat-derived PLP even in bat cells . PLP function is a viral virulence trait that varies among reservoir-borne viruses .
An amino acid sequence alignment of the PLP region shows obvious similarities between SA-PLP and SR-PLP , and less so between these PLPs and SO-PLP . The PLP core domains in SA-PLP and SR-PLP each comprise 315 amino acids , and in SO-PLP 320 amino acids . SA-PLP and SR-PLP are 86% ( 271/315 amino acids ) identical . SO-PLP share 39% ( 125/324 positions including insertions/deletions ) and 36% ( 118/324 positions including insertions/deletions ) identical amino acids with SA- and SR-PLP , respectively ( Table 1 ) . A catalytic triad consisting of the three residues cysteine C1651 , histidine H1812 and aspartic acid D1826 was previously shown to be responsible for cleavage of the SARS-CoV replicase polyprotein , and is present in SR-PLP ( Fig 1B , grey arrows ) [46] . In SO-PLP the aspartic acids ( D1826 ) are replaced by an asparagine ( N ) . This alternative type of catalytic domain was previously described for other cysteine proteases [47] . Another indispensable feature of SA-PLP is the zinc-binding domain , comprised of four cysteine residues , which connect the left- and right-hand domains of the papain-like fold by a zinc atom [48] . All these residues are also present within SR- and SO-PLP amino acid sequences ( Fig 1B , marked with asterisks ) . To functionally compare the PLPs , protease activities were assessed by a trans-cleavage assay [49] . The assay was based on coexpression of a SARS-CoV nsp2/3-GFP substrate with the respective PLPs , testing the cleavage of substrate into truncated products nsp2 and nsp3-GFP [49] . During establishment of the assay we noticed a considerable level of mRNA splicing while expressing SR-PLP under the control of a chicken β-actin promoter ( S1 Fig ) . Codon-optimized constructs were therefore generated . The truncated nsp3-GFP product was detectable by Western blot using anti-GFP epitope tag antibodies with all PLPs ( Fig 2A; lanes 2 , 4 and 6 ) . To confirm that the protease activity of SR-PLP and SO-PLP depended on the same typical catalytic domain as in SA-PLP , the catalytic cysteines of each PLP ( C1651 ) were changed to alanines . For SA-PLP this mutation was previously shown to abolish PLP activity [46] . The mutants are henceforth referred to as CA-mutants . As expected , each PLP CA-mutant was unable to process the SARS nsp2/3-GFP substrate ( Fig 2A; lanes 3 , 5 and 7 ) . To confirm PLP expression in all cases , Western blots were done on the same cell lysates using antibodies directed against the FLAG epitope tag fused to PLPs . It was found that the expression levels of all PLPs were equal ( Fig 2A , lower panel ) . To enable a quantitative comparison of protease activities , a PLP biosensor luciferase assay was done in which a split Firefly luciferase is coexpressed together with the PLP of interest . Upon cleavage of an LXGG protease cleavage site in the split luciferase construct , luciferase activity is reconstituted and measured after equilibration with cell membrane-penetrating luciferase substrate [50] . A time-course experiment confirmed that all PLPs had similar protease activities ranging between 1- to 6-fold within 2 to 6 h when compared to CA-mutants corresponding to each PLP ( Fig 2B ) . To investigate if the amino acid differences between the PLPs affect the stereostructure of the catalytic site , a small molecule competitive inhibitor known to be specific for the PLP catalytic site , named 3e [51] , was tested side-by-side on the PLPs . Inhibition of protease activity by the inhibitor was successful for both SA-PLP and SR-PLP , indicating structural similarity of both catalytic sites . The inhibitor was slightly more efficient for SA-PLP ( EC50 = 26 . 20 μM ) than for SR-PLP ( EC50 = 30 . 28 μM ) which is plausible because the inhibitor was designed to target a beta-loop structure ( BL2 ) of SA-PLP located close to the catalytic site of the protease ( Fig 1B ) . The inhibitor had very low efficiency towards SO-PLP , whose inhibitor-binding site has only 46% amino acid identity ( 6/13 amino acids identical to SA-PLP ) to the inhibitor target site in SA-PLP ( Fig 2C ) . To compare DUB activities , HEK-293T cells were cotransfected with increasing doses of plasmids encoding each PLP along with constant doses of plasmids encoding HA-tagged ubiquitin . The decrease of protein ubiquitination conferred by PLP was determined by Western blot using anti-HA antibodies . Each PLP deconjugated ubiquitin in a dose-dependent manner suggesting that the PLPs have comparable DUB efficiencies ( Fig 2D ) . Because it is known that SA-PLP has deISGylating activity [34] the efficiency to deconjugate ISG15 from cellular proteins was determined by deISGylation assay . The PLPs were coexpressed with myc-tagged ISG15 in HEK-293T cells . The extent of deISGylated proteins was determined by Western blot using anti-myc antibodies . Each PLP was highly efficient in deconjugating ISG15 from the cellular proteins ( Fig 2E ) . Taken together , these results suggested comparable levels of protease activity and identical DUB and deISGylating activities of SA- and SR-PLP when overexpressed in a human cell context . Differential efficiency towards a PLP inhibitor hint at structural differences even between PLPs from conspecific viruses that occur in zoonotic reservoirs . However , complete failure of the inhibitor was only seen with SO-PLP that falls outside the limits of current CoV species classification . To quantitatively compare the functions of the two closely related PLPs ( SA- and SR-PLP ) in the context of the full virus replication cycle , we constructed a chimeric SARS-CoV in which SA-PLP was replaced by SR-PLP ( Fig 3A ) . Viral plaques with similar morphologies were observed for the recombinant wild type virus ( rSCV , Fig 3B ) and the chimeric virus ( SR-PLP-rSCV , Fig 3C ) , indicating that the SR-PLP was able to functionally compensate the SA-PLP . As already suggested by protease cleavage assays , SR-PLP-rSCV was slightly less sensitive against protease inhibitor 3e than SA-PLP-rSCV ( rSCV: IC50 = 2 . 36 μM; SR-PLP-rSCV: IC50 = 11 . 02 μM ) , providing additional evidence for functional and structural integrity of SR-PLP in the context of SARS-CoV replication ( Fig 3D ) . To exclude the possibility that 3e was cytotoxic , a cell viability assay was conducted . The number of viable cells decreased with increasing amounts of inhibitor to a minimum of 80% at the highest dose of 50 μM ( S2 Fig ) . This confirms the specific action of 3e . To obtain a more quantitative comparison of SA- and SR-PLP during virus replication , multistep growth curve experiments were done for both viruses in Vero cells . Both viruses grew to the same titers at all tested time points ( range: 8 . 2x102 PFU/ml to 8 . 2x106 PFU/ml; 8 , 14 , 24 , 48 hours post infection [hpi] , Fig 4A ) . Notably , growth curves differed when both viruses were grown in the type I IFN-competent primate cell line MA-104 . SR-PLP-rSCV grew to significantly lower titers than rSCV ( general linear regression model , p = 0 . 038/R-square = 0 . 612; differences 3 . 7-fold at 14 hpi and 3 . 6-fold at 24 hpi , Fig 4B ) . The reduced growth of SR-PLP-rSCV in MA-104 cells may indicate a less efficient viral counteraction against type I IFN . In order to compare the IFN sensitivity of both viruses , Vero cells were treated with a defined concentration of pan-species IFN-α . A significantly increased IFN sensitivity ( 4 . 2-fold at 100 IU/ml; p = 0 . 008 in a two-sided t test ) compared to rSCV confirmed that the anti-IFN activity of SR-PLP was decreased in primate cells ( Fig 4C ) . To further confirm that growth of SR-PLP-rSCV is also reduced in context of the human cell environment , we performed multistep growth curve experiments in the type I IFN competent human lung epithelial cell line Calu-3 ( Fig 4D ) . Again , SR-PLP-rSCV grew to significantly lower titers than rSCV ( 2 . 6 fold at 24 hpi [p = 0 . 023] and 1 . 5 fold at 48 hpi [p = 0 . 016] in a two-sided t test ) . A 1 . 5-fold increased detection of IFN-β mRNA expression levels in SR-PLP-rSCV- compared to rSCV-infected Calu-3 cells further confirmed that the SR-PLP is less efficient in blocking IFN induction ( Fig 4E ) . To better reflect the human respiratory tract and to generalize the notion that SR-PLP-rSCV grows less efficiently in presence of an active type I IFN response , multistep growth curve experiments were conducted in human airway epithelial cells ( HAE ) . In accordance with our previous studies [42] virus growth was generally delayed in HAE compared to the monoclonal primate and human cell cultures . Importantly , a significantly reduced growth was detected for SR-PLP-rSCV compared to rSCV at 96 hpi ( 10 . 3 fold; p = 0 . 006 in a two-sided t test ) providing further confirmation that the anti-IFN activity of SR-PLP may be decreased in type I IFN competent cells ( Fig 4F ) . The treatment with IFN-α in the above experiment ( Fig 4C ) would broadly affect IFN induction , signaling , and response . Because PLP IFN antagonism functions have been linked to IRF-3 function , a sensitive assay for IRF-3 nuclear translocation was established . Nuclear translocation of an overexpressed IRF-3/GFP fusion protein was stimulated by superinfection with Rift Valley fever virus clone 13 ( RVFV Cl 13 ) , an RVFV-mutant devoid of the IFN induction antagonist NSs . RVFV Cl 13 is known to trigger a strong IFN response [52 , 53] . The proportion of cells with nuclear translocation of GFP signal was counted microscopically . As summarized in Fig 5A and 5B , ectopic expression of SA- and SR-PLP blocked the nuclear translocation of IRF-3 to comparable levels ( SA-PLP: 18% and SR-PLP: 20% of cellular IRF-3/GFP fusion proteins located in the nucleus ) . The SO-PLP of the outlying virus blocked the nuclear translocation of IRF-3 even more efficiently ( 3% of cellular IRF-3/GFP fusion proteins located in the nucleus ) . For all PLPs , CA-mutants were included in the experiment to determine whether anti-IFN effects depended on protease function . The inhibitory capacity of the SA- and SO-PLP CA-mutants were strongly reduced , but still detectable at significant levels ( SA-PLP: 71% and SO-PLP: 45% of cells with IRF-3/GFP fusion protein located in the nucleus ) . In contrast , the SR-PLP CA-mutant had essentially lost its ability to block IRF-3 nuclear translocation ( 94% of cells with nuclear IRF-3/GFP signal ) . To see whether the PLP activity against IRF-3 translocation determined IFN counteraction , the abilities to block IFN promoter activation were compared in a reporter gene assay . All PLPs efficiently blocked IFN-β promoter activation in a dose-dependent manner ( 6 . 3% , 2 . 3% and 2 . 2% residual IFN-β promoter activation , respectively , at plasmid concentrations of 50 ng for SA-PLP , SR-PLP and SO-PLP; Fig 5C–5E ) . However , differences were observed with CA-mutants . The SA-PLP and SO-PLP CA-mutants still blocked the IFN-β promoter activity in a dose-dependent manner , albeit with reduced efficiency as compared to wild type ( SA-PLP: 26 . 4% [Fig 5C] and SO-PLP: 12 . 9% [Fig 5E] residual IFN-β promoter activation at a plasmid concentration of 50 ng ) . The SR-PLP CA-mutant lost any significant blocking functions against IFN-β promoter activation ( Fig 5D ) . The differential capabilities of SA-PLP , SR-PLP and SO-PLP to counteract IFN induction may be attributable to differential capabilities to inactivate ubiquitin or ubiquitin-like modifiers such as ISG15 . SR-PLP may have to rely on protease-dependent cleavage , while SA-PLP and SO-PLP may be able to provide additional IFN-antagonistic activity by binding of ubiquitin or ubiquitin-like modifiers . To understand if within the SARSr-CoV species the SA-PLP as compared to SR-PLP provides additional anti-IFN activity depending on ubiquitin-binding but not cleavage , we decided to introduce a modification in the ubiquitin-binding surface of SA-PLP that should not affect the protease-processing activity . Amino acid residue M209 of SA-PLP interacts directly with a binding patch comprised of I44 , V70 , L8 , R167 and D168 of ubiquitin based on co-crystallization studies [38 , 54] . In the MERS-CoV PLP , replacement of methionine at a corresponding position by the bulkier arginine residue was found to prevent the interaction between ectopically expressed PLP and ubiquitin [41] . We thus decided to study the effect of a mutation that selectively destabilizes ubiquitin-binding in SARS-CoV via an M209R mutation . In order to specifically focus on the consequences of the mutation on IFN antagonism , we had to exclude that M209R disturbs the protease activity of the PLP . Expression plasmids were generated , and protease activities were compared by trans-cleavage assays , confirming that the mutation had no impact on the cleavage activity ( Fig 6A ) . M209R PLP processed the nsp2/3 substrate with the same efficiency as SA-PLP . To confirm that ubiquitin-binding of SA-PLP is reduced by the M209R mutation , a DUB assay was conducted with wild type and M209R SA-PLP expression plasmids as previously described . DUB efficiency of M209R PLP ( Fig 6B , lane 4 ) was reduced compared to the wild type PLP ( Fig 6B , lane 2 ) as indicated by the increased number of ubiquitinated protein bands in lane 4 . The reduced ubiquitin-binding efficiency should result in a decreased IFN-antagonistic activity of M209R PLP . We thus compared IFN-β promoter activation in a reporter gene assay . Both , the M209R and the corresponding M209R CA mutant , had less anti-IFN activity compared to SA-PLP ( Fig 6C ) . The combination of M209R and CA ( Fig 6C , right black column ) resulted in the complete depletion of significant anti-IFN activity . To explore the consequences of the disturbed ubiquitin-binding on virus growth , the M209R mutation was introduced into rSCV . As expected , the resulting virus , termed rSCV-PLP/M209R , grew to titers similar to rSCV in a multistep growth curve in Vero cells ( Fig 6D ) . As observed with SR-PLP , the congruence with wild type changed when both viruses were grown in type I IFN competent MA-104 cells . In MA-104 cells rSCV-PLP/M209R grew to significantly lower titers than rSCV ( general linear regression model , p = 0 . 034/R-square = 0 . 217; differences 1 . 9-fold at 14 hpi and 3 . 5-fold at 24 hpi , Fig 6E ) . To confirm that the destabilizing mutation in the ubiquitin-binding domain caused increased IFN sensitivity , Vero cells were pretreated with 100 IU of pan-species IFN-α ( Fig 6F ) . Upon treatment , the growth difference between both viruses was augmented , confirming a significantly increased IFN sensitivity of rSCV-PLP/M209R ( 4 . 3-fold at 0 IU/ml [p = 0 . 023] and 5 . 2-fold at 100 IU/ml [p = 0 . 007] in a two-sided t test ) . To further confirm the type I IFN-dependent growth differences of rSCV and rSCV-PLP/M209R , multistep virus growth kinetics were done in the human lung epithelial cell line Calu-3 ( Fig 6G ) . Compared to wild type virus-infected cells , the M209R mutation led to significantly lower titers in rSCV-PLP/M209R-infected Calu-3 cells ( 8 . 3 fold at 24 hpi [p = 0 . 007] and 2 . 8 fold at 48 hpi [p = 0 . 003] in a two-sided t test ) . In order to directly detect the virus-dependent effect on IFN-β gene expression , Calu-3 cells were infected with both viruses with an MOI of 1 for 24 h and IFN-β mRNA upregulation was detected by quantitative real time RT-PCR assay ( Fig 6H ) . To determine the fold-induction of the IFN-β mRNA , the 2−ΔΔCt method was applied with TATA binding protein ( TBP ) as housekeeping gene [55] . The rSCV-PLP/M209R-infected cells had a more pronounced IFN-β mRNA expression level ( 3 . 2-fold difference between rSCV and rSCV-PLP/M209R at 24 hpi; p = 0 . 031 in a two-sided t test ) compared to rSCV . The type I IFN-inducing RVFV Cl 13 was included as an IFN-β mRNA expression control showing that the cells had a fully functional IFN response . The phenotypic similarity between SR-PLP-rSCV and rSCV-PLP/M209R suggests that SA-PLP , but not SR-PLP , provides additional IFN antagonism via binding of ubiquitin or ubiquitin-like modifiers . As ubiquitin ( and its derivatives ) are conserved in mammalian species , we expected that ubiquitin-dependent anti-IFN activities should be independent of the host species . As all members of the species SARS-CoV are carried by bats of the genus Rhinolophus , IFN-competent Rhinolophus alcyone lung epithelial cells ( RhiLu ) were made transgenic for the human ACE2 receptor ( RhiLu-hACE2 ) by lentiviral transduction [56] , and were infected with rSCV and SR-PLP-rSCV ( Fig 7A ) . In multistep growth curves , SR-PLP-rSCV grew to significantly lower titers compared to rSCV ( 13-fold lower at 24 hpi [p = 0 . 016] and 18-fold lower at 48 hpi [p = 0 . 005] , respectively ) . Decreased viral growth of SR-PLP-rSCV in RhiLu-hACE2 bat cells may indicate more efficient IFN counteraction by rSCV compared to SR-PLP-rSCV , as in IFN-competent MA-104 cells . Remarkably , the growth differences were even more prominent in the bat cells compared to IFN competent primate cells ( 13-fold in RhiLu-hACE2 versus 3 . 6-fold in MA-104 cells at 24 hpi ) . In order to investigate if the growth differences in bat cells were reflected by differences in IFN suppression , a quantitative real-time-PCR assay specific for the Rhinolophus IFN-β gene was established . To determine the specific IFN-β mRNA expression levels , RhiLu-hACE2 cells were infected with rSCV , SR-PLP-rSCV and , as IFN induction control , RVFV Cl 13 ( MOI of 1 ) . Total RNA was extracted from cell lysates at 24 hpi . To determine the fold-induction of the IFN-β mRNA expression levels , the 2−ΔΔCt method was applied with β-actin as housekeeping gene [55] . In comparison to rSCV , SR-PLP-rSCV infection led to significantly higher IFN-β expression levels ( 6 . 6 fold at 24 hpi [p = 0 . 001] in a two-sided t test , Fig 7B ) . In summary , these results suggest that the increased IFN sensitivity of SR-PLP is independent of the host cell species .
SARS-CoV variants exist across Europe in widespread Rhinolophus bat species . It is important to understand whether these viruses constitute a risk for human infection [57] . Host tropism is mainly thought to be determined by the spike protein , but studies have shown that CoV populations in natural reservoirs can contain a plethora of spike variants , including variants that can directly mediate entry into human cells [16 , 20] . Because of widespread recombination , spike proteins can be exchanged between viral genetic lineages in the reservoir . The spike protein may not sufficiently represent the variability of virulence traits in the reservoir . Here we studied a conserved viral function related to host interaction by focusing on PLP , a multifunctional protein that drives essential steps in the infection cycle including cleavage of the viral polyprotein as well as deubiquitination and deISGylation . While these latter functions interfere with molecules of the IFN pathway and cytokine production , it was unknown to what extent they contribute to the replication phenotype of SARS-CoV . Our initial comparisons of PLP functions of epidemic and reservoir-associated virus were based on ectopically-expressed protein , confirming that essential properties such as cleavage of the viral polyprotein and sensitivity to a known inhibitor were almost identical between epidemic- and reservoir-derived PLP . However , overexpression assays cannot reflect the whole complexity of the viral life cycle including the timing and compartmentalization of functions . To detect more the subtle differences that we expected to occur between natural virus variants , we generated an infectious clone carrying the reservoir-derived PLP in an otherwise unmodified SARS-CoV backbone . Growth properties of the chimeric virus in IFN-deficient Vero cells confirmed that the reservoir-derived PLP effectively cleaved the viral polyprotein and supported recombinant virus replication kinetics comparable to wild type SARS-CoV . Only when grown in cells with a fully active IFN system , wild type SARS-CoV replicated better than the PLP-chimeric virus . The small but significant growth differences ranged between 2 . 6 ( Calu-3 ) and 10 . 3-fold ( HAE ) depending on the time point post infection and the applied cell cultures . We assume that other viral proteins with IFN antagonistic functions , like protein 6 and 3b , might have compensated for the decreased immune-modulating functions of SR-PLP and PLP/M209R preventing a more pronounced growth difference [27] . Importantly , we showed that the effect was increased when artificial IFN was added to the cell cultures and was additionally accompanied by a more efficient blocking of IFN-β mRNA upregulation , obtaining further confirmation for the specificity of the PLP function within the IFN response . We showed by studies on catalytically-inactivated PLP constructs that the anti-IFN function of the reservoir-derived SR-PLP depended on the protease function while SARS-CoV PLP exerted an additional , protease-independent activity in IFN evasion . The likely mechanism for this additional activity has been identified by in-vitro studies to involve binding of ubiquitin or ubiquitin-like modifiers [34 , 54 , 58] . Here we confirmed the functional link between DUB functions and IFN antagonism for the first time in a replicating CoV . Because the cytopathogenic nature of SARS-CoV infection prevented sensitive cell-based assays for DUB- and deISGylating functions , we relied on knowledge of structure-function relationships from a study on equine arteritis virus ( EAV ) , an arterivirus related to CoVs [35] . In EAV PLP2 , a ubiquitin-binding surface distinct from the protease active site was identified and mutagenized without causing a loss of protease processing . A definite link to IFN antagonism was established by engineering of a destabilizing mutation into the ubiquitin-binding surface in an EAV infectious clone , causing increased induction of IFN during replication . By introduction of a homologous mutation in our SARS-CoV infectious clone , we found an even clearer loss of IFN antagonism than with EAV . The choice of the mutation was informed by another study that defined the primary ubiquitin-binding surface in SARS-CoV PLP and found position M209 to critically interact with a conformational binding patch ( I44 , V70 , L8 ) on ubiquitin but to not affect protease activity [54] . Concordantly , we found no signs of catalytic inactivation of the M209 mutant in Vero cells where it replicates like wild type virus , and a pattern of fully protease-dependent IFN antagonism like that of the reservoir-derived PLP . We conclude that ubiquitin binding constitutes an anti-IFN function of SARS-CoV PLP that is independent of the conserved protease function , is phenotypically relevant for replication level and immune evasion , and is variable among viral variants . The IFN-related effects observed in the present study seem to be relevant as they resemble effects observed upon deletion of the 2’-O-methyltransferase activity , another essential replicase function that modifies IFN-recognition [59] . In that study , a D130A mutation in the nsp16 2’-O-methyltransferase domain replicated like wild type in Vero cells but caused about 10-fold lower replication in an IFN-competent human airway epithelial cell line . When engineered in a mouse-adapted backbone , the mutant was cleared faster from the lungs of infected mice , and caused significantly less weight loss . The present study did not use in vivo models to evaluate the consequences of the introduced mutations on replication . Mouse experiments may not reflect the role of PLP in host switching because ISG15 ( expectably the most important ubiquitin-like modifier processed by SARS-CoV PLP [38] ) is not conserved between humans , bats , and mice [60] . However , we can derive from the existence of SR-PLP in bats that the ubiquitin-related function of PLP is not essential for SARS-CoV replication in the natural host , and functional variants reflect natural diversity . While it has been suggested that SARS-CoV may have gained virulence for humans during human-to-human passage and adaptation [12 , 61] , the differences in immune interaction observed in the present study are not dependent on the host cells used . Even in bat cells from the natural host of SARS-CoV , epidemic SARS-CoV has additional anti-IFN activity . Differential virulence via PLP may therefore have pre-existed within viral reservoir , rather than having evolved by positive selection upon adaptation within the human host . Therefore , we postulate that PLPs are a variable virulence trait among members of the species SARS-CoV that exist in natural reservoirs . Based on the PLP structure , there is huge potential for natural sequence variants to influence ubiquitin-binding . For instance , PLP has a second ubiquitin-binding site enabling a bi-dentate binding mechanism that may explain the PLP preference for di-ubiquitins and ISG15 [54] . Crystallization of PLP in complex with K48-linked di-ubiquitin identified the structure of the second binding site ( Ub2 ) , which seems to be more critical for K48-linked di-ubiquitin-binding than the primary binding surface [38] . The primary ubiquitin-binding surface ( Ub1 ) is more critical for ISG15 binding [38 , 60] . The SARS-CoV strains found in reservoirs in China and Europe show several variants at both ubiquitin-binding sites that could be studied for effects on replication level or other virulence traits ( Fig 8A ) . In silico modeling of the ubiquitin-binding sites of SA- and SR-PLP shows that sequence variations may even cause structural changes in- and adjacent to the binding sites ( Fig 8B and 8C ) . The fact that , apart from our observed functional differences , there may also be structural changes between two closely related PLPs of a single virus species ( the species “SARS-related CoV” ) is particularly intriguing as such differences were previously only seen among CoVs that belong to different virus genera . The PLP of HCoV-NL63 , for example , has similar protease-independent anti-IFN functions as the PLP of SARS-CoV whereas MERS-CoV PLP anti-IFN activity clearly depends on protease function [34 , 37] . The existence of functional diversity in the reservoir makes sense in light of viral evolution and emergence . For instance , the capacity to modulate processing and substrate binding in separate domains of PLP opens the possibility for reservoir-associated CoVs to adjust fitness or virulence levels based on changes in host population structure . Increased virulence may confer increased transmission rates , but may come at the cost of host population decline . Based on our data it seems that the reservoir contains a considerable degree of functional variability that is based on highly conserved host functions such as the ubiquitin systems . Virus adaptation to the human host may not only be an evolved property at the beginning of virus emergence . Rather , the reservoir may contain pre-formed virulence traits that can be predicted by experiments informed by detailed knowledge of molecular virus functions , enabling new approaches to forecast emergence potential .
Fecal pellets of Hipposideros bats were provided by author Christian Drosten . All fecal samples were collected non-invasively by trained field biologists and stored at -80°C until analysis . In detail , bats were caught with mist nets , which were checked at intervals of 5 min . Captured bats were freed from nets immediately and put into cotton bags for several minutes . While being kept in bags , bats produced fecal pellets that were collected and transferred into RNAlater RNA stabilization solution ( QIAGEN , Hilden , Germany ) . Procedures were previously described in [62] , and were consistent with guidelines of the American society of mammalogists [63] for the use of wild mammals in research and national guidelines for the capture , handling , and care of bats . For all capturing , sampling and sample export , permission was obtained from the Wildlife Division of the Ministry of Lands , Forestry , and Mines in Ghana ( permit no . CHRPE49/09; A04957 ) as described previously in [45] . Primary human airway epithelial cells were procured from patients who underwent surgical lung resection for any pulmonary disease and who gave informed consent . This was done in accordance with the ethical approval EKSG 11/044 , EKSG 11/103 and KEK-BE 302/2015 . HEK-293T ( Friedemann Weber , University of Gießen ) and MA-104 ( Friedemann Weber , University of Gießen ) , Vero ( Jindrich Cinatl , University of Frankfurt ) , Vero E6 ( ATCC , ATCC CRL-1586 ) , Calu-3 ( ATCC HTB-55 ) and RhiLu-hACE2 ( provided by author Marcel A . Müller [56] ) cells were cultivated in Dulbecco’s modified Eagles medium ( DMEM ) supplemented with 10% fetal bovine serum ( ThermoFisher Scientific ) , 1% penicillin/streptomycin , 1% non-essential amino acids , 1% L-glutamine and 1% sodium pyruvate in a 5% CO2 atmosphere at 37°C . HAE were generated and maintained as previously described [64] . RVFV Cl 13 was a kind gift from Friedemann Weber ( University of Gießen ) . Infection experiments with rSCV were done under biosafety level 3 conditions with enhanced respiratory personal protection equipment . To ensure high level protein production of SA- , SR- and SO-PLPs in eukaryotic cells the codon-usage was optimized based on the human codon-usage frequency . In addition , potential splice sites and polyadenylation signal sequences were eliminated before the sequences were cloned into the eukaryotic expression plasmid pCAGGS along with a carboxy-terminal FLAG epitope tag . Regions of SA-PLP were PCR amplified from parental plasmid pPLpro1541-1855 [48] . SR- and SO-PLPs were synthesized by Geneart . Primers , used for cloning are listed in S1 Table . Site directed mutagenesis was done to change the catalytic cysteines into alanines ( QuikChange Mutagenesis ) and replace the ubiquitin-binding M209 residue to arginine ( Gibson assembly , NEB ) using the indicated primers below ( S1 Table ) . The insertion of correct mutations was verified by DNA sequencing . Nsp2/3-GFP , pCAGGS-HA-Ub , pISG15-myc and pRL-TK plasmids were kindly provided by Ralph S . Baric ( University of North Carolina , USA ) , Adriano Marchese ( Medical College of Wisconsin , USA , Min-Jung Kim ( Pohang University , Republic of Korea ) and Karl-Klaus Conzelmann ( University of Munich , Germany ) . PcDNA3-Ube1L , pcDNA3-UbcH8 and pcDNA3-Herc5 were kind gifts from Robert M . Krug ( University of Texas , USA ) . P125-Luc , pEF-BOS-MDA5-3xFLAG His10 , GFP-IRF-3 and pCAGGS were previously described [65] . The synthesis and characterization of SA-PLP inhibitor 3e are described in [51] . The SO-PLP sequence was obtained from a Hipposideros bat fecal sample ( BUO2-B-F114 ) as previously described by Pfefferle et al . [45] . According to a nucleotide sequence alignment , containing related PLP gene sequences , primers were located to the most conserved regions within- and downstream of the PLP domain . Two different primer sets were applied in two successively performed PCR reactions . The nucleotide sequence information gained was used for the design of primers specifically targeting the SO-PLP sequence . The sequencing strategy and the primers are shown in S2 and S3 Tables . SR-PLP-rSCV was constructed using the previously established cDNA clone [66] . This approach was based on bacterial artificial chromosomes ( BAC ) for keeping the full-length CoV cDNA stable . For the construction of the full-length infectious cDNA clone seven subclones ( referred to as pA1 , pA2 , pB , pC , pD , pE and pF ) covering the whole SARS-CoV genome were generated and assembled in a stepwise procedure . For the generation of SR-PLP-rSCV a chimeric SR-PLP subclone , named pBG-ABCD2 , containing approximately one-half of the SARS-CoV genome and the desired SR-PLP replacement , was used . PBG-ABCD2 was joined with subclone pDEF into a full-length BAC cDNA clone and rescued as in [66] . SR-PLP-rSCV was sequenced to confirm the presence of SR-PLP and the absence of any further mutations with the following primers: primer for reverse transcription ( Brev: 5’-TGAACCGCCACGCTGGCTAAACC-3’ ) , sequencing primers ( B4622F: 5’-CTTAAAGCTCCTGCCGTAGTG-3’ , BG4792F: 5’-TATTAAGGTGTTCACAACTGTAG-3’ , BG5631F: 5’-AAATTGATGGTGCTCTCTTGAC-3’ ) Cells were seeded at a concentration of 3 . 5x105 cells/ml . After 24 h , virus stocks were diluted in serum-free medium according to the desired MOI . For virus adsorption 200 μl ( 24-well ) or 1 ml ( 6-well ) of virus master mix was added to the cells and incubated for 1 h at 37°C . After 1 h , the virus dilutions were removed and the wells were washed twice with 1x PBS and refilled with supplemented DMEM . Supernatants were taken at the indicated time points and studied further . For infection of HAE air liquid interface cultures , the apical surface was washed twice with 200 μl Hank’s balanced salt solution ( HBSS ) to remove mucus . Virus stocks were diluted in HBSS and HAE were infected with an absolute infectious dose of 40 , 000 PFU . Cells were incubated for 1 . 5 h at 37°C in a 5% CO2 atmosphere with 95% humidity . After adsorption , virus dilutions were removed and the wells were washed three times with 200 μl HBSS . Samples were taken at the indicated time points by applying 200 μl HBSS to the apical surface 10 minutes prior to the actual time points . Basolateral medium was exchanged at 48 hpi . Assessment of protease activity was conducted by a trans-cleavage assay . 2x105 cells/ml were seeded in a 12-well plate 24 h prior to transfection . 300 ng of PLP-encoding plasmids were coexpressed with 25 ng of SARS-CoV nsp2/nsp3-GFP substrate [49] . After 16 h , cells were lysed with 100 μl lysis buffer ( 20 mM Tris [pH 7 . 5] , 150 mM NaCl , 1 mM EGTA , 1 mM EDTA , 1% Triton X-100 , 12 . 5 mM Na pyrophosphate , 1 mM β-glycerophosphate , 1 mM Na ortho-vanadate , 1 mg/ml leupeptin , 1 mM PMSF ) and lysates were separated by SDS-PAGE using a semi-dry transfer blotter . After protein transfer , the membrane was blocked by 5% dried milk in TBST buffer ( 0 . 9% NaCl , 10 mM Tris-HCl [pH 7 . 5] , 0 . 1% Tween 20 ) for 1 h at room temperature . The membrane was incubated with a rabbit-antibody directed against the GFP-epitope tag ( ThermoFisher Scientific ) . The membrane was washed three times for 15 min in TBST buffer . Next , the membrane was incubated with HRP-coupled donkey anti-rabbit secondary antibody ( SouthernBiotech ) . After 1 h , the membrane was washed three times for 15 min in TBST buffer . To confirm PLP expression , the membrane was probed with mouse anti-FLAG ( Sigma-Aldrich ) followed by goat anti-mouse HRP-coupled ( SouthernBiotech ) antibodies . Mouse anti-β-actin ( Sigma-Aldrich ) was used to detect host cell proteins as a loading control . Secondary detection was performed using goat anti-mouse HRP antibody ( SouthernBiotech ) . The assay was performed using HEK-293T cells in black 96-well plates with clear bottom . Cells were transfected with 37 . 5 ng pGlo-30F-RLKGG [50] and 50 ng PLP-expressing plasmids using Fugene HD ( Promega ) according to the manufacturer’s instructions . At 14 hours post transfection ( hpt ) , cells were equilibrated with GloSensor reagent ( Promega ) . For inhibitor quantification , cells were equilibrated at 13 hpt with GloSensor reagent as indicated above and at 14 hpt diluted inhibitor or DMSO was added . Luminescence was measured after 1 h for the following 6 h . For data analysis , the fold luciferase induction was calculated independently for each PLP by calibrating the values , obtained for every time point to the respective starting values . HEK-293T cells ( 2x105 cells/ml ) were transfected in the 12-well format with 300 ng of HA-ubiquitin ( HA-Ub ) and 50 , 100 or 200 ng PLP-expressing plasmids using TransIT-LT1 transfection reagent ( Mirus ) . At 18 hpt , cells were treated with 100 μl of lysis buffer . Proteins were separated by SDS-PAGE and analyzed by Western blotting as described above . Western blot analysis was done using mouse anti-HA serum ( Covance ) and goat anti-mouse HRP ( SouthernBiotech ) antibodies . HEK-293T cells were transfected with plasmids encoding for myc-epitope tagged ISG15 ( 250 ng of pISG15-myc ) and its conjugation machinery , comprising a set of three ligases ( UbcH8/125 ng pUbcH8 , Ube1L/125 ng pUbe1L and Herc5/125 ng pHerc5 ) as in [40] . PLP-encoding plasmids were cotransfected in amounts of 100 ng . After 18 h , cells were lysed and separated by SDS-PAGE as indicated above . The extent of ISGylated cellular proteins was analyzed by Western blotting using mouse anti-myc ( MBL Life science ) and goat anti-mouse HRP ( SouthernBiotech ) antibodies . 2x105 HEK-293T cells/ml were transfected with plasmids encoding Renilla ( RL ) and Firefly ( FF ) luciferases . The FF luciferase gene was under control of an IFN-β promoter ( p125-luc ) . RL luciferase gene was cloned behind a herpes simplex thymidine-kinase ( TK ) promoter . IFN-β promoter activation was triggered by the overexpression of the cellular MDA5 , which led to an auto-activation of the IFN pathway [67] . Transfection of DNA plasmids was done with Fugene HD ( Promega ) according to the manufacturer’s instructions . Plasmid amounts are given for the 24-well plate format and are as follows: pRL-TK ( 5 ng ) , p125-luc ( 250 ng ) , pEF-BOS-MDA5-3xFLAG His10 ( 100 ng ) and 1 to 50 ng of PLP-expressing plasmids . Cells were lysed at 17 hpt . 20 μl of luciferase-containing lysate was transferred to an opaque , white 96-well microtiter plate and used for assessment of luciferases activity by a bioluminescence detection reader . The IRF-3 translocation assay was performed as described elsewhere [65] . Briefly , MA-104 cells were transfected with plasmids encoding 250 ng GFP-IRF-3 and either EV or 1000 ng of FLAG-tagged PLP-expressing plasmids using Fugene HD ( Promega ) . To activate the IFN pathway , cells were infected with RVFV Cl 13 and an MOI of 5 at 17 hpt . At 8 hpi cells were fixed with 4% paraformaldehyde and permeabilized with 0 . 1% TritonX-100 . Immunofluorescence analysis was done as in [68] . Samples transfected with EV were treated with anti-RVFV mouse serum ( Friedemann Weber , University of Gießen , [53] ) and goat anti-mouse Cy3 ( Dianova GmbH ) secondary antibody . PLP-expressing cells were stained with mouse anti-FLAG ( Sigma-Aldrich ) and goat anti-mouse Cy3 ( Dianova GmbH ) antibodies . Samples were analyzed by a fluorescence microscope ( Zeiss ) . Depending on the number of transfected cells , at least three images were taken , and the number of cells double-positive for GFP and FLAG was divided by the number of cells showing IRF-3 nuclear translocations . 3 . 5x105 VeroE6 cells/ml were seeded in a 24-well plate 24 h prior to infection . A 1:10 serial dilution of samples was generated , 200 μl dilution was added to the cells and incubated for 1 h at 37°C for adsorption . After virus samples were removed , 500 μl overlay ( 2 . 4% avicel diluted in 2x DMEM ) was added to each well . The overlay was discarded at 3 days post infection ( dpi ) and cells were fixed for 30 min in 6% formaldehyde . The cells were washed once with 1x PBS and stained with crystal violet working solution for 15 min . Plaque forming units were determined from at least two dilutions for which distinct plaques were detectable . Viral RNA was extracted with the NucleoSpin RNA virus isolation kit ( Macherey-Nagel ) after the manufacturer’s instructions . 1 μl of viral RNA was applied to each reaction . Quantification of genomic SARS-CoV RNA was done using the SuperScript III one-step reverse transcriptase-PCR system ( Invitrogen ) with the Platinum Taq DNA polymerase according to the manufacturer’s recommendations and these primers ( SARS-F: 5’-CCCGCGAAGAAGCTATTCG-3’ , SARS-P: Fam-5’-ACGTTCGTGCGTGGATTGGCTTTG-3’-BHQ , SARS-R: 5’-AGTTGCATGACAGCCCTCTACA-3’ ) . An established SARS-CoV standard curve was used in each run [69] to quantify RNA copies per ml . To analyze IFN-β expression in Calu-3 and Rhinolophus alcyone cells ( provided by authors Christian Drosten and Marcel A . Müller ) , mRNA was extracted with the NucleoSpin RNA isolation kit ( Macherey-Nagel ) . Real-time PCR was done as described previously [70] using the following primers: Hu-IFNB1-F: 5’-AGGATTCTGCATTACCTGAAGG-3’ , Hu-IFNB1-P: Fam-5’- TCCACTCTGACTATGGTCCAGGCA-3’-ZEN and Hu-IFNB1-R: 5’-GGCTAGGAGATCTTCAGTTTCG-3’ ( Calu-3 ) and Rhi-IFNB1-F: 5’-AAATAATGGAGGAGGAAAACTTCAC-3’ , Rhi-IFNB1-P: Fam-5’-CGAAACATGAGCACGCTGCACCTG-3’-BHQ and Rhi-IFNB1-R: 5’- CGCCTGATCCTTAGGTAGTAATTCT-3’ ( Rhinolophus alcyone ) . To determine the induction of IFN-β relative to TBP ( Calu-3 ) and β-actin ( Rhinolophus alcyone ) the 2−ΔΔCt method was applied [55] . Phylogenetic analysis was performed using the neighbor joining algorithm in Geneious 9 . 1 . 5 under the assumption of a Tamura-Nei genetic distance model . Alignments were done based on the amino acid codes by the BLOSUM62 algorithm in the Geneious 9 . 1 . 5 software package . The identities and similarities of the listed proteins were calculated from the amino acid alignment using the BLOSUM62 substitution matrix with a threshold of 1 . SO-PLP: MG916963 | Novel detection and sequencing technologies have greatly improved our knowledge of virus diversity in nature . Metaviromic screening of zoonotic animal reservoirs has become an established approach in pathogen surveillance and pandemic preparedness research . However , knowledge of viral genomes and host-associated viromes is insufficient to predict zoonotic spillover events of reservoir-borne viruses . Phenotypic characterization of important viral functions will be necessary to identify virulence traits that determine the potential of viral emergence . As proof-of-principle , the present study demonstrates relevant functional differences between one of the main host immune antagonists of bat-borne viruses that belong to the same virus species as the epidemic agent of SARS . The antagonist , the papain-like protease , shows double action against IFN-mediated antiviral effects in epidemic SARS-coronavirus ( binding and processing of ubiquitin ) , whereas the homologous protein in bat-borne viruses has only the processing function . This finding is surprising as the papain-like protease is a highly conserved protein domain that was not expected to vary among conspecific coronaviruses . PLP function may represent a variable virulence trait among reservoir-borne viruses . The preformed virulence of the primordial genetic lineage may have supported the emergence of SARS-CoV as a human epidemic agent . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"vero",
"cells",
"medicine",
"and",
"health",
"sciences",
"coronaviruses",
"luciferase",
"pathology",
"and",
"laboratory",
"medicine",
"enzymes",
"pathogens",
"sars",
"coronavirus",
"biological",
"cultures",
"microbiology",
"enzymology",
"viruses",
"rna",
"viruses",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"zoonoses",
"proteins",
"medical",
"microbiology",
"oxidoreductases",
"microbial",
"pathogens",
"viral",
"replication",
"cell",
"lines",
"biochemistry",
"virology",
"viral",
"pathogens",
"interferons",
"biology",
"and",
"life",
"sciences",
"proteases",
"organisms"
] | 2018 | The papain-like protease determines a virulence trait that varies among members of the SARS-coronavirus species |
Pituitary endocrine cells fire action potentials ( APs ) to regulate their cytosolic Ca2+ concentration and hormone secretion rate . Depending on animal species , cell type , and biological conditions , pituitary APs are generated either by TTX-sensitive Na+ currents ( INa ) , high-voltage activated Ca2+ currents ( ICa ) , or by a combination of the two . Previous computational models of pituitary cells have mainly been based on data from rats , where INa is largely inactivated at the resting potential , and spontaneous APs are predominantly mediated by ICa . Unlike in rats , spontaneous INa-mediated APs are consistently seen in pituitary cells of several other animal species , including several species of fish . In the current work we develop a computational model of gonadotropin releasing cells in the teleost fish medaka ( Oryzias latipes ) . The model stands out from previous modeling efforts by being ( 1 ) the first model of a pituitary cell in teleosts , ( 2 ) the first pituitary cell model that fires sponateous APs that are predominantly mediated by INa , and ( 3 ) the first pituitary cell model where the kinetics of the depolarizing currents , INa and ICa , are directly fitted to voltage-clamp data . We explore the firing properties of the model , and compare it to the properties of previous models that fire ICa-based APs . We put a particular focus on how the big conductance K+ current ( IBK ) modulates the AP shape . Interestingly , we find that IBK can prolong AP duration in models that fire ICa-based APs , while it consistently shortens the duration of the predominantly INa-mediated APs in the medaka gonadotroph model . Although the model is constrained to experimental data from gonadotroph cells in medaka , it may likely provide insights also into other pituitary cell types that fire INa-mediated APs .
Several types of excitable cells elicit electrical pulses called action potentials ( APs ) , which , depending on cell type , can trigger neurotransmitter release , cellular contraction , hormone release or other actions . APs are generated by a combination of ion channels in the plasma membrane , which are typically characterized by the type of ions they are permeable to , and their voltage and/or Ca2+ dependent gating properties . The primary role of APs in endocrine pituitary cells is to regulate the cytosolic Ca2+ concentration , which in turn controls the hormone secretion rate in these cells [1] . Hormone secretion often occurs as a response to input from the hypothalamus , peripheral endocrine glands , or from other pituitary cells . However , many endocrine cells are also spontaneously active [1–10] . The spontaneous activity is partly a means to regulate the re-filling of intracellular Ca2+ stores , but in several cells also leads to a basal release of hormones . An understanding of the mechanisms regulating the electrodynamics of these cells is therefore fundamental for understanding their overall functioning . While neuronal APs are predominantly mediated by TTX-sensitive Na+ currents ( INa ) , AP generation in endocrine cells depends strongly on high-voltage-activated Ca2+ currents ( ICa ) , which in addition to their role in affecting the voltage dynamics of the cell , also are the main source of Ca2+ entry through the plasma membrane [3 , 11 , 12] . In some studies of endocrine cells , APs were exclusively mediated by ICa , and the spontaneous membrane excitability was insensitive or nearly so to TTX [1 , 2 , 13–16] . In other studies , APs were evoked by a combination of ICa and INa [4 , 7 , 17–19] . In one of these studies , TTX was found to block single , brief action potentials , while action potentials of long duration and low amplitude persisted [18] , indicating the roles and different time courses of the ICa and INa components . The strong involvement of ICa could explain why pituitary APs typically last longer ( typically some tens of milliseconds [8] ) than neuronal APs ( a few milliseconds ) , which are mainly mediated by INa . All endocrine cells express INa [8] , and TTX sensitive APs can typically be triggered by current injections from hyperpolarized holding potentials even in cells where they are not elicited spontaneously [4 , 17 , 20 , 21] . The reason why the spontaneous activity in many cases is TTX insensitive is likely that a major fraction of INa is inactivated at the resting membrane potential [15 , 16] . The reason why this is not always the case , may be that the resting potentials vary greatly between different studies . Only for rat somatotrophs , resting potentials ranging as wide as from −30 mV [13] to −80 mV [18] have been reported . Computational models constructed to capture the essential activity of pituitary cells have predominantly relied on rat data . The typical resting potentials for rat pituitary cells lie in the range between −50 mV and −60 mV , and at these resting levels , INa tends to be inactivated and the spontaneous activity TTX insensitive ( see reviews in [8 , 22] ) . Models based on rat data have therefore typically excluded INa [3 , 9 , 23–30] . As TTX-sensitive spontaneous APs have been seen in mice corticotrophs [19] , INa was included in a recent computational model of these cells [31] , and in a more generic murine pituitary cell model based on the previous rat and mice models [32] . However , the role of INa in these models was mainly in modulating the firing patterns , and it was not essential for AP firing as such [31 , 32] . Furthermore , ICa and INa were in these models described by simplified kinetics schemes that were adjusted to give the models the desired firing properties , but not explicitly adapted to voltage-clamp recordings of the respective currents in the cell species being modeled . There are reasons to believe that the dynamical properties of the above cited pituitary cell models are not well suited to represent teleost pituitary cells . Firstly , TTX-sensitive spontaneous activity has been seen in goldfish gonadotrophs resting at −60 mV [4] , and TTX sensitive APs has been evoked from a holding potential as high as −50 mV in pituitary cells in cod [7] , suggesting that INa may be more available in resting pituitary cells in fish [4] . Secondly , data from gonadotrophs and somatotrophs in goldfish [4 , 20] and unspecified pituitary cells in tilapia [5] show APs with very short duration ( < 10 ms ) compared to the APs in the previous rat models ( several tens of ms ) , putatively indicative of a stronger involvement of INa . A third difference between fish and rat pituitary cells may be in the role of the big conductance K+ current ( IBK ) , which has been shown to have a paradoxical role in some rat pituitary cells , i . e . , it can prolong the duration of ICa-mediated APs , and sometimes give rise to pseudo-plateau bursts , contrary to what one would expect from a hyperpolarizing current [9 , 25] . IBK is almost absent in rat gonadotrophs [25] , and this was proposed as an explanation to why these cells tend to be less bursty than other pituitary cell types in rats [1 , 9 , 32] . In contrast , IBK is highly expressed in medaka gonadotrophs , but without making these cells bursty [12] . The indication that there are differences between rat and fish pituitary cells are further supported by experiments presented in the current work , performed on luteinizing hormone ( LH ) -producing gonadotroph cells in medaka . We show that these cells elicit brief spontaneous APs that , unlike spontaneous APs in the previous murine pituitary cell models , predominantly are mediated by TTX sensitive Na+ currents ( INa ) . Furthermore , we show that IBK acts to make APs narrower in medaka gonadotrophs , and thus have the opposite effect from what they have in rat gonadotrophs . As previous computational models based on murine data seem unsuited to describe the spontaneous activity of teleost pituitary cells , we here present a novel pituitary cell model constrained to data from medaka gonadotrophs . Given the putatively complex interplay between ICa and INa during the AP upstroke , we put extra effort into developing accurate models of these two currents , and constrained their kinetics to voltage-clamp recordings of the individual currents . In addition to ICa and INa , we included a leak current and three K+ currents in the model . These we adopted from previous pituitary cell models , and adjusted to adapt the firing properties of the model to current-clamp recordings from medaka gonadotrophs under control conditions , after application of TTX , and after application of the IBK blocker paxilline . By comparing the medaka gonadotroph model , which predominantly fires INa-mediated APs , with three previous models of murine pituitary cells , which predominantly [32] or exclusively [9 , 27] fire ICa-mediated APs , we explore the consequences of having different AP-generating mechanisms . We find that the medaka gonadotroph model produces spontaneous APs that are faster than those in the murine models , and thus more suited to describe the firing properties of teleost pituitary cells . Furthermore , we show that while IBK may broaden APs in the murine pituitary models [9 , 27 , 32] , it consistently had a narrowing effect on APs in the medaka gonadotroph model , and propose explanations to these model differences . By this , we add to the discussion of the role played by IBK in shaping pituitary APs [8] , and suggest that the effect of IBK on APs is mainly determined by the timing of IBK-activation relative to the AP peak , as also proposed previously [9] . Although the model presented here was tailored to represent gonadotroph cells in medaka , we believe that it is of a more general value for improving our understanding of INa-based APs in the pituitary , which are elicited by several endocrine cell types and in several animal species , depending on biological conditions [4 , 7 , 17 , 18 , 33–35] .
The general electrophysiological properties of gonadotroph cells in medaka were assessed through a series of voltage-clamp and current-clamp experiments . The voltage-clamp experiments used to develop kinetics models of Na+ and Ca2+ currents are presented in the Methods section . Here , we focus on the key properties of spontaneous APs as recorded in current clamp . Selected , representative experiments are shown in Fig 1 . Although variations were observed , the medaka gonadotrophs typically had a resting potential around −50 mV , which is within the range found previously for goldfish [4] and cod [10] gonadotrophs . As for goldfish gonadotrophs , the majority of medaka gonadotrophs fired spontaneous APs with peak voltages slightly below 0 mV . The spontaneous APs were always regular spikes ( i . e . , not bursts or plateau-like ) and the AP width varied between 3 and 7 ms ( blue traces in Fig 1 ) . Similar brief AP waveforms have been seen in previous studies on fish [4 , 5 , 20] , while the APs reported for rat gonadotrophs are typically slower , i . e . from 10-100 ms [8] . The spontaneous AP activity was completely abolished by TTX application ( Fig 1B ) . Finally , we explored how paxilline ( an IBK blocker ) affected the spontaneous activity of medaka gonadotrophs . In the experiment shown in Fig 1D , paxilline increased the firing rate and slightly reduced the mean AP peak amplitude , but these effects were not seen consistently in experiments using paxilline application . However , in all experiments , paxilline application was followed by a small increase of the resting membrane potential ( Fig 1C ) , and a broadening of the AP waveform ( Fig 1D2 and 1D3 ) . Similar effects have been seen in goldfish somatotrophs , where application of tetraethylammonium ( a general blocker of K+ currents ) lead to broadening of APs [20] . The effect of IBK in goldfish and medaka gonadotrophs is thus to make APs narrower , which is the opposite of what was found in rat pituitary somatotrophs and lactotrophs , where IBK lead to broader APs and sometimes to burst-like activity [9 , 25] . The model for medaka gonadotrophs is described in detail in the Methods section , but a brief overview is given here . The dynamics of the membrane potential is determined by the differential equation: C m d V d t = - ( I C a + I N a + I K + I B K + I S K + I l e a k ) . ( 1 ) The three K+ currents , IK , IBK and ISK , were based on previously published models ( [9 , 32] ) , but adjusted ( see Methods ) so that the final model had an AP shape and AP firing rate that were in better agreement with the experimental data in Fig 1 . IK denotes the delayed rectifier K+ channel [9] , ISK denotes the small-conductance K+ channel , activated by the intracellular Ca2+ concentration [9] , and IBK denotes the big-conductance K+ channel . The latter was assumed to be both voltage and Ca2+-dependent . As it is often co-localized high-voltage-activated Ca2+ channels , it was assumed to sense a domain-Ca2+ concentration proportional to ICa [32] . The depolarizing membrane currents consisted of a high-voltage activated Ca2+ current ( ICa ) and a Na+ current ( INa ) , both of which were novel for this model , and adapted to new voltage-clamp data from gonadotroph cells in medaka ( see Methods ) . INa activated in the range between −50 mV and −10 mV , with half activation at −32 mV ( black curves , Fig 2A1 ) , quite similar to what was previously found in goldfish gonadotrophs [4] . INa inactivated in the range between −90 mV and −40 mV , with half-inactivation at −64 mV , which was lower than in goldfish , where the half-inactivation was found to be around −50 mV [4] . With the activation kinetics adapted to medaka data , only 6% of INa was available at the typical resting potential of −50 mV . The fact that medaka still showed TTX-sensitive spontaneous activity thus suggests that INa is highly expressed in these cells . In comparison , in the generic murine pituitary model [32] , INa activation required depolarization to voltages far above the resting potential ( red curves , Fig 2A1 ) ) , meaning that this model could not elicit spontaneous INa-based APs . Both INa and ICa had fast activation in medaka gonadotrophs , INa being slightly faster with a time constant of about 0 . 5–0 . 8 ms in the critical voltage range ( Fig 2A2 ) , whereas ICa had a time constant >1 ms in the critical voltage range ( Fig 2B2 ) . ICa activated in the range between −40 mV and +10 mV , with a half activation at 16 mV ( red curve in Fig 2B1 ) . This activation curve was much steeper than ICa in the rat models ( colored curves ) . The high activation in medaka gonadotrophs threshold suggests that ICa is unsuitable for initiating spontaneous APs , making spontaneous activity critically dependent on INa , unlike in all rat models , where ICa was highly available around the resting potential . With the right tuning , the medaka gonadotroph model reproduced the essential firing patterns seen in the experiments ( Fig 1 ) . In control conditions , it fired sharp APs ( AP width was 6 ms ) with relatively low peak voltages ( around -10 mV ) , and had a spontaneous firing rate slightly below 1 Hz ( Fig 3A ) . AP firing was completely abolished when the Na+-conductance gNa was set to zero , mimicking the effect of TTX application in the experiments . A series of previous models have shown that IBK may act to broaden APs and promote bursting in rat pituitary cells [9 , 25 , 27 , 32] . In contrast , a high IBK expression in medaka gonadotrophs [12] does not make these cells bursty . On the contrary , the experiments in Fig 1D showed that medaka gonadotrophs fired broader APs when BK channels were blocked . This was also seen in the model simulations , when the BK-channel conductance ( gBK ) was reduced relative to its value during control ( Fig 3B–3F ) . Generally , a reduction in gBK lead to a broadening of the AP event . The resemblance with paxilline data was strongest in simulations with partial reduction , such as in Fig 3C and 3D , where gBK had been reduced to 16% and 15% of its control value , respectively . Then , AP events were about 60 ms wide , and included plateau potentials that presumably reflected an interplay between ICa and repolarizing currents activating/inactivating after the initial AP peak . When gBK was set to zero , the oscillations were not seen , and the AP was prolonged by a flatter and more enduring > 100 ms plateau . It is reasonable to assume that also in the experiments , the blockage of BK by paxilline was not complete . To explore in further detail how the various membrane mechanisms affected the AP firing , we performed a feature-based sensitivity analysis of the medaka gonadotroph model ( Fig 4 ) . We then assigned the maximum conductances of all included currents uniform distributions within intervals ±50% of their default values ( Table 1 ) , and quantified the effect that this parameter variability had on selected aspects of the model output ( see Methods ) . An exception was made for gBK , which was assigned a uniform distribution between 0 and the maximum value given in Table 1 ) , i . e . , from fully available to fully blocked , motivated by the fact that this was the possible range spanned in the paxilline experiments ( Fig 1D ) . We note the total-order Sobol sensitivity indices considered in the current analysis reflects complex interactions between several nonlinear mechanisms , and that mechanistic interpretations therefore are difficult . Below , we have still attempted to extract the main picture that emerged from the analysis . Three features of the model responses were considered: ( i ) IsBursting , ( ii ) IsRegular , and ( iii ) IsNotSpiking . Following the definition used by Tabak et al . [9] , plateau potentials of duration longer than 60 ms ( such as those in Fig 3C2–3F2 ) were defined as bursts . For simplicity , we used the definition loosely , and referred to enduring plateau potentials as bursts even in cases where they did not contain any oscillations ( such as in Fig 3C2–3F2 ) ) . APs of shorter duration than this ( such as those in Fig 3A2 and 3B2 ) were defined as regular spikes . All the features ( i-iii ) were binary , meaning e . g . , that IsBursting was equal to 1 in a given simulation if it contained one or more bursts , and equal to 0 if not . The mean value of a feature ( taken over all simulations ) then represented the fraction of simulations that possessed this feature . For example , IsBursting had a mean value of 0 . 11 , IsRegular had a mean value of 0 . 35 , and IsNotSpiking had a mean value of 0 . 55 , which means , respectively , that 11% of the model parameterizations fired bursts , 35% fired regular APs , and 55% did not fire any kind of APs . AP activity was thus seen in less than half of the parameterizations . This reflects that the default configuration had a resting potential only slightly above the AP generation threshold , so that any parameter re-sampling that would make the cell slightly less excitable , would abolish its ability for AP generation . We note that the mean values of the three features sum up to 1 . 01 and not to unity . This was because a few of the parameterizations fired both bursts and regular APs within the same simulation . The total-order Sobol indices , shown in histograms in Fig 4 , quantify how much of the variability ( between different simulations ) in the response features that are explained by the variation of the different model parameters , i . e . , the maximum conductances . When interpreting these results , we should keep in mind that the feature sensitivities are not independent , i . e . , if Isbursting equals 0 for a given implementation , it means that either IsRegular or IsNotSpiking must equal 1 . When the sensitivity to gNa was small for IsBursting , but quite large for IsRegular and IsNotspiking , it then means that gNa was important for switching the model between not firing and regular firing , while it contributed less to prolonging the APs into possible bursts . In contrast , IsNotSpiking was almost insensitive to gBK , while IsBursting and IsRegular had a high sensitivity to gBK . A little simplified , we can thus say that gNa determined whether the model fired an AP ( Fig 4C ) , while gBK determined whether the AP , if fired , became a burst or a regular spike ( Fig 4A and 4B ) . All three features had a high sensitivity to gK , which indicates that gK played multiple roles for the firing properties of the model . Firstly , IK had a nonzero activity level around rest ( cf . Fig 2C1 ) , and was important ( along with INa and the leakage current , Il ) for determining whether the resting potential was above the AP threshold , hence the high sensitivity of IsNotSpiking to gK . Having a broad activation range , IK was also important for repolarizing the membrane potential after the AP peak , and thus for the duration of the AP . Therefore , also IsBursting had a high sensitivity to gK . The mechanisms behind burst generation are reflected in the sensitivity of IsBursting to gBK , gK and gCa . Here , ICa is responsible for mediating the plateaus that prolong APs into possible bursts , while gBK and gK may prevent bursts by facilitating a faster down-stroke . The interaction between gBK and gK in mediating the AP downstroke is complex , as we comment on further in the next subsection . The sensitivity to the last K+-channel , gSK , was very low in all the features considered here . gSK had very little impact on the AP shape or the ability of the model to elicit APs , but was included in the model since it was important for regulating the firing rate . As we have seen , IBK consistently had a narrowing effect on APs in the medaka gonadotroph model , while it has previously been reported to broaden APs in several models based on data from murine pituitary cells [9 , 27 , 32] . To gain insight into this dual role , we here explore the relationship between gBK expression and AP shape in four different models , including ( i ) the medaka gonadotroph model ( Fig 5A ) , ( ii ) a previously published models of a rat lactotroph ( Fig 5B ) , ( iii ) a generic pituitary cell model based on data from rats and mice ( Fig 5C ) , and ( iv ) a model of a rat somatotroph ( Fig 5D ) . In the medaka gonadotroph model , APs were predominantly mediated by INa , while in the murine models , APs were predominantly mediated by ICa . Despite several differences , all models contained IBK and IK , which were the most important ion channels for mediating the AP downstroke . In all models , an increase in gBK consistently lead to a reduction of the AP-peak voltage ( Fig 5A1–5D1 ) , as one might expect from a hyperpolarizing current . Additional simulations on the medaka gonadotroph model revealed that the magnitude of the reduction depended strongly on the IBK-activation time constant ( τBK ) . In the default configuration , τBK was set to 3 ms ( orange curve in Fig 5A1 ) . With a slower τBK , IBK activation remained low during the AP upstroke , and its effect on the AP-peak voltage was low ( red curve in Fig 5A1 ) . Contrarily , when τBK was faster , IBK activation largely occurred during the AP upstroke , and IBK then had a larger effect on the AP-peak value ( blue curve in Fig 5A1 ) . In general , IBK could affect both the upstroke ( reducing the peak voltage ) and downstroke ( repolarizing the membrane ) of the AP , and the relative contribution to the two phases would depend on the relative timing of IBK activation and the AP peak . The effect on gBK on AP width ( Fig 5A2–5D2 ) and afterhyperpolarization ( AHP ) depth , defined simply as the minimum voltage reached between two APs ( Fig 5A3–5D3 ) , was more complex and less intuitive . To gain an insight into the mechanisms at play , we start by exploring how gBK affected the AHP depth ( Fig 5A3–5D3 ) . The AHP was predominantly due to the joint effect of the two hyperpolarizing currents , gBK and gK . It may therefore seem counterintuitive that , for low gBK , an increase in gBK actually decreased the AHP ( less hyperpolarization means higher AHP voltages ) . The explanation lies in the simultaneous effect that gBK had on reducing the AP-peak voltage ( Fig 5A1–5D1 ) . Lower AP-peak values generally meant less IK activation [9 , 25] , as this current activates at high voltage levels . Hence , gBK had a dual affect on the AHP . It could facilitate AHP through its direct , hyperpolarizing effect , but at the same time counteract AHP indirectly , by limiting gK activation . When gBK was small , the AHP was predominantly mediated by gK , and the indirect effect dominated , so that an increase in gBK reduced the AHP up to a certain point , where the direct effect to took over , so that a further increase in gBK enhanced the AHP . The dual role of gBK is also reflected in the effect that gBK had on the AP width ( Fig 5A2–5D2 ) . An increase in gBK could either lead to briefer APs , through the direct hyperpolarizing effect of IBK , or broader APs , through the indirect effect of gBK reducing IK activation . This dual role of gBK is seen most clearly in the rat lactotroph model ( Fig 5B2 ) . For low values of gBK , the direct effect dominated , and the AP width decayed monotonically with increasing gBK up to a certain threshold value , where a further increase in gBK gave a sharp transition to very broad APs ( pseudo-plateau bursts ) . The paradoxical role of IBK as a burst-promoter in the lactotroph model was explored in detail in the original study [9] , and in a later re-implementation of the model [37] . The same dual role of gBK on the AP shape was seen in the generic pituitary model , although the effect of gBK on narrowing APs for low gBK was there very small ( Fig 5C2 ) . In the rat somatotroph model , the indirect effect dominated for all values of gBK , and the AP width increased monotonically with increasing gBK ( Fig 5D2 ) . Oppositely , in the default parameterization of the medaka gonadotroph model , the direct effect dominated for all values of gBK , and the AP width decreased monotonically with increasing gBK ( orange curve , Fig 5A2 ) . Only by decreasing τBK to unrealistically low values , gBK could have a broadening effect on APs in the medaka gonadotroph model ( blue curve , inset in Fig 5A2 ) . We note that the complex interplay of mechanisms is only partly captured by the simplified , heuristic explanations presented above . In particular , for high gBK values , neither the AP width or AHP increased monotonically with gBK in all models ( Fig 5B2 and 5C3 ) . This non-monotonic behavior putatively reflects a complex and highly sensitive interplay between several mechanisms active in the aftermath of the AP peak , and we did not attempt to explore it in further detail . As we have seen , IBK facilitated bursting in all the considered models based on murine data , but not in the default parameterization of the medaka gonadotroph model . As the IBK kinetics in the medaka gonadotroph model was essentially the same as in the generic murine pituitary cell model [32] , we hypothesized that the different role played by IBK in the medaka gonadotroph model versus the murine pituitary cell models was due to differences in AP shape , rather than IBK kinetics . By comparing the AP upstrokes of the different models , we see that the fastest upstroke was found in the medaka gonadotroph model where APs were predominantly INa mediated ( blue curve in Fig 6 ) . The second fastest upstroke was seen in the generic murine model in the case where its APs were mediated by a combination of ICa and INa ( purple curve in Fig 6 ) . Hence , the addition of INa to this model made the AP upstroke steeper , and , as we saw in Fig 5C2 , this made the model less susceptible to bursting , i . e . , the transition to bursting occurred for a much higher value of gBK . In the remaining models , APs were mediated solely by ICa , and had slower upstroke . Hence , in the murine models , IBK had more time to activate during the AP upstrokes , which explains why IBK indirectly could promote bursting in these models , by reducing AP-peak value and thereby IK activation . In order to have the same effect in the medaka gonadotroph model , τBK needed to be speeded up dramatically , as illustrated in the blue curve in Fig 5A2 . This scenario is hypothetical , as no experiments suggest that IBK does promote bursting in medaka gonadotrophs . However , it is interesting to note that this parameterization of the model fired bursts ( i . e . , APs with width > 60 ms ) both for very low and very high values for gBK , while it fired regular AP for intermediate gBK value . Also a previous model of a rat corticotroph showed such bursting behaviour in two disjoint regions in gBK-space ( see Fig 3 in [31] ) . In summary , IBK had an inhibitory effect on IK by reducing the AP amplitude , and a collaborative effect with IK in mediating the AP downstroke . With slow AP upstrokes , as in the murine pituitary cell models , the inhibitory effect of IBK on IK could dominate , and IBK could result in a net reduction of hyperpolarization and as such promote broader APs and sometimes bursts . With faster AP upstrokes , such as in the medaka gonadotroph models , the collaborative effect of IBK always dominated , and IBK consistently facilitated narrower APs . In a broader scope , this suggests that IBK can act as a mechanism that reduces the duration of already fast APs , and prolongs the duration of already slow APs .
TTX-sensitive Na+ currents ( INa ) are present in all pituitary cells , but are in many cases inactive during spontaneous activity [8] . Previous models of the electrical activity of pituitary cells have focused on conditions where INa is of lesser importance , and where AP generation is predominantly mediated by high-voltage activated Ca2+ currents [3 , 9 , 23 , 25–28] . To our knowledge , we have in the current work presented the first models that describe pituitary cells under conditions where AP generation is INa-mediated . The model was adapted to experimental data from LH-producing gonadotrophs in medaka , whose spontaneous activity is highly INa-dependent . Voltage-clamp data was used to develop models for the activation kinetics for INa and ICa currents , and the firing properties of the model were further adapted to current-clamp data from spontaneously active cells ( under control conditions , and after application of TTX and paxilline ) . To examine the consequences of having different AP generation mechanisms , we performed a comparison between the the medaka gonadotroph model , which fired INa-mediated APs , and three models of murine pituitary cells which fired APs that were exclusively mediated by ICa [9 , 27] , or by a combination of ICa and INa [32] . The most interesting result that came out of this comparison was that IBK had a dual role on AP shape , and could under some conditions broaden APs and promote bursting , and under other conditions make them narrower . We suggested that the broadening effect could only occur in scenarios where IBK had sufficient time to activate during the AP upstroke , and thus required either a very fast IBK activation time constant , or a relatively slow AP upstroke . In the murine models , the AP upstrokes were slower than in the medaka gonadotroph model , and we suggest that this explains why increased IBK can promote bursting in many murine pituitary cells [8 , 9 , 25 , 32] , but not in medaka gonadotrophs [12] . Also other K+ channels have been shown to have such a burst-promoting role in murine pituitary cells [38 , 39] . It should be noted that the effect on reducing AP width is a commonly reported role for IBK in many excitable cells [40–44] , while the AP-broadening and burst promoting effect that IBK is less conventional . The role of IBK as a burst promoter has not been found consistently in rat lactotrophes . In the study by Miranda et al . 2003 , AP width in rat GH3 , a widely used model for pituitary lactotrophs , was instead found to increase when IBK was blocked with paxilline [43] , i . e . , similar to what we found for medaka gonadotrophs ( Fig 1D ) . The different effects of IBK on AP width observed in different laboratories [25 , 40 , 43] was addressed by Tabak et al . 2011 [9] , who proposed possible explanations that could reconcile the conflicting results . One possible explanation could be there is a variability in terms of how BK channels are localized in various cells , and that BK channels that are co-localized with Ca2+ channels will respond rapidly to voltage fluctuations and promote bursting , while BK channels that are not co-localized with Ca2+ channels will react more slowly to voltage fluctuations and have the opposite effect [9] . A second possible explanation , also suggested by Tabak et al . 2011 , was that IBK might have different kinetic properties in different cells due to variations in their phosphorylation state [9] . A third explanation could be that different cells have different BK splice variants [45] , or different regulatory sub-units . The model comparison in Fig 5 provides additional possible explanations to the conflicting conclusions regarding the role of IBK in lactotrophs . Firstly , the fact that IBK has affects the AP shape differently in different cells does not by necessity reflect differences in IBK kinetics or localization . Simulations on the rat lactotroph model showed that the same IBK could both have a broadening and narrowing effect on the APs within one and the same model ( Fig 5B ) . That is , APs could be made broader either by reducing gBK to very low values , or by increasing it to very high values . This dual effect of IBK was even more pronounced in a version of the medaka gonadotroph model ( blue curve in Fig 5B ) , and a previous model of rat corticotrophs [31] , which elicited bursts both under full gBK blockage and for large gBK expression , while they fired regular APs for intermediate gBK expression . Hence , in general , gBK blockage could affect the AP width in either way , depending on the initial level of gBK expression , and conflicting conclusions regarding the role of IBK could reflect that variations in gBK expression under control conditions . Secondly , the way on which IBK will affect the APs in a given cell can not be predicted from gBK kinetics/expression alone , but also depends on the AP generating mechanisms in the cell . Our simulations suggested that APs with a steep upstroke were prone to be made briefer by IBK , while APs with a slower upstroke were prone to be prolonged by IBK , as also suggested in the dynamic clamp experiments by Tabak et al . [9] . Putatively , INa mediated APs will generally have a steeper upstroke than ICa mediated APs , as was the case in the models studied here . If this is the case , the role of IBK in a given cell may thus largely be determined by which membrane mechanisms that mediate its AP upstroke , and especially the degree to which INa is involved , which is highly resting-potential dependent , and likely to depend strongly on experimental conditions . Differences in INa involvement could in principle explain the conflicting experiments on rat lactotrophs [25 , 43] . In the experiment by Van Goor et al . 2001 , where IBK was found to broaden APs , APs were predominantly mediated by ICa [9 , 25] . In the experiment by Miranda et al . 2003 , where IBK was found to narrow APs ( i . e . , blocking IBK lead to broader APs ) , it was reported that this only occurred under conditions in which short APs were present . It is likely that the events described in that work as short APs were largely INa-mediated , so that the differences between the studies by Van Goor et al . 2011 and Miranda et al . 2003 suggest that different AP generation mechanisms dominated in the two experiments . Although the medaka gonadotroph model captured the essential firing properties of medaka gonadotrophs , the agreement between model and data was not perfect . For example , the AP width during control conditions ( Fig 3A2 ) was in the upper range of that seen in the experiments , while AP peak voltage was in the lower range of what was seen in the experiments ( Fig 1A and 1B ) . We were not able to obtain briefer APs with larger peak values without compromising the agreement between the experimental data and other model features , such as afterhyperpolarization , firing rate and response to IBK-blockage . The conductances selected for the default model ( Table 1 ) were thus a compromise made to obtain an acceptable match to several features simultaneously . The fact that we were not able to obtain a more accurate match between model and data likely reflect that some of the ion channels present in the model are imperfect representations of the ion channels present in the real cell . For example , the simplified kinetics schemes used for IK , IBK and ISK were adopted from models of rat pituitary cells [9 , 32] , and were not constrained to data from medaka gonadotrophs . In addition , the biological cell is likely to contain a variety of additional ion channels that were not included in the model . To our knowledge , the medaka gonadotroph model is the first computational model of an endocrine cell that fires APs that are predominantly INa-based . Although it was adapted to experimental recordings from LH-producing gonadotrophs in medaka , we believe that the model has a more general value . Different types of pituitary cells in several different species share many of the same membrane mechanisms [8] . In particular , INa-based APs are elicited by several pituitary endocrine cell types and in several animal species , depending on biological conditions [4 , 7 , 17 , 18 , 33–35] . It is thus likely that the response patterns of related cell types may be captured by up- or down-regulation of selected mechanisms included in medaka gonadotroph model . Insight in which parameters that should be adjusted in order to obtain desired changes in the model’s firing properties could then be obtained through a sensitivity analysis , such as that presented in Fig 4 , or in previous , more comprehensive studies that consider a larger number of model features [37 , 39] .
The electrophysiological experiments were conducted using the patch-clamp technique on brain-pituitary slices from adult female medaka ( as described in [46] ) . To record spontaneous action potentials and Ca2+ currents we used amphotericin B perforated patch configuration , while for Na+ currents we used whole cell configuration . Extracellular ( EC ) solution used for recording spontaneous action potentials ( current clamp ) contained 134 mM NaCl , 2 . 9 mM KCl , 2 mM CaCl2 , 1 . 2 mM MgCl2 1 . 2 , 10 mM HEPES , 4 . 5 mM glucose . The solution was adjusted to a pH of 7 . 75 with NaOH and osmolarity adjusted to 290 mOsm with mannitol before sterile filtration . Before use , the EC solution was added 0 . 1% bovine serum albumin ( BSA ) . For Na+ current recordings ( voltage clamp ) we used a Ca2+ free and Na+ fixed ( 140 mM ) EC solution , pH adjusted with trizma base . In addition , 10 μM nifedipine , 2 mM 4-Aminopyridine ( 4-AP ) and 4 mM Tetraethylammonium ( TEA ) was added to the EC solution just before the experiments . To record Ca2+ currents , we substituted NaCl with 120 mM choline-Cl and added 20 mM Ca2+ , 2 mM 4-AP and 4 mM TEA . The patch pipettes were made from thick-walled borosilicate glass using a horizontal puller ( P 100 from Sutter Instruments ) . The resistance of the patch pipettes was 4-5 MΩ for perforated patch recordings and 6-7 MΩ for whole-cell recordings . For recordings of spontaneous action potentials , the following intracellular ( IC ) electrode solution was added to the patch pipette: 120 mM KOH , 20 mM KCl , 10 mM HEPES , 20 mM Sucrose , and 0 . 2 mM EGTA . The pH was adjusted to 7 . 2 using C6H13NO4S ( mes ) acid , and the osmolality to 280 mOsm using sucrose . The solution was added 0 . 24 mg/ml amphotericin B to perforate the cell membrane ( see [46] for details ) . In voltage clamp experiments the K+ was removed from the intracellular solution to isolate Na+ and Ca2+ currents . This was achieved by substituting KOH and KCl with 130 mM Cs-mes titrated to pH 7 . 2 with CsOH . The electrode was coupled to a Multiclamp 700B amplifier ( Molecular Devices ) and recorded signal was digitized ( Digidata 1550 with humsilencer , Molecular Devices ) at 10 KHz and filtered at one-third of the sampling rate . In selected experiments , voltage-gated Na+ channels were blocked using 5 μM TTX , and BK channels were blocked using 5 μM paxilline . Both drugs were dissolved in EC solution and applied using 20 kPa puff ejection through a 2 MΩ pipette , 30-40 μm from the target cell . Under the experimental ( voltage-clamp ) conditions used for recording Na+ currents , and under the experimental current-clamp conditions , a liquid junction potential of about −9 mV was calculated and corrected for in the data shown in Fig 1 , and in the kinetics model for INa ( Fig 2A ) . A liquid junction potential of about −15 mV was calculated for the experimental ( voltage-clamp ) conditions used for recording Ca2+ currents , and was corrected for in the kinetics model for I C a m ( Fig 2B ) . Handling , husbandry and use of fish were in accordance with the guidelines and requirements for the care and welfare of research animals of the Norwegian Animal Health Authority and of the Norwegian University of Life Sciences . As stated in the Results-section , the medaka gonadotroph model was described by the equation: C m d V d t = - ( I N a + I C a + I K + I B K + I S K + I l e a k ) . ( 2 ) The membrane capacitance was set to the standard value Cm = 1μF/cm2 , and the leak conductance was described by I l e a k = g l e a k ( V - E l e a k ) . ( 3 ) with a reversal potential Eleak = −45 mV . Due to a nonzero-activation of IK around the resting level , this gave an effective resting potential around −50 mV , similar to that in the experimental data in Fig 1 . The kinetics of all ion channels were summarized in Fig 2 , but are described in further detail here . INa was modeled using the standard Hodgkin and Huxley-form [47]: I N a = g N a q 3 h ( V - E N a ) , ( 4 ) with a reversal potential ENa = 50 mV , and gating kinetics defined by: d q d t = q ∞ - q τ q , d h d t = h ∞ - h τ h . ( 5 ) The steady-state activation and time constants ( q∞ , h∞ , τq and τh ) were fitted to voltage-clamp data from medaka gonadotrophs , as described below , in the subsection titled “Model for the voltage-gated Na+ channels” . ICa was modelled using the Goldman-Hodgkin-Katz formalism , which accounts for dynamics effect on Ca2+ reversal potentials [48]: I C a m = g C a m 2 F 2 R T V [ C a ] - [ C a ] e exp ( - V F R T ) 1 - exp ( - V F R T ) , ( 6 ) with d m d t = m ∞ - m τ m ( 7 ) Here , R = 8 . 314J/ ( mol ⋅ K is the gas constant , F = 96485 . 3C/mol is the Faraday constant , T is the temperature , which was set to 293 . 15 K in all simulations . [Ca] and [Ca]e were the cytosolic and extracellular Ca2+ concentrations , respectively . The former was explicitly modelled ( see below ) , while the latter was assumed to be constant at 2 mM . As Eq 6 shows , we used two activation gates m . This is typical for models of L-type Ca2+ channels ( see e . g . [49–52] ) , which are the most abundantly expressed HVA channels in the cells studied here [11] . The steady-state activation and time constant ( m∞ and τh ) were fitted to voltage-clamp data from medaka gonadotrophs , as described below , in the subsection titled “Model for high-voltage activated Ca2+ channels” . We note that gCa in the Goldman-Hodgkin-Katz formalism ( Eq 6 ) is not a conductance , but a permeability with units cm/s . It is proportional to the conductance , and for simplicity , we have referred to it as a conductance in the text . The delayed rectifyer K+ channel was modelled as I K = g K n ( V - E K ) , ( 8 ) with reversal potential EK = −75 mV , and a time dependent activation gate described by d n d t = n ∞ - n τ K . ( 9 ) The steady-state activation was described by: n ∞ = [ 1 + e x p ( ( v n - V ) / s n ) ] - 1 . ( 10 ) with a slope parameter sn = 10 mV , and half-activation vn = −5 mV . The model for IK was identical to that in a previous rat lactotroph model [9] , with the exception that the constant ( voltage-independent ) time constant τK was made faster ( 5 ms ) in the medaka gonadotroph model to account for the more rapid APs elicited by these cells . The model for the BK-channel kinetics was a was taken from a previous model ( where it was called BK-near ) of murine corticotrophs [29 , 30] , which has also been used in a generic rat pituitary cell model [32]: I B K = g B K f ( V - E K ) . ( 11 ) The activation kinetics was given by: d f d t = f ∞ - f τ B K , ( 12 ) The constant ( voltage-independent ) activation time constant τBK was set to 3 ms . The steady-state activation was given by [29]: f ∞ = [ 1 + e x p ( ( v f - V ) / 3 ) ] - 1 , ( 13 ) with v f = 0 . 1 - 18 · log ( c d o m / c r e f ) ( 14 ) As BK channels are often co-localized with high-voltage activated Ca2+ channels , BK activation was assumed to depend on a domain concentration cdom in Ca2+ nanodomains , which in turn was assumed to be proportional to the instantaneous Ca2+ influx through ICa . We therefore set: c d o m = - A I C a , ( 15 ) where A = 1 . 21 mmol⋅cm−1⋅C−1 is a parameter that converts a current density into a concentration , and cref = 2μM is a reference concentration . The parameter A was not taken from previous studies , but tuned so that the model obtained suitable BK activation in simulations on the medaka gonadotroph model . Finally , the SK channel was the same as in the previous model of a rat lactotroph [9] , and was modelled as: I S K = g S K s ∞ ( [ C a ] ) ( V - E K ) , ( 16 ) with an instantaneous , Ca2+ dependent , steady-state activation: s ∞ ( [ C a ] ) = [ C a ] 2 [ C a ] 2 + k s 2 ( 17 ) where [Ca] denotes the cytosolic Ca2+ concentration , and ks is a half-activation concentration of 0 . 4 μM . ICa and ISK were dependent on the global cytosolic Ca2+ concentration . This was modelled as a simple extrusion mechanism , receiving a source through ICa , and with a concentration dependent decay term assumed to capture the effects of various ion pumps and buffering mechanisms: d [ C a ] d t = - f c ( α I C a + k c [ C a ] ) . ( 18 ) Here , fc = 0 . 01 is the assumed fraction of free Ca2+ in the cytoplasm , α = 0 . 015mM ⋅ cm2/μC converts an incoming current to a molar concentration , and kc = 0 . 12ms−1 is the extrusion rate [9] . The conductances used in the default parameterization of the model are given in Table 1 . The steady-state values and time courses of the gating kinetics were determined using standard procedures ( see e . g . [35 , 47 , 53 , 54] ) , and was based on the experiments summarized in Fig 7 . To determine activation , the cell was held at −60 mV for an endured period , and then stepped to different holding potentials between −80 to 100 mV with 5 mV increments ( Fig 7A2 ) , each for which the response current ( INa ) was recorded ( Fig 7A1 ) . The inactivation properties of Na+ were investigated using stepwise pre-pulses ( for 500 ms ) between −90 and 55 mV with 5 mV increments before recording the current at −10 mV ( Fig 7B2 ) . The resulting Na+ current then depended on the original holding potential ( Fig 7B1 ) . Finally , the recovery time for the Na+ current was explored by exposing the cell to a pair of square pulses ( stepping from a holding potential of −60 mV to −10 mV for 10 ms ) separated by a time interval Δt ( Fig 7C2 ) . The smallest Δt was 10 ms , and after this , Δt was increased with 100 ms in each trial . The cell responded to both pulses by eliciting Na+-current spikes ( Fig 7C2 ) . When Δt was small , the peak voltage of the secondary spike was significantly reduced compared to the first spike , and a full recovery required a Δt in the order of 1/2-1 s . When estimating the steady-state values and time constant we followed procedures inspired from previous studies of L-type Ca2+ channel activation , we did not use Eq 6 , but used the simpler kinetics scheme ICa = gHV Am2 ( V − ECa ) ( see e . g . [50] ) ) assuming a constant reversal potential . | Excitable cells elicit electrical pulses called action potentials ( APs ) , which are generated and shaped by a combination of ion channels in the cell membrane . Since one type of ion channels is permeable to Ca2+ ions , there is typically an influx of Ca2+ during an AP . Pituitary cells therefore use AP firing to regulate their cytosolic Ca2+ concentration , which in turn controls their hormone secretion rate . The amount of Ca2+ that enters during an AP depends strongly on how long it lasts , and it is therefore important to understand the mechanisms that control this . Pituitary APs are generally mediated by a combination of Ca2+ channels and Na+ channels , and the relative contributions of from the two vary between cell types , animal species and biological conditions . Previous computer models have predominantly been adapted to data from pituitary cells which tend to fire Ca2+-based APs . Here we develop a new model , adapted to data from pituitary cells in the fish medaka , which APs that are predominantly Na+-based , and compare its dynamical properties to the previous models that fire Ca2+-based APs . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"medicine",
"and",
"health",
"sciences",
"action",
"potentials",
"fish",
"goldfish",
"nervous",
"system",
"membrane",
"potential",
"vertebrates",
"electrophysiology",
"neuroscience",
"animals",
"simulation",
"and",
"modeling",
"osteichthyes",
"animal",
"models",
"ion",
"channels",
"model",
"organisms",
"experimental",
"organism",
"systems",
"endocrine",
"cells",
"research",
"and",
"analysis",
"methods",
"animal",
"cells",
"pituitary",
"gland",
"proteins",
"animal",
"studies",
"mouse",
"models",
"biophysics",
"physics",
"biochemistry",
"eukaryota",
"neuroanatomy",
"anatomy",
"cell",
"biology",
"physiology",
"endocrine",
"system",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"physical",
"sciences",
"neurophysiology",
"organisms"
] | 2019 | A computational model for gonadotropin releasing cells in the teleost fish medaka |
Cutaneous leishmaniasis ( CL ) is treated with parenteral drugs for decades with decreasing rate cures . Miltefosine is an oral medication with anti-leishmania activity and may increase the cure rates and improve compliance . This study is a randomized , open-label , controlled clinical trial aimed to evaluate the efficacy and safety of miltefosine versus pentavalent antimony ( Sbv ) in the treatment of patients with CL caused by Leishmania braziliensis in Bahia , Brazil . A total of 90 patients were enrolled in the trial; 60 were assigned to receive miltefosine and 30 to receive Sbv . Six months after treatment , in the intention-to-treat analyses , the definitive cure rate was 53 . 3% in the Sbv group and 75% in the miltefosine group ( difference of 21 . 7% , 95% CI 0 . 08% to 42 . 7% , p = 0 . 04 ) . Miltefosine was more effective than Sbv in the age group of 13–65 years-old compared to 2–12 years-old group ( 78 . 9% versus 45% p = 0 . 02; 68 . 2% versus 70% p = 1 . 0 , respectively ) . The incidence of adverse events was similar in the Sbv and miltefosine groups ( 76 . 7% vs . 78 . 3% ) . Vomiting ( 41 . 7% ) , nausea ( 40% ) , and abdominal pain ( 23 . 3% ) were significantly more frequent in the miltefosine group while arthralgias ( 20 . 7% ) , mialgias ( 20 . 7% ) and fever ( 23 . 3% ) were significantly more frequent in the Sbv group . This study demonstrates that miltefosine therapy is more effective than standard Sbv and safe for the treatment of CL caused by Leishmania braziliensis in Bahia , Brazil . Clinicaltrials . gov Identifier NCT00600548
The principal species causing cutaneous leishmaniasis ( CL ) in Brazil is Leishmania braziliensis which most often leads to a cutaneous form of the disease characterized by one or more ulcers with raised borders , most frequently located on the upper and lower extremities , but also found on the head , face , and trunk [1] , [2] . Although CL is a self-limited disease , approximately 3 to 5% of subjects infected with L . braziliensis will eventually develop mucosal disease or disseminated leishmaniasis , both considered severe forms of leishmaniasis [2]–[4] . Pentavalent antimony ( Sbv ) by intramuscular or intravenous route remains the first-line drug for the treatment of CL , a therapy that is moderately toxic and difficult to administer in poor rural areas . In an endemic area of L . braziliensis transmission in Bahia , Brazil , cure rates after Sbv therapy are becoming increasingly lower and vary from 50% to 90% [5]–[7] . Factors contributing to this variability are not yet fully understood . The development of parasite resistance [8] to a drug used for decades and irregular adherence by the patients due to the daily schedule of parenteral route during 20 days , could be main factors in determining the increasingly failure rate to Sbv [5]–[7] . Other alternative drugs like pentamidine and amphothericin B are also of parenteral use and the former may require hospitalization . Miltefosine , a phosphatidylcholine analogue , is an active antileishmanial oral drug used for the treatment of visceral leishmaniasis in India [9] . The clinical efficacy of miltefosine for New World CL was investigated in trials conducted in Central and South America . The cure rates varied from one leishmania species to another , with L . panamensis having 82% of cure rate , L . mexicana 60% and L braziliensis 33% respectively [10] . More recently in CL caused by L . braziliensis in patients from Bolivia , Soto et al reported a 88% cure rate [11] . These data shows that CL cure rate upon miltefosine treatment varies among Leishmania species and also between the same species from different endemic areas . Accordingly , it is demonstrated that different treatment outcome after antimony therapy is found in CL caused by L . braziliensis from different endemic regions in Brazil [5] , [6] , [12] . There is no data about the use of miltefosine in CL caused by L . braziliensis in Brazil , where this species is considered the most aggressive and prevalent agent of the disease [13] . This randomized and controlled trial evaluated the efficacy and safety of miltefosine versus meglumine antimoniate ( Sbv ) for the treatment of CL in a L . braziliensis endemic rural area in Bahia , Brazil .
This trial has been conducted according to the principles expressed in the Declaration of Helsinki . Prior to enrollment in the study , all patients received a written copy of study policy which was reviewed with them individually by an independent party . A written informed consent was obtained for all adult patients , and from parents or guardians of minors . This study was approved by the Ethics Committee of the Federal University of Bahia , in Salvador , Brazil ( CEP/MCO/UFBA-Par/Res 034/2007 ) . All subjects were recruited at the health post of Corte de Pedra , located 260 km southeast of Salvador , the capital of Bahia , Brazil . This clinic is a referral center for the diagnosis and treatment of CL , with an average of 1 , 000 new cases per year . L . braziliensis has been the unique causal agent identified in this area in the last 15 years [14] . The criteria used for the diagnosis of CL were the presence of a typical ulcerated lesion and a positive Montenegro intradermal skin test in a subject living in the endemic area . A typical CL ulcerated lesion is characterized by a round shape and raised borders associated with regional adenopathy . This classical clinical picture together with a positive leishmania skin test is highly specific for CL in the endemic area [2] , [5] . Patients were then selected based on the following inclusion criteria: 1 ) age between 2 and 65 years; 2 ) a maximum of 5 ulcers with no more than 2 body regions involved; 3 ) lesion size between 10 and 50 mm in a single dimension; and 4 ) a period of less than 90 days from the onset of the first ulcer . All subjects were submitted to a punch biopsy to obtain material for leishmania culture and PCR . Patients with a prior history of CL or antimony use , patients with evidence of mucosal or disseminated disease , pregnant or breastfeeding mothers , and patients with HIV or any systemic severe disease were excluded . A total of 90 patients were enrolled in the study . A randomization table was obtained with Statacorp LP 9 , Texas USA . Group assignments were made after assessment that patient had met all eligibility criteria and no patients were withdrawn after randomization because of ineligibility . The 90 patients were randomly assigned at a rate of 2∶1 allocation to receive miltefosine for 28 days or Sbv for 20 days in two age groups ( 2–12 years-old and 13–65 years-old ) . The Leishmania antigen used for intradermal skin testing was obtained from Leishmania amazonensis strain ( MHOM-BR-86BA-125 ) . The volar surface of the left forearm was injected with 25 µg of antigen in 0 . 1 mL of distilled water , and the largest diameter of induration was measured at 48–72 hours . The test was considered positive for induration greater than 5 mm . DNA isolation was carried out from biopsy samples using the Wizard Genomic DNA purification kit ( Promega Corporation – USA ) . Purified DNA was resuspended in TE buffer and stored frozen at −20°C until use . Detection of the subgenus Viannia applied the primers 5′-GGGGTTGGTGTAATATAGTGG-3′ and 5′-CTAATTGTGCACG-3′ . For Leishmania genus detection we have used the primers 5′- ( G/C ) ( G/C ) ( C/G ) CC ( A/C ) CTAT ( A/T ) TTA CAC CAA CCC C-3′ and 5′-GGG GAG GGG CGT-3′ . Amplification mixes consisted of 25 pmol each primer; 1 . 2 mM MgCl2; 0 . 2 mM dNTP; 2 . 5 U Taq DNA polymerase; 10× PCR buffer; 2 uL of target DNA . Amplifications were carried out in a Veriti 96-well thermal cycler ( Applied Biosystems – USA ) . Viannia detection applied 35 cycles of 1 minute at 94°C , 1 minute at 60°C and 1 minute at 72°C . Amplicons were fractionated in 1 . 3% agarose gels , stained with ethidium bromide and photographed under UV light using a UVP Vision Works LS apparatus ( UVP – USA ) . The Leishmania specific band consists of 120 base pairs , and that for Viannia of 750 base pairs . All study volunteers were treated as outpatients . Miltefosine ( Impavido , Zentaris GmbH ) was supplied in blisters containing 10 mg or 50 mg capsules . Meglumine antimoniate ( Glucantime , Aventis ) was supplied in vials of 5 ml containing 81 mg/Sbv/ml . Miltefosine was administered orally at the total target daily dosage of 2 . 5 mg/kg of body weight ( maximum daily dose of 150 mg ) for 28 consecutive days . Daily dose was divided in two or three intakes , given always with meals according to the following weight scale: patients with ≥15 kg and ≤29 kg - total dose of 50 mg/day; patients with ≥30 kg and ≤45 kg - total dose of 100 mg/day; patients with ≥46 kg - total dose of 150 mg/day . Meglumine antimoniate ( Sbv ) was administered intravenously at a dose of 20 mg Sbv/kg/day for 20 consecutive days ( maximum daily dose of 3 ampoules or 1215 mg/Sbv ) . At every weekly visit patients returned the blisters for verification of regular use and adherence . Sbv was administered daily in health posts near from patient's home . In these cases the administration's date and dosage were registered and signed by the health care provider . All women in child bearing age were submitted to beta HCG test to exclude pregnancy . The use of a parenteral contraceptive during and for 2 months after treatment was done in all women in child bearing age . Complete hemogram , aminotransferases ( AST , ALT ) , alkaline phosphatase , potassium , sodium , urea , creatinine , and urine chemistry were determined in all patients on day -1 , weekly thereafter up to the end of therapy , and after 15 and 30 days of the end of therapy . Those with abnormal parameters were followed-up until normalization . At each weekly return for drug dispensation patients were monitored for adverse events ( AE ) . Bidirectional measurements of ulcers were taken of the patients' lesions at the initial visit , and at each follow-up visit with standardized caliper . The area involved was calculated as the product of the two measurements . A standardized digital photograph was also taken from each patient's lesions at the same time points . Patients were seen for follow-up at 2 weeks , 1 , 2 , 4 and 6 months post-therapy . In the event that a patient did not return for follow-up at the specified time , visits were conducted in the patient's home in the same day or within 7 days of the missed appointment . The results are presented as proportions , interquartile ranges ( IQR ) , 95% confidence intervals ( 95% CI ) , means and standard-deviations ( SD ) . The normally distributed variables were compared using the t test . The proportions were compared with the Chi-square or Fisher test when appropriate . The sample size of 90 patients was obtained by calculating the number of participants needed for 80% power ( β = 0 . 2 ) to detect an absolute difference as large as 25% in the rate of cure between the two treatment groups with a statistical significance of 5% ( α = 0 . 05 ) . The intention-to-treat ( ITT ) analysis was used to calculate the cure rates . All statistical analyses were performed with the software SPSS 9 . 0 for Windows . A value of p<0 . 05 established the level of statistical significance .
Two months after the end of the treatment , 60% ( 18/30 ) of patients in the Sbv group had cured lesions , compared with 81 . 7% ( 49/60 ) in the miltefosine group . The cure rates at 6 months of follow-up were 53 . 3% ( 16/30 ) in the Sbv group and 75% ( 45/60 ) in the miltefosine group ( p = 0 . 04 ) ( Table 2 ) . The absolute difference was 21 . 7% ( 95% CI 0 . 08% to 42 . 7% ) . The mean time to cure was 75±15 days in the Sbv group and 73±17 days in the miltefosine group ( p = 0 . 63 ) . In age groups 2–12 years-old and 13–65 years-old cure rates at 6 months for miltefosine and Sbv were 68 . 2% ( 15/22 ) versus 70% ( 7/10 ) ( p = 1 . 0 ) , and 78 . 9% ( 30/38 ) versus 45% ( 9/20 ) ( p = 0 . 02 ) , respectively . Two patients ( one from miltefosine group and the other from Sbv group ) were lost for follow-up after the end of the treatment . One patient in the miltefosine group was excluded by irregular use of the medication . Relapses occurred in both groups ( two patients treated with Sbv and four patients treated with miltefosine ) between the 2 months and 6 months after the end of the treatment . In four patients ( miltefosine = 3 and Sbv = 1 ) there was a reactivation of the ulcer and the two others ( miltefosine = 1 and Sbv = 1 ) had reactivation of the infiltration in the borders of the healed ulcer . To make sure that the absence of leishmania identification did not influence the outcome , we compared the 6 month cure rate for patients with negative culture or PCR versus those with positive culture or PCR . In the miltefosine group , the cure rate of patients with positive culture or PCR ( n = 41 ) was 75 . 6% compared to 73 . 7% in the patients with negative culture or PCR ( n = 19 ) . In the Sbv group , patients with positive culture or PCR ( n = 23 ) had 47 . 8% of cure rate compared to 71 . 4% in the patients with negative culture or PCR ( n = 7 ) . However , the number of Sbv patients without parasitological confirmation is very low to allow any conclusion . Indeed if we exclude all parasitological negative patients from the analysis of the primary outcome , the cure rate in the miltefosine group ( 75 . 6% ) is higher as compared to Sbv group ( 47 . 8% ) . The frequency of AE was similar in the Sbv and miltefosine groups ( 76 . 7% vs . 78 . 3% , p = 0 . 86 ) but they were reported more commonly in patients ≥13 years-old than those <13 years-old ( 88 . 3% vs . 66 . 7% , p = 0 . 07 ) . Although the detection of AE was similar , the type of side effects varied widely between the treatment arms ( Table 3 ) . The AE that were significantly more frequent in the miltefosine group were vomiting ( 41 . 7% ) , nausea ( 40% ) and abdominal pain ( 23 . 3% ) . In the Sbv group , arthralgias ( 20 . 7% ) , mialgias ( 20 . 7% ) and fever ( 23 . 3% ) were significantly more frequent than in the miltefosine group ( Table 3 ) . Others common AE were diarrhea ( 10% of miltefosine patients ) and headache ( 43% of Sbv patients ) . None of the reported AE required complete discontinuation of therapy in any patient . In the miltefosine group , one patient presented CTC grade 3 vomiting ( 6 episodes in 24 hrs but did not require IV fluids ) and other patient grade 3 diarrhea ( increase of ≥7 stools per day over baseline and interfering with activities of daily living , but did not require hospitalization ) . Both were able to continue the treatment with miltefosine after a period of 3 to 5 days of interruption and oral fluid supplementation . One patient in the Sbv group presented one episode of CTC grade 3 urticaria at the end of therapy . This patient needed to use oral antihistaminic for 7 days with total regression of urticaria . Mild and transient raised liver enzymes was detected in less than 5% of patients in both groups . No raised levels of urea and creatinine or sodium and potassium abnormalities were detected ( data not shown ) .
The treatment of CL caused by L . braziliensis with Sbv in rural endemic areas in Brazil has been associated with decreasing cure rates , in a setting where the parenteral route can be implicated with unsatisfactory adherence . Additionally , the monotherapy of CL with Sbv after several decades may induce Leishmania resistance . It also has been shown that CL caused by L . braziliensis has higher therapeutic failure when compared to CL due to other Leishmania species [15] . Therefore the development of new therapeutic strategies associated to a better patient compliance and higher efficacy is needed to a better control of CL . Our study is the first to evaluate the efficacy of miltefosine in CL caused by L . braziliensis in adults and children in Brazil . Our data shows that a better therapeutic outcome is found after miltefosine treatment with 75% of cure , compared to 53% in the Sbv group irrespective of the age group . While we found no difference between Sbv ( 70% ) and miltefosine ( 68% ) cure rates in children , we observed that in patients above 12 years-old the superiority of miltefosine is higher , with 79% of cure compared to 45% with Sbv treatment . The reasons for miltefosine be more effective than antimony on adults and have similar efficacy in children is not completely understood . A decrease in the efficacy of miltefosine in children could be explained by the difficulty to ingest the capsules and also by differences in the biodisposition of the drug in children . However , the efficacy of miltefosine in children and adults was similar in the present study , and the difference found in cure rates of miltefosine and Sbv was due to the high rate of antimony failure in the adult population . It can not be ruled out that no difference between the two drugs was documented in the children population due to the small number of children in the study , and consequently lack of power to detect the differences . Alternatively the genetic polymorphism of L . braziliensis could explain the decreased susceptibility to Sbv in isolates infecting the adult population . We have shown that L . braziliensis is polymorphic in the endemic region of Corte de Pedra [14] , and severe forms of the disease with lower response to antimony therapy are observed predominantly in the adult population with agricultural labor [3] , [4] , [16] . For instance , in disseminated leishmaniasis less than 40% of patients are cured with one Sbv course [4] , and atypical CL failure to Sbv treatment is documented in up to 95% of the patients [16] . Although the use of miltefosine is associated with several AE ( as well as Sbv therapy ) , we reported no SAE or biochemical abnormalities that caused treatment to be abandoned . Miltefosine is a safe and effective oral treatment of CL caused by L . braziliensis in Bahia , Brazil and should be regarded as an option for CL therapy in rural endemic areas of L . braziliensis transmission in Brazil . | Cutaneous leishmaniasis ( CL ) is characterized by skin ulcerations and occurs in rural poor areas of developing countries . It is treated with daily injections of antimony for 20 days , which is associated with irregular use and increasingly lower cure rates . Miltefosine is an oral medication with activity against the agent of CL ( Leishmania ) . We have studied the efficacy and safety of miltefosine compared with antimony in patients with CL caused by Leishmania braziliensis in Bahia , Brazil . A total of 90 patients participated; 60 received miltefosine and 30 were treated with antimony . Six months after treatment , 75% of patients treated with miltefosine were cured , compared with 53% of the patients in the antimony group , a difference considered significant ( p = 0 . 04 ) . We also found that miltefosine was more effective than antimony in adults than in children . The incidence of side effects was similar with both drugs ( 76 . 7% vs . 78 . 3% ) , but all patients were able to finish the treatments . Our study shows that miltefosine is more effective than antimony for the treatment of CL in Bahia , Brazil and can contribute to the control of this disease due to its activity and easier administration . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"infectious",
"diseases/protozoal",
"infections",
"infectious",
"diseases/neglected",
"tropical",
"diseases",
"infectious",
"diseases/skin",
"infections",
"dermatology/skin",
"infections"
] | 2010 | Miltefosine in the Treatment of Cutaneous Leishmaniasis Caused by Leishmania braziliensis in Brazil: A Randomized and Controlled Trial |
North America is currently home to a number of grey wolf ( Canis lupus ) and wolf-like canid populations , including the coyote ( Canis latrans ) and the taxonomically controversial red , Eastern timber and Great Lakes wolves . We explored their population structure and regional gene flow using a dataset of 40 full genome sequences that represent the extant diversity of North American wolves and wolf-like canid populations . This included 15 new genomes ( 13 North American grey wolves , 1 red wolf and 1 Eastern timber/Great Lakes wolf ) , ranging from 0 . 4 to 15x coverage . In addition to providing full genome support for the previously proposed coyote-wolf admixture origin for the taxonomically controversial red , Eastern timber and Great Lakes wolves , the discriminatory power offered by our dataset suggests all North American grey wolves , including the Mexican form , are monophyletic , and thus share a common ancestor to the exclusion of all other wolves . Furthermore , we identify three distinct populations in the high arctic , one being a previously unidentified “Polar wolf” population endemic to Ellesmere Island and Greenland . Genetic diversity analyses reveal particularly high inbreeding and low heterozygosity in these Polar wolves , consistent with long-term isolation from the other North American wolves .
Grey wolves ( Canis lupus ) currently occupy a wide range of habitats across North America , including the tundra , taiga , desert , plain , and boreal forest . Analysing ~40–50 , 000 SNPs from genotype arrays , the hitherto most comprehensive studies have identified seven North American grey wolf populations and ecotypes , which are referred to as West Forest , Boreal Forest , Arctic , High Arctic , British Columbia , Atlantic Forest , and Mexican wolves [1 , 2] . While this represents a major step forward in terms of describing the population structure , much remains to be learned . For example , nuclear DNA-based studies remain to include the full range of North American continental populations , omitting , for example , the Greenland wolves , despite mitochondrial DNA evidence suggesting it might represent an isolated population [3] . Furthermore , previous nuclear-DNA ( nuDNA ) based studies analysed SNP markers that were initially identified in the domestic dog ( C . l . familiaris ) [1 , 2] . Although dogs and wolves are closely related , phylogenetic analyses based on their nuclear genomes show that dogs are a distinct monophyletic clade within wolves [4–6] . Therefore , dog-ascertained markers may not be able to reveal the full genetic structure of wolves , and underestimate their true genetic diversity [7] . Outstanding questions also pertain to the taxonomic status of the North American wolf-like canids . These include the Southeastern red wolf ( C . rufus or C . l . rufus ) ( subsequently referred to as the red wolf ) , as well as the Northeastern groups that are frequently referred to as Eastern timber wolves , Eastern wolves , Algonquin wolves or Great Lakes wolves ( C . lycaon or C . l . lycaon ) ( subsequently referred to as the ‘Eastern timber/Great Lakes wolf’ ) . While recent studies of both SNP-chip and whole genome resequencing data have shown that the genetic makeup of modern C . l . rufus and C . l . lycaon can be explained through admixture of various grey wolf and coyote populations [1 , 8 , 9] , others argue for the possibility of a cryptic third ancestral canid species [10–12] , sparking debate within the field of a two versus three species origin of C . l . rufus and C . l . lycaon [1 , 8 , 9 , 11–16] . Given this debate the definition and integrity of C . l . rufus and C . l . lycaon remains interesting , and clearly requires more research before the scientific community can agree on a fulfilling explanation for their origin and evolution . In light of the above , we undertook an analysis of the genomic structure in , and admixture among , the full range of extant North American grey wolves , coyotes and wolf-like canids , by mapping the hitherto largest dataset of nuclear genome sequences against a de novo assembled wolf reference genome sequence [7] . To specifically test if wolves in Greenland are a unique population , and if the here analysed large genome data set potentially could bring further insight into the evolution of North American wolf like canids .
We generated resequencing data from 15 new canid samples , representing 13 North American grey wolves , one red wolf and one Eastern timber/Great Lakes wolf . Between 56 and 400 million paired end reads were generated per sample . After quality control , including removal of adapters , discarding of low quality reads and removal of duplicates , these reads were aligned to the de novo wolf reference genome [7] , which resulted , for most of the samples , in depth of coverage between 3 . 8–15 . 3x . The exception being the ‘Krummelangsø’ wolf from Greenland , with coverage of only 0 . 4x . We complemented this dataset with 25 previously published samples , all of which were re-mapped following our mapping pipeline and yielded genomes with coverages of 2 . 1–26 . 4x . Additional details for the samples can be found in supplementary S1 Table . The error rates estimated for the different samples ( S1 Table and S1 Fig ) were estimated in ANGSD [17] , using the ‘Daneborg’ Greenland wolf sample as the “error-free” model sample and the ‘Golden Jackal’ as the outgroup . For the newly sequenced samples the error rates ranged between 0 . 039–0 . 076% , except for the ‘Krummelangsø’ Greenland wolf whose error rate was 0 . 146% , consistent with lower sequencing coverage . We also noted elevated error rates in the data from several of the previously published samples ( 0 . 146%-0 . 636% ) , including three coyote samples ( ‘Illinois’ , ‘Quebec’ and ‘Alabama’ ) , the ‘Red wolf 2’ sample , and the wolves ‘Eurasia 3’ and ‘Yellowstone 1’ . Because error rates can affect the results of some analyses , for example the terminal branch lengths estimated using Treemix , they must be considered when drawing conclusions from the results . When inferring ancestry clusters using admixture , with two ancestry clusters ( K = 2 ) , all samples split into two separate clusters representing the grey wolf-like and coyote-like ( S2 Fig ) . When the number of clusters is increased to three ( K = 3 ) , the grey wolves subdivide into one cluster represented by Polar wolves , and a second cluster represented by Eurasian , Mexican and Pacific wolves . All other wolf lineages derive from these two clusters . At K = 4 , the red wolves split from coyotes , and at K = 5 , Eastern timber/Great Lakes wolves form their own cluster ( Fig 1A ) , while the wolves remain as two additional clusters , one containing the Eurasian , Yellowstone , Mexican and Pacific wolves , and the other represented by the East Arctic , West Arctic and Polar wolves . The remaining wolves are mostly represented as a combination of ancestries from these two wolf clusters . However , some wolves showed low levels of shared ancestry with the other three non-grey-wolf clusters . As we increased the number of clusters to K = 15 ( Fig 1A ) , a pattern emerged that is consistent with both the results of the phylogenetic reconstruction and the PCA , making us choose K = 15 at the upper justifiable number of ancestry clusters . Grey wolves split into 9 clusters , each identifying a population of North American wolves , specifically: ( 1 ) Mexican , ( 2 ) Pacific , ( 3 ) Yellowstone , ( 4 ) Central , ( 5 ) Alaskan , and ( 6 ) Atlantic wolves , as well as three groups from the high Arctic , namely ( 7 ) West Arctic ( representing the Banks and Victoria Islands ) , ( 8 ) East Arctic ( representing the Baffin Islands ) , and ( 9 ) Polar ( representing Ellesmere Island and Greenland ) . The ( 10 ) red wolves , ( 11 ) coyotes and ( 12 ) Eurasian wolves each grouped into separate clusters , while individuals from ( 13 ) the Algonquin Provincial Park formed a cluster that is henceforth referred to as to as Eastern timber wolves . The samples from ( 14 ) Isle Royale National Park and Minnesota formed a cluster referred to as Great Lakes wolves , that is closely related to the Eastern timber wolves . The final cluster ( 15 ) contained the ‘Golden Jackal’ outgroup . Phylogenetic reconstruction based on 40 nuclear genomes ( Fig 1B and S3 Fig ) revealed three major clades: one containing the ‘Golden Jackal’ outgroup , a second containing the red wolves and coyotes , and a third containing all grey wolves together with Eastern timber/Great Lakes wolves . These observations of affinity between red wolves and coyotes , and Eastern timber/Great Lakes wolves alongside grey wolves , were also supported by the admixture , Treemix and D-statistics . Within the wolf clade , we observed Old and New World wolves to be reciprocally monophyletic , and within the New World grey wolves , we found the Mexican wolves to be the most divergent from all others . Although we note that ( i ) the overall phylogenetic relationships between the golden jackal , coyotes and grey wolves , and ( ii ) the divergence of Mexican from other New World wolves , were recovered in a previous nuDNA-based analysis [4] , the inclusion of additional North American samples improved the resolution of these relationships . Specifically , it was noted that after the aforementioned basal divergence of Mexican , Yellowstone and Pacific wolves , the remaining North American populations formed a monophyletic clade . Surprisingly , the 2 individuals identified as Central wolves did not form a clade; the Saskatchewan individual was basal to the one from Qamanirjuaq Lake ( Nunavut ) , which in turn was sister group to the remaining Arctic and Polar wolves . However , the position of the Qamanirjuaq individual is only poorly supported . Given that admixture and PCA analyses indicate that its genetic background is largely similar to the Saskatchewan individual , we believe its phylogenetic placement is likely the result of gene flow from other Northern wolf populations . We caution , however , that any conclusions drawn from the phylogenetic tree must be tempered by the large amounts of allele sharing observed in the population genomic analyses ( D-statistics , Admixture and Treemix ) . Further , the amount of incomplete lineage sorting between the different wolf populations that relates to their recent divergence from each other , suggests that several equally likely alternative placements exist for many of these nodes ( S3 Fig ) . Principal component analyses were used to project the SNP variation of the wolves in two dimensions ( Fig 1C , S4 Fig and S5 Fig ) . The wolf diversity expressed in PC1 vs . PC2 ( variance explained 9 . 26–7 . 76% ) , and PC3 vs PC4 ( variance explained 6 . 92–6 . 82% ) ( Fig 1C ) clearly showed a signal that correlates with the geographical distribution of samples running North-South and East-West . Polar , Pacific and Atlantic wolves exhibited highest variation in PC1 and PC2 . Furthermore , Polar and East Arctic wolves were also clearly distinct in PC3 and PC4 . The grouping of individuals was congruent with the clusters identified by NGSadmix [18] , and the tree topology delineated in the phylogenetic reconstruction . Evidence of gene flow among the North American canids was obtained from the D-statistic analyses on the genomes ( S6 Fig and S7 Fig ) . A test of coyote ancestry among the different North American canids ( S6 Fig ) revealed that all North American wolf-like canid populations had a significant , but varying , degree of coyote ancestry , consistent with previously published findings [8 , 9] . Specifically , the highest levels of coyote ancestry were observed in the red wolves , and somewhat lower levels were found in the Eastern timber/Great Lakes wolves . Lowest , although still identifiable , values were observed in the Mexican and the Atlantic wolves . Our expanded dataset also enabled testing for gene flow between North American and Eurasian wolves . The results indicated gene flow between the East Siberian ( Chukchi ) wolf ‘Eurasia 2’ and the Alaskan wolves , consistent with their geographic proximity ( S7 Fig ) . Treemix analyses ( S8 Fig , S9 Fig and S10 Fig ) yielded results that were consistent with the phylogenetic reconstruction ( Fig 1B and S3 Fig ) , with migration events indicating allele sharing between the wolf-like canids , and likely shared coyote ancestry in the Yellowstone , Mexican and Pacific wolves . Using admixture graphs ( Fig 2 ) , we modelled the genomic makeup of red , Eastern timber and Great Lakes wolves , as composed of genomic variation found in North American grey wolves and coyotes . When using admixture graphs ( Fig 3 ) and ( S11 Fig ) to investigate the relationships between Eurasian , Mexican and other North American wolves , the best fitting graph ( Z = -0 , 556 ) assigns Eurasian wolves as sister to all North American wolves , with the Mexican wolf sister to other American wolves , containing considerable coyote introgression . The most parsimonious explanation for this outcome is that all extant North American grey wolves descend from the same ancestral wolf diversity , although whether this ancestral “population” had colonised the North American continent prior to , or post ( possibly on multiple occasions ) the divergence between Mexican and other North American wolves remains a open question . We used f4 ratios to investigate proportion of coyote and grey wolf ancestries in the North American wolf-like canids , setting aside the Polar wolf ( ‘Daneborg’ ) and coyote ( ‘Mexico’ ) as references ( Fig 4A ) . These samples were chosen based on their respective distance to the coyote or wolf cluster in the PCA ( S4 Fig ) , which suggests they may represent the “purest” examples of coyote and North American wolf in our dataset . The f4 ratio estimates showed that the coyotes from Alabama , California , Quebec and Alaska harbour negligible wolf ancestry , while those from Missouri , Illinois and Florida contained between 5–10% wolf ancestry . Much higher levels of wolf versus coyote admixture were observed in red wolves ( 40%:60% ) , the Eastern timber wolves ( 60%:40% ) , and the Great Lakes wolves ( 75%:25% ) . Within wolves , coyote ancestry was highest in the Mexican wolves and the Atlantic Coast wolves ( 10% ) , followed by the Pacific Coast and Yellowstone wolves ( ~5% ) . The wolves from the Canadian archipelago showed less than 3% coyote ancestry . The higher than 100% combined admixture proportions estimated for the wolf ‘Alaska 1’ , likely result from the tree configuration , with the ‘Eurasia 1’ wolf being a fixed member of the quartets used to compute the admixture proportions and indicate Eurasian wolf gene flow into ‘Alaska 1’ , something also supported by D-statistics ( S7 Fig ) . The admixture proportion estimates do not need to add up to 100% because they are estimated separately for the ‘Daneborg’ wolf and the ‘Mexico’ coyote component . Nevertheless , nearly all estimates summed up to 100% , indicating that most samples can be modelled as a mixture between just two components , the wolf and the coyote . f3 statistics were also computed to assess the affinity of the various North American wolf-like canids to the ‘Daneborg’ Polar wolf . As expected from their geographic proximities , wolves from the Canadian Arctic archipelago displayed the highest affinity ( Fig 4B ) , while the amount decreased in populations from further West and South . Furthermore , populations such as the Eastern timber/Great Lakes and red wolves that had substantial amounts of coyote ancestry , showed the lowest affinity with Polar wolves . An inverse pattern was observed when affinities were assessed with the ‘Mexico’ coyote , yielding lowest coyote affinity with the most Northern and Eastern populations ( S12 Fig ) . Our pan-population dataset also enabled us to undertake the first whole-genome based , continental-scale investigation of heterozygosity and inbreeding levels in these canids ( S1 Table ) . The 6 samples with highest estimated error rates ( marked with * , S1 Table ) also have the highest estimates of heterozygosity and low inbreeding coefficients . Given the error rate , heterozygosity and inbreeding coefficients must be interpreted with care in these individuals . The estimates for the remaining grey wolves , coyotes and wolf-like canids ( Fig 5 ) allow for more robust interpretation . The heterozygosity estimates indicated that higher diversity exists among the coyotes , red wolves and Eastern timber/Great Lakes wolves , than in any of the North American grey wolf populations ( Fig 5 ) . Further , within the “true” wolves , the Polar and Mexican wolves showed the lowest heterozygosity , while the Eurasian wolves had the highest . In order to estimate the inbreeding coefficients for these samples , we split the samples into 2 groups , as indicated by the phylogeny , i . e . the red wolves and the coyotes in one group , and the Eastern timber/Great Lakes wolves and the grey wolves in another . To avoid overestimating the inbreeding coefficients ( caused by the Wahlund effect ) , we estimated the allele frequencies in each of these clusters separately , and used these allele frequencies to estimate inbreeding coefficients . Overall , values of inbreeding were relatively low , and the highest values were obtained for the Mexican , Pacific , and one Great Lakes ( Isle Royale National Park ) wolf ( 0 . 2<F<0 . 7 ) ( Fig 5 ) . The ‘Ellesmere 2’ Polar wolf showed rather low ( 0 . 1<F ) levels of inbreeding , which we ascribe to likely admixture ( Fig 1A ) . The ‘Daneborg’ and ‘Ellesmere 1’ Polar wolves showed higher ( F<0 . 5 ) levels of inbreeding , which is probably a more accurate representation of the inbreeding levels in the “Polar wolf” population . To further examine the levels of inbreeding , we estimated the fraction of the genome in long runs of homozygosity ( ROH ) on a subset of seven selected wolves with high coverage , including the ‘Daneborg’ Polar wolf ( S13 Fig and S1 Table ) . The Mexican wolf—Mexico 1—showed the highest proportion of the genome contained in ROH longer than 1 Mb , followed by IRNP and Daneborg . The Polar wolf contained more than ~15% of its genome in long ROH , but none of these segments were longer than 4 Mb , in contrast to IRNP and Mexico 1 , which contained ROH segments longer than 6 Mb . When comparing Daneborg to East and West arctic wolves , represented by Banks Island and North Baffin respectively , the Polar wolf showed significantly longer and more abundant ROH , implying higher levels of inbreeding in the Polar wolf compared to its Arctic conspecifics .
Whole genome sequencing of North American grey wolves and wolf-like canids showed complex mixing of the wolf and coyote lineages . We find the ancestral genomic makeup in the controversial red , Eastern timber and Great Lakes wolves , can be explained as admixture between modern grey wolves and coyotes . However , there were also population specific divergences in these lineages , which distinguish them from modern wolves and coyotes . All in all—to explain modern genomic structure , if a third cryptic canid species have been involved in the formation of the wolf-like canids , this lineage must also be admixed into modern coyotes or grey wolves . Finally , three distinct grey wolf populations were identified among high arctic wolves , including a novel and highly distinct Polar wolf population endemic to Ellesmere Island and Greenland . Overall , our study provides results for future research in canid evolution and relevant knowledge about North American grey wolves and wolf-like canids .
Our dataset consists of 25 previously published canid genomes , 21 of which are derived from North American grey wolves , coyotes , wolf-like canids and a golden jackal ( Canis aureus ) [4–6 , 9 , 31] , as well as new data from 15 additional New World canid specimens sequenced to a coverage of between 0 . 4 and 15x These additional samples consist of one red wolf , one Eastern timber/Great Lakes wolf and 13 grey wolves . Four of the grey wolves are from the High Arctic . Details on samples can be found in supplementary S1 Table and Fig 4B . Samples originating from Canada or the USA were obtained under Article VII , paragraph 6 CITES convention for import as scientific exchange between CITES institution Natural History Museum of Denmark ( DK-003 ) , U . S . Fish and Wildlife Service ( US 096 ( A/P ) ) , University of New Mexico Museum of Southwestern Biology ( US 119 ( A/P ) ) , University of Alaska Museum of the North ( US 130 ( A/P ) ) and University of Alberta Museums & Collection Services ( CA-010 ) . DNA was extracted using the DNeasy Blood & Tissue Kit ( Qiagen ) following the manufacturer’s protocol . DNA was converted into double stranded blunt-end libraries with Illumina-specific adapters [45] using the NEBNext DNA Sample Prep Master Mix Set 2 ( E6070S - New England Biolabs Inc . , Beverly , MA , USA ) following the manufacturer’s protocol . Libraries were sequenced on Illumina HiSeq 2500 platforms using 100 base pair paired-end read chemistry . The short-read data from each sample ( including the previously published genomes ) were mapped against a recently published wolf reference genome [7] . The PALEOMIX ( v1 . 2 . 5 ) [46] pipeline was used to process the reads and to remove adapters . Subsequently , the reads were mapped to the reference genome using the bwa ( v0 . 7 . 10; aln algorithm ) [47] . Picard ( v1 . 128 , https://broadinstitute . github . io/picard ) was used to exclude reads that were PCR or optical duplicates , and to exclude reads that mapped to multiple locations in the genome . GATK ( v3 . 3 . 0 ) [48 , 49] was used to perform an indel realignment step to adjust for increased error rates at the end of short reads in the presence of indels . In the absence of a curated dataset of indels in wolves , this step relied on a set of indels identified in the specific sample being processed . The samples in this study have very disparate coverages across the genome . Instead of calling genotypes at variant sites , which have been shown to introduce biases [50] , the uncertainty in genotypes was propagated to downstream analyses using genotype likelihoods . The genotype likelihoods at variant sites were computed in ANGSD ( v0 . 919 ) [17] using the aligned reads obtained from PALEOMIX , under the model proposed in samtools ( v1 . 2 ) [47] . Nucleotides with base qualities lower than 20 and reads with mapping quality lower than 20 were discarded . Sites with coverage at fewer than 38 out of the 40 samples were excluded . Finally , only sites with an estimated minor allele frequency greater than 0 . 05 were retained . Clusters of ancestry and the associated ancestry proportions were estimated using NGSadmix [18] taking into account the genotype likelihoods obtained from ANGSD [17] . Since low frequency markers are uninformative for admixture analyses , only markers with minor allele frequency greater than 0 . 1 were used for this analysis , which resulted in a total of approximately 4 . 47 million SNPs being retained . Admixture analyses were performed using a range of values for the number of estimated ancestry clusters ( K = 2–15 ) , to explore the structure in the dataset . To avoid convergence to local optima , the analysis was repeated 100 times , and the replicate with the highest likelihood was chosen . For the principal components analysis , a variance covariance matrix was computed from the genotype likelihoods of the various samples using ngsCovar [51 , 52] . For this analysis , only polymorphic sites with a minor allele frequency greater than 0 . 05 were used . Finally , the principal components of the genotype likelihood data were calculated by eigen-decomposition of the variance covariance matrix in R ( v3 . 2 . 1 ) [53] . For each sample , the consensus sequence was generated in ANGSD [17] using the -doFasta 1 option . Regions with missing data were filtered out using trimal ( v1 . 4 . 1 ) [54] with parameters -gappyout , -resoverlap 0 . 60 and -seqoverlap 60 . The phylogenetic trees for each scaffold were constructed using FastTree2 ( v2 . 1 . 10 ) [55] , which uses a generalized time-reversible model for sequence evolution . Only the trees with a minimum of 4 samples were retained to infer the phylogenetic relationship between the samples using ASTRAL-II [56] with default parameters . D-statistics were computed in ANGSD [17] using a single randomly sampled allele at each site that was covered by at least one read . Sites with mapping or base quality less than 30 were discarded . The D-statistic was computed for all possible triplets of samples from the data , using an Israeli golden jackal [5] as outgroup , i . e . the tree configuration used to compute the D-statistic was ( H1 , H2; H3 , ‘Golden Jackal’ ) . While golden jackals in Israel have been documented to admix with dogs , grey wolves , and African golden wolves ( Canis anthus ) [57] , the specific sample perform well as an outgroup for the configurations tested . Between 0 . 2–2 . 1 million sites were used to compute the D statistic , depending on the triplet being used for analysis . Only a subset of the triplets lead to trees that allowed to test hypotheses relating to gene flow between the North American wolves and other canids . Following standard procedure , blocks containing 500 markers each were used to perform the block jackknife [58] procedure to estimate the variance of the statistic . We fitted f-statistics based admixture graphs as implemented in qpGraph from the ADMIXTOOLS package [59] to evaluate the position of the Mexican wolf among Eurasian and American grey wolf diversity . As well as to evaluate the position of red , Eastern timber and Great Lakes wolves among modern coyote and American grey wolf diversity . Specific samples used in the graphs are given in S1 Table . We explain the genomic diversity of red , Eastern timber and Great Lakes wolves as a mix between variation found in coyotes and modern American grey wolves . We considered graphs placing the Mexican wolf as either sister to Eurasian wolves or American wolves under several scenarios of gene flow with various genetic clusters in the graph . We obtained one model with a specific topology , which explained the data well ( Fig 3 ) and present all considered graphs in the supplementary ( S11 Fig ) . Specific migration events between populations were estimated using Treemix [60] . As with the D-statistic analyses , informative sites were identified for each sample by randomly sampling one allele at each site , where both nucleotides and reads with quality lower than 30 were excluded . Only sites where 2 different alleles were sampled were retained for the analysis , leading to ~158K-1 . 938M sites being used , depending on the subset the analysis was performed on . Using these sites and treating each sample as its own initial population , a global tree without any migration edges was constructed . This tree was used as the initial tree for all subsequent Treemix analyses . Treemix graphs with 1–5 migration edges were estimated . For each setting , the best Treemix graph was obtained from 100 replicates . Genetic affinity between pairs of samples ( X and Y ) was estimated by the f3 [61 , 62] statistic using the triplet ( ‘Golden Jackal’; X , Y ) to assess the shared drift between X and Y from the outgroup ‘Golden Jackal’ . The genetic affinity of the samples ( X ) to the ‘Daneborg’ Greenland wolf and the ‘Mexico’ coyote were contrasted by computing the two f3 statistics—f3 ( ‘Golden Jackal’; X , ‘Daneborg’ ) and f3 ( ‘Golden Jackal’; X , ‘Mexico’ ) . These were computed by the threepop program included as part of the Treemix package [60] , using the same set of sites that were used to estimate the Treemix tree . The f4 ratio was used to estimate the amount of coyote and Greenland wolf-like ancestry in all samples included in this study . The program fourpop , part of the Treemix package [60] , was used to compute two f4 statistics for each sample ( X ) —f4 ( ‘Daneborg’ , X; ‘Eurasia 1’ , ‘Golden Jackal’ ) and f4 ( X , ‘Mexico’; ‘Eurasia 1’ , ‘Golden Jackal’ ) . The proportion of ancestry related to the ‘Daneborg’ Greenland wolf was estimated by computing the ratio f4 ( ‘Daneborg’ , X; ‘Eurasia 1’ , ‘Golden Jackal’ ) /f4 ( ‘Daneborg’ , ‘Mexico’; ‘Eurasia 1’ , ‘Golden Jackal’ ) . Similarly , the proportion of ancestry related to the ‘Mexico’ coyote in sample X was computed using the ratio f4 ( X , ‘Mexico’; ‘Eurasia 1’ , ‘Golden Jackal’ ) /f4 ( ‘Daneborg’ , ‘Mexico’; ‘Eurasia 1’ , ‘Golden Jackal’ ) . Further details on the f4 ratio and its use in estimating the admixture proportions can be found in Patterson et al . [61] . For each sample , the heterozygosity was computed using ANGSD [17] under a probabilistic framework based on genotype likelihoods . Reads with mapping quality lower than 20 , and bases with base qualities less than 20 , were excluded from the analyses . The heterozygosity and its variance were calculated from 100 sets of variant sites obtained by bootstrapping on the polymorphic sites . The inbreeding coefficient for each sample was estimated under a probabilistic framework using ngsF [63] , which allows for estimation of inbreeding coefficients without calling genotypes . The genotype likelihoods dataset that was previously calculated for the NGSadmix [18] analysis , was used for computing inbreeding . To avoid convergence to local maxima , the approximated-EM algorithm was started 20 times from random initial values , and the run with the highest likelihood was used as starting values for the final EM run . We selected 7 wolf samples—Mexico 1 , IRNP , Banks Island , North Baffin , Daneborg , Pacific Coast and Yellowstone 2—for the ROH analysis since they spanned all the interesting wolf clades , and had a minimum genome coverage of 10x ( except IRNP , which has a genome coverage of 9x ) . Genotype calling was performed using GATK ( v3 . 3 . 0 ) [49] haplotype caller , restricting the analysis to only variable sites identified in the full set of samples ( ~ 10 . 5 million variable sites ) . Subsequently , we identified ROH using plink ( v1 . 9 ) [64] , only allowing regions longer than 1 Mb , with a minimum of 100 SNPs . | Full genome sequencing is becoming an increasingly valuable tool for both the management of animal populations , as well as fundamental to improving our understanding of their evolutionary history . The grey wolf ( Canis lupus ) is a keystone species in North America whose population structure and admixture has yet to be fully investigated in this way . We compiled a dataset of 40 full genomes spanning their total geographic range on the continent . In addition to confirming general population structure among them and previous reports of admixed origins for several wolf-like canid species , we identify three particularly interesting groups: two in Arctic Canada and one novel “Polar wolf” population on Ellesmere Island and Greenland . The particularly low genetic diversity of the Polar wolves suggests a small and isolated population . Overall we provide new information of relevance for the future management of wolves in Arctic Canada and Greenland . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Material",
"and",
"methods"
] | [
"timber",
"infographics",
"taxonomy",
"ecology",
"and",
"environmental",
"sciences",
"plant",
"products",
"wolves",
"vertebrates",
"animals",
"mammals",
"ethnicities",
"phylogenetics",
"data",
"management",
"aquatic",
"environments",
"phylogenetic",
"analysis",
"bodies",
"of",
"water",
"latin",
"american",
"people",
"computer",
"and",
"information",
"sciences",
"crop",
"science",
"lakes",
"marine",
"and",
"aquatic",
"sciences",
"evolutionary",
"systematics",
"inbreeding",
"agriculture",
"people",
"and",
"places",
"coyotes",
"agronomy",
"eukaryota",
"freshwater",
"environments",
"data",
"visualization",
"heredity",
"earth",
"sciences",
"graphs",
"genetics",
"biology",
"and",
"life",
"sciences",
"population",
"groupings",
"mexican",
"people",
"evolutionary",
"biology",
"amniotes",
"organisms"
] | 2018 | Population genomics of grey wolves and wolf-like canids in North America |
Messenger RNA ( mRNA ) synthesis and export are tightly linked , but the molecular mechanisms of this coupling are largely unknown . In Saccharomyces cerevisiae , the conserved TREX complex couples transcription to mRNA export and mediates mRNP formation . Here , we show that TREX is recruited to the transcription machinery by direct interaction of its subcomplex THO with the serine 2-serine 5 ( S2/S5 ) diphosphorylated CTD of RNA polymerase II . S2 and/or tyrosine 1 ( Y1 ) phosphorylation of the CTD is required for TREX occupancy in vivo , establishing a second interaction platform necessary for TREX recruitment in addition to RNA . Genome-wide analyses show that the occupancy of THO and the TREX components Sub2 and Yra1 increases from the 5′ to the 3′ end of the gene in accordance with the CTD S2 phosphorylation pattern . Importantly , in a mutant strain , in which TREX is recruited to genes but does not increase towards the 3′ end , the expression of long transcripts is specifically impaired . Thus , we show for the first time that a 5′-3′ increase of a protein complex is essential for correct expression of the genome . In summary , we provide insight into how the phospho-code of the CTD directs mRNP formation and export through TREX recruitment .
Gene expression is a fundamental process of every living cell . In eukaryotes , RNA polymerase II ( RNAPII ) transcribes protein-coding genes to synthesize messenger RNA ( mRNA ) . In addition to RNAPII , a plethora of transcription factors is needed for efficient and regulated transcription in vivo . Transcription initiation factors recruit RNAPII to the promoter , whereas transcription elongation factors ensure efficient passage of RNAPII through the transcribed region [1]–[3] . Termination factors are required at the 3′ end of the gene to end the synthesis of mRNA . The newly synthesized mRNA is processed , i . e . capped , spliced and polyadenylated , and packaged into a ribonucleoprotein ( mRNP ) before its nuclear export . Interestingly , these downstream processes occur co-transcriptionally and are intimately linked to each other and to transcription to ensure efficient mRNA biogenesis [4] . The transcription cycle and the co-transcriptional processing of the mRNA are coordinated by the differential phosphorylation of the C-terminal domain ( CTD ) of Rpb1 , the largest subunit of RNAPII [5] . The CTD consists of heptad repeats with the consensus sequence YSPTSPS with 26 or 52 repeats in budding yeast or humans , respectively . It serves mainly as a recruitment platform for transcription and mRNA processing factors , whose association is largely regulated by phosphorylation of the CTD at positions Y1 , S2 , T4 , S5 and S7 [6] . S5 phosphorylation of the CTD is high during transcription initiation , decreases rapidly during the early phase of elongation , and persists at a low level throughout the body of the gene [7]–[10] . Consistently , the capping complex is recruited via direct interaction with the S5 phosphorylated CTD [6] . S2 phosphorylation appears early during transcription elongation , increases during the elongation phase and drops shortly 3′ of the polyadenylation ( polyA ) site [7]–[9] . Transcription elongation , splicing , termination and polyadenylation factors interact with the S2 or S2/S5 ( di ) phosphorylated CTD [5] . In Saccharomyces cerevisiae , the pattern of Y1 phosphorylation ( Y1P ) resembles that of S2 phosphorylation ( S2P ) , with the exception of an earlier decrease at the poly ( A ) site [11] . Y1P stimulates binding of the transcription elongation factor Spt6 and prevents recruitment of termination factors [11] . S7 phosphorylation appears early at the 5′ end of genes , persists at a lower level throughout the open reading frame ( ORF ) and is required for the transcription and correct processing of human snRNA genes [7] , [9] , [12] , [13] . Phosphorylation of T4 increases in the 3′ region of genes subsequently to the increase in S2P and is required for transcription elongation and histone mRNA 3′end processing [14] , [15] . Hence , the CTD plays a pivotal role in the coordination of transcription with downstream processes . In addition to the CTD , a multitude of proteins and protein complexes link transcription to one or several downstream events . In S . cerevisiae , the conserved TREX complex couples transcription to mRNA export [16]–[20] . TREX consists of the heteropentameric subcomplex THO , comprised of Tho2 , Hpr1 , Mft1 , Thp2 and Tex1 , the mRNA export factors Sub2 and Yra1 and the mRNA-binding proteins Gbp2 and Hrb1 [16] . TREX is essential for efficient transcription elongation and links transcription to mRNA export by recruiting the mRNA exporter Mex67-Mtr2 to the mRNA [18]–[20] . Furthermore , TREX also functions in 3′ end processing through its subunit Yra1 , which is recruited to the polyadenylation factor by its interaction with Pcf11 [21] , [22] . In addition , TREX prevents hyper-recombination events associated with inefficient mRNP assembly and functions in transcription-coupled DNA repair ( TCR ) [23] , [24] . Thus , TREX is important for a multitude of co-transcriptional processes . The TREX components Sub2 , Yra1 , Gbp2 and Hrb1 are thought to be transferred to the mRNA during packaging of the mRNA into a ribonucleoparticle ( mRNP ) [25] , [26] . Yra1 ( Aly in metazoans ) directly interacts with the conserved mRNA exporter Mex67-Mtr2 in yeast ( TAP-p15 or NXT-NXF in metazoans ) and functions as an adaptor protein between this heterodimer and mRNA [27] , [28] . Mex67-Mtr2/TAP-p15 binds directly to the mRNA as well as nuclear pore proteins and mediates export of the mRNP through the nuclear pore complex [18]–[20] . In addition to Yra1 , the THO component Hpr1 and the mRNA-binding proteins Nab2 and Npl3 , both of which are also recruited to the mRNA co-transcriptionally , are thought to function as adaptor proteins for Mex67-Mtr2 [29]–[31] . However , the specific function of the different proteins serving as adaptors has remained elusive . TREX is recruited to all protein-coding genes but seems to be especially important for the expression of long and GC-rich transcripts since these are less abundant in deletion mutants of TREX [23] . The downregulation of long transcripts in TREX deletion mutants is consistent with the finding that THO is necessary for RNAPII processivity [32] . In addition , it has been long known that TREX moves along the gene together with RNAPII [16] , but the molecular basis of this interaction has remained enigmatic . The Prp19 complex ( Prp19C ) interacts with TREX and RNAPII and is important to ensure TREX occupancy at the gene , especially at the 3′-end [33] , [34] . However , Prp19C is not responsible for the recruitment of TREX to genes at the 5′ end [33] . Furthermore , the TREX subunit Yra1 has been shown to interact directly with the phospho-CTD , but it is currently unknown whether this interaction is needed for TREX recruitment [35] . Thus , it remained an open question how TREX is recruited to genes and how it interacts with the transcription machinery . Here , we show that the occupancy of THO , Sub2 and Yra1 at genes increases from the 5′ to the 3′ end of the ORF and with gene length using ChIP-chip . A ChIP-based assay with a reporter construct containing a self-cleaving ribozyme shows that recruitment of TREX is at least partially RNA-dependent , but its 5′ to 3′ increase cannot be explained by RNA length . Instead , increasing TREX occupancy is most likely mediated by direct interaction of its subcomplex THO with the S2/S5 diphosphorylated CTD of RNAPII . Consistently , phosphorylation of the CTD on S2 is necessary for TREX recruitment in vivo . In contrast , THO recruitment is independent of Yra1's CTD-binding domain . Importantly , the 5′-3′ increase in TREX occupancy is crucial for the correct expression of long genes . This suggests that the CTD phospho-code dictates mRNP assembly and export through recruitment of TREX .
The Mex67-Mtr2 heterodimer is recruited to mRNAs co-transcriptionally via association with multiple , distinct mRNA-binding proteins [29]–[31] . In S . cerevisiae , several proteins have been proposed to function as Mex67-Mtr2 adaptor proteins to the mRNA . This includes the TREX complex members Yra1 and Hpr1 as well as the mRNA-binding proteins Nab2 and Npl3 . To assess whether these mRNA adaptors are differentially recruited to genes we assessed the genome-wide occupancy of individual TREX components , Nab2 and Npl3 by using high density tiling arrays for the analysis of chromatin immunoprecipitation ( ChIP ) experiments . TREX , Nab2 and Npl3 are recruited to all actively transcribed protein-coding genes with a slight preference of Npl3 for intron-containing genes and of Yra1 for intron-less genes ( data not shown ) . This is consistent with previous data showing that Hpr1 and Sub2 are recruited to all ORFs [23] . TREX , Nab2 and Npl3 are also recruited to RNAPII-transcribed sn- and snoRNA genes , but at a lower level ( Figure S1 ) . This is consistent with the recent finding that in fission yeast THO is recruited to snoRNA genes , negatively regulating their expression [36] . Taken together , however , there is no marked difference in the recruitment of different mRNA export adaptors . TREX travels along the gene together with RNA polymerase II [16] but the molecular basis for this interaction has remained elusive . In order to gain insight into the recruitment mechanism , we calculated meta gene occupancy profiles for each protein . To do this , the average nucleotide occupancy for each analyzed protein was plotted for the top 50% of the most highly transcribed genes ( 1 , 538–2 , 895 bp in length ) after gene length normalization ( Figure 1A ) . THO components , Sub2 and Yra1 appear at the transcription start site and their occupancies increase steadily from the 5′ to the 3′ end of genes ( Figure 1A ) . Consistent with a function in mRNA export , the occupancy levels of all proteins drop at the polyA site and before the termination site ( Figure 1A ) . These results are in accordance with genome-wide data published recently by Aguilera and coworkers for Hpr1 and Sub2 and data by the Rosbash lab for Yra1 , Sub2 and Hpr1 recruitment to selected genes [23] , [26] . According to the meta profiles , Sub2 and Yra1 dissociate from the transcription machinery slightly before the THO complex ( Figure 1A ) . This might be due to a transfer of Sub2 and Yra1 to the mRNA ( also see discussion ) . The increase of TREX components from 5′ to 3′ is striking since the occupancy of bona fide transcription elongation factors , such as Spt5 , Spt6 , Spt16 , Bur1 and Paf1 , does not increase over the length of the ORF ( Figure S2 ) . Thus , TREX might be the only transcription elongation complex whose occupancy increases with gene length . The 5′ to 3′ increase in TREX occupancy could either lead to a maximal occupancy independent of gene length or to a higher maximal occupancy for longer genes in case the occupancy constantly increases until the 3′ end of the gene . In order to distinguish between these two possibilities , genes were subdivided into eight length classes between 500 bp and 5000 bp . The peak occupancy of TREX components at each gene within one length class was normalized to the occupancy of RNAPII ( Rpb3 ) to correct for transcription activity , and the average normalized occupancy of TREX components was plotted for each length class ( Figure 1B ) . As expected for the observed 5′ to 3′ increase , the average occupancy of all factors increases with gene length , i . e . , the longer a gene , the higher TREX occupancy . For genes shorter than 1500 bp this increase in TREX occupancy is roughly linear and lower for genes longer than 1500 bp . The 5′ to 3′ increase in TREX occupancy is also evident in meta gene occupancy profiles calculated for different length classes; the maximal occupancy of Tho2 , Hpr1 , Mft1 , Sub2 and Yra1 increases with length of the gene class ( Figure S3I , K , L , M , N ) . This increase in TREX occupancy with gene length is not caused by antisense transcription as the same increase is observed when genes containing CUTs or SUTs are omitted from the calculation ( Figure S4 ) . Taken together , the occupancy of THO , Sub2 and Yra1 increases from the 5′ to 3′ end of genes . In contrast , the occupancy of TREX components Gbp2 and Hrb1 decreases slightly from 5′ to 3′ ( Figure 1C ) and increases only slightly with gene length ( Figure 1D ) . This is also evident in the meta gene occupancy profiles of Gbp2 and Hrb1 for different length classes ( Figure S3O , P ) . Interestingly , the occupancy of Nab2 and Npl3 , two other mRNA-binding proteins in yeast important for mRNA export , is also constant over the ORF and increases only slightly with gene length ( Figures 1D and S3Q , R ) . This difference in distribution over the length of the gene is also reflected by the correlations of the peak occupancies between the different proteins . For example , the THO subunits , Yra1 and Sub2 correlate highly with each other , whereas Gbp2 and Hrb1 correlate well with Nab2 and Npl3 and with general transcription elongation factors ( Figure S5 ) . In a study examining mRNP composition and structure it was suggested that the amount of Nab2 increases with RNA length [37] , a finding that may apply to other mRNA-binding proteins present in the mRNP . Thus , Nab2 , and other mRNA binding proteins such as Gbp2 , Hrb1 and Npl3 , may be removed from chromatin by transfer to the nascent mRNA . Regardless , this shows that the occupancy of a core TREX complex consisting of THO , Sub2 and Yra1 increases from 5′ to 3′ of the gene ( also see Discussion ) . The 5′ to 3′ increase in TREX recruitment may be explained by interaction with the mRNA and/or the C-terminal domain ( CTD ) of Rpb1 , the largest subunit of RNA polymerase II ( RNAPII ) . First , we assessed whether the increased association of TREX towards the 3′ end of genes is caused by interaction with the growing mRNA . In order to assess the occupancy of a chosen protein dependent on the length of the RNA , it is necessary to cut the mRNA at a specific position to shorten the nascent RNA to a defined length in relation to which TREX occupancy can be measured . To do this , we established a hepatitis δ ribozyme based ChIP assay in S . cerevisiae according to [38] ( Figure 2A ) . As the mRNA is synthesized , the internal ribozyme sequence folds into an enzymatically active RNA and initiates co-transcriptional self cleavage . This cleavage event releases the 5′ portion of the nascent mRNA and any proteins bound to it from chromatin , while RNAPII and the 3′ portion remain at the transcription site ( Figure 2B ) . For each protein we measured its occupancy at a defined distance to the cleavage site . This occupancy was compared to the occupancy at the same position but with an uncleaved and thus longer RNA . As expected , the occupancy of RNAPII is not dependent on RNA ( Figure 2C , Rpb3 ) . In contrast , the occupancy of all TREX components , as well as Nab2 and Npl3 , significantly decreases to about 70% 100 bp downstream of the ribozyme site ( Figure 2C; P2 ) . The efficiency of ribozyme cleavage in this context is not known . However , since protein occupancy decreases , a subset of transcripts must cleave soon after synthesis . This reduced occupancy indicates that the recruitment of TREX , Nab2 and Npl3 is at least partially dependent on RNA . It is of note that prior studies of the TREX components Sub2 , Yra1 and Hpr1 , demonstrated varying degrees of RNA-dependent interaction with chromatin [26] . Because our ribozyme cleavage assay affects each of these complex members equivalently , we suggest that discrepancies within this previous study are due to the use of RNAse digestion to assess RNA-dependent recruitment . More specifically , because nuclease digestion follows formaldehyde crosslinking steps , we speculate that this treatment may result in RNAse-resistant interactions that were RNA dependent in vivo . Our results suggest that the occupancy of all TREX components , as well as Nab2 and Npl3 , is at least partially dependent on RNA at actively transcribed genes . Importantly , though , we used this assay to determine , whether the 5′ to 3′ increase in the occupancy of TREX components is caused by the nascent RNA chain . With increasing RNA length proteins bound to the RNA are taken further and further away from the DNA template and might not be crosslinked to chromatin any more ( Figure 2A , B ) . To test TREX occupancy dependent on nascent mRNA length , we analyzed the association of these factors at different downstream portions of the gene , distal to the ribozyme cleavage site ( Figure 2A , P3–P6 ) . Indeed , while the occupancy of Hpr1 and Sub2 is decreased 0 . 1 and 0 . 4 kb 3′ of the cleavage site compared to the inactive ribozyme sequence , i . e . at the same genomic position but with a longer nascent mRNA , it is unaffected 0 . 7 kb and further downstream of the cleavage site ( Figure 2D ) . Thus , TREX occupancy is independent of RNA length once the nascent mRNA is longer than approximately 550 nt suggesting that the 5′ to 3′ increase in TREX occupancy observed over several kilobases is not caused by the growing mRNA chain . Another recruitment platform for TREX could be the CTD of RNAPII . The CTD is differentially phosphorylated during the transcription cycle and is well established to recruit a plethora of mRNA processing factors [5] . To assess this possibility , we compared recruitment of the TREX complex and RNA-binding proteins to the phosphorylation pattern of the CTD . Specifically , the meta gene occupancy profiles of Y1P and S2P are very similar to the ones of TREX with a biased distribution towards the 3′ end ( Figure 3A ) . However , Y1P occupancy drops at the polyA site as does TREX occupancy , whereas S2P levels drop slightly downstream of the polyA site ( Figure 3A ) . In addition , the peak occupancies of Y1P and S2P increase with gene length similar to TREX ( Figure 3B ) . To test whether TREX occupancy depends on Y1 or S2 phosphorylation we used an S2A mutant with nine wild-type ( wt ) and six S2A repeats [39] and engineered an Y1F mutant carrying five wt and nine Y1F repeats . The remaining wild type repeats are necessary for survival since mutation of all S2 or Y1 residues is lethal [39] . A CTD truncated to 14 repeats served as wild-type control [39] . Interestingly , the S2A mutation leads to a decrease in Y1 phosphorylation and vice versa ( Figure 3C , S2A and Y1P ) . This is not due to decreased RNAPII association , as the occupancy of RNAPII ( Rpb1 ) is largely unaffected in both mutants at the PMA1 and the ADH1 gene ( Figure 3C , RNAPII ) . This suggests that Y1 and S2 phosphorylation are interdependent , although we cannot exclude that our results are a reflection of epitope masking . Importantly , the occupancy of Hpr1 , and likely the whole THO complex , is also decreased in the Y1F and the S2A mutant showing that recruitment of THO is dependent on proper Y1 and/or S2 phosphorylation in vivo ( Figure 3C ) . Consistently , occupancy of the TREX subunits Yra1 and Sub2 is impaired in the S2A mutant ( Figure S6 ) . S2 rather than Y1 phosphorylation is probably essential for TREX recruitment in vivo since sn/snoRNA genes are low in TREX occupancy and S2P but high in Y1P ( Figure S1 ) . However , the levels of Y1 and S2 phosphorylation most likely decrease in both CTD mutants , making it impossible to determine unambiguously which one of the two phosphorylation events is necessary for TREX occupancy in vivo . It has been shown recently that Yra1 binds to the S2/S5 diphosphorylated CTD in vitro and that deletion of the N-terminal 76 amino acids of Yra1 abrogates this interaction as well as recruitment of Yra1 to genes [35] . Thus , this N-terminal domain of Yra1 was named PCID for phospho-CTD interaction domain . In addition , the PCID also contains the NLS of Yra1 and is thus necessary for efficient nuclear localization of Yra1 [35] . However , it remained unclear whether this domain is also responsible for TREX recruitment to genes . In order to test whether the PCID of Yra1 is required for recruitment of TREX components in vivo , we assessed Yra1 , Hpr1 and Mft1 occupancy in the yra1-ΔPCID mutant ( Figure 4A ) . As shown before , Yra1 occupancy is greatly decreased in the yra1-ΔPCID mutant whereas RNAPII occupancy is not affected ( Figure 4B , RNAPII and Yra1 , and [35] ) . In contrast to Yra1 , THO recruitment is not affected in the absence of Yra1 ( Figure 4B , Hpr1 and Mft1 ) . Thus , recruitment of THO is independent of Yra1 . Since recruitment of THO is not dependent on Yra1 , we asked whether the interaction of Yra1 with THO could be impaired by deletion of the PCID . Full-length Yra1 copurified with TREX whereas yra1-ΔPCID did not ( Figure 4C ) . The lack of yra1-ΔPCID incorporation into the TREX complex might be due to three reasons: 1 . the PCID being necessary for the interaction of Yra1 with the other TREX components , 2 . the mislocalization of yra1-ΔPCID to the cytoplasm , and/or 3 . an impaired interaction of Yra1-ΔPCID with Pcf11 , which is needed for recruitment of Yra1 [21] and from which Yra1 – after recruitment – could be transferred to THO . Important in this context , THO is recruited to the transcription machinery independently of Yra1 . In order to assess whether recruitment of THO is mediated by direct binding of THO to the phosphorylated CTD and which phosphorylation event is necessary for this interaction , we performed pulldown experiments . CTD peptides that were either unphosphorylated , monophosphorylated on Y1 , S2 or S5 or diphosphorylated on Y1 and S2 , Y1 and S5 or S2 and S5 were immobilized on beads and incubated with the endogenous THO complex purified from yeast under high salt conditions . This purification method yields a pure THO complex composed of Tho2 , Hpr1 , Mft1 , Thp2 and Tex1 but lacking Sub2 , Yra1 , Gbp2 and Hrb1 ( Figure S7 ) . The unrelated Rix1 complex , which is required for processing of ITS2 sequences from the 35S pre-rRNA , served as negative control [40] , [41] . Pcf11 served as a positive control since this 3′end processing factor binds to the S2 phosphorylated CTD [42] , . THO binds to the S2 and the S5 monophosphorylated CTD and exhibits the strongest interaction with the S2/S5 diphosphorylated CTD ( Figure 5 , upper panel ) . In contrast , THO did not bind to the Y1 phosphorylated CTD peptides ( Figure 5 ) . When the S2/S5 diphosphorylated CTD peptides were treated with alkaline phosphatase ( AP ) the interaction between THO and the CTD was abrogated , showing that the interaction is indeed phosphorylation dependent ( Figure 5 , upper panel ) . Thus , THO associates directly with the S2/S5 diphosphorylated CTD . This is consistent with the requirement for S2 phosphorylation for TREX occupancy in vivo ( Figure 3 ) and the increase in occupancy towards the 3′ end ( Figure 1 ) . Since S2P increases from 5′ to 3′ and with the length of the gene while S5P peaks at the 5′ end and persists at a basal level throughout the gene ( Figure S3C , D and [44] ) the binding of THO to the S2/S5 diphosphorylated CTD is most likely the molecular basis for the 5′ to 3′ increase of THO , Sub2 and Yra1 . Importantly , we asked whether the 5′-3′ increase of TREX is physiologically relevant . Analysis of the transcriptomes of TREX knock-out mutants would make effects due to the lack of the whole protein indistinguishable from effects caused by the lack of the 5′-3′ increase . Thus , we exploited an allele of THO2 encoding a C-terminally TAP-tagged Tho2 that fortuitously results in defective recruitment of TREX towards the 3′ end . In contrast to the N-terminally TAP-tagged Tho2 , TAP-Tho2 , the occupancy of Tho2-TAP neither increases from 5′ to 3′ nor with gene length genome-wide ( Figure 6A , B ) . Since the signals obtained from genome-wide experiments are not quantitative , we determined the levels of Tho2-TAP and TAP-Tho2 recruitment by ChIP followed by quantitative RT-PCR for different regions of the PMA1 gene . Importantly , Tho2-TAP and TAP-Tho2 are recruited to similar levels to the 5′ end of PMA1 ( Figure 6C ) . Next , we wanted to assess whether recruitment of the whole TREX complex is impaired similarly to Tho2-TAP . Since the protein A moiety of the TAP tag interferes with the use of any antibody , the other TREX components were tagged with the avidin epitope tag ( Avi-tag ) . This bacterial biotin-acceptor peptide is biotinylated in cells expressing the corresponding biotin ligase BirA and can be immunoprecipitated with streptavidin beads [45] . Hpr1 , Sub2 and Yra1 are recruited to similar levels to the 5′end of genes in wt and THO2-TAP cells , but do not increase towards the 3′ end in the THO2-TAP mutant ( Figures 6D and S8A , B ) . Transcription by RNAPII is largely unaffected as judged by the fact that RNAPII occupancy does not change significantly in the THO2-TAP strain ( Rpb1-Avi , Figure 6E ) . In addition , TAP-Tho2 and Tho2-TAP are assembled into the TREX complex ( Figure S8C ) . Thus , the TREX complex is intact and recruited to the 5′ end of genes , but no longer shows 3′ end biased occupancy in the THO2-TAP strain . In order to assess the physiological relevance of TREX's increasing occupancy , the transcriptomes of the THO2-TAP and a corresponding wild-type strain were analyzed for transcripts with differential expression . Importantly , the expression of long transcripts is decreased in the THO2-TAP strain ( Figure 6F ) . In contrast , the expression of other gene classes , including highly expressed , highly transcribed , GC-rich , convergent and divergent genes , does not change when the 5′-3′ increase in TREX recruitment is impaired ( Figures 6F and S9 ) . In addition , the position of the first nucleosome or the promoter type does not influence expression in the THO2-TAP strain ( Figure S9 ) . Thus , the 5′ to 3′ increase in TREX occupancy is important for the expression of long transcripts . Interestingly , THO2-TAP is synthetically lethal with yra1-ΔPCID , i . e . when Yra1 is largely mislocalized to the cytoplasm and mRNA export compromised ( Figure S10 ) . This finding underlines the physiological importance of the 5′ to 3′ increase of TREX . In summary , we identified a direct interaction of TREX with the S2/S5 diphosphorylated CTD of RNAPII that most likely mediates the 5′-3′ increase in TREX occupancy important for the expression of long genes . Thus , the differential phosphorylation of the CTD not only coordinates transcription and mRNA processing , but also couples transcription to mRNA export via TREX recruitment .
The TREX complex is essential for gene expression through its functions in transcription elongation , 3′end processing and mRNA export . It has been known for years that TREX is recruited to the transcription machinery . However , how TREX interacts with RNAPII has remained enigmatic . Here , we show that TREX binds directly to RNAPII through the direct interaction of its subcomplex THO with the S2/S5 diphosphorylated CTD of Rpb1 ( Figure 7 ) . THO is thus a new member of a small but growing class of protein complexes that bind to the S2/S5 double mark . Other S2/S5 diphosphorylated CTD binding proteins are Set2 , which methylates histone H3 during transcription elongation , and Rco1 , a subunit of the RPD1S complex , which deacetylates H3 and H4 , preventing cryptic transcription [46]–[49] . TREX is recruited to genes early during transcription elongation and increases in occupancy as elongation proceeds . This increase in TREX occupancy is most likely mediated by the increase in S2 phosphorylation ( Figures 3 and 5 ) . The importance of S2 phosphorylation is consistent with the finding that Ctk1 , the S2 kinase , physically and genetically interacts with TREX [25] . In addition to the CTD , RNA is necessary for TREX recruitment ( Figure 2C ) . Interestingly , it was recently shown that the C-terminus of Tho2 interacts with nucleic acids and is necessary for occupancy of THO at transcribed genes [50] . Thus , the lack of the 5′-3′ increase in TREX occupancy in the Tho2-TAP mutant could be due to the fact that the C-terminal TAP tag interferes with nucleic acid binding , i . e . either of DNA or RNA . This is consistent with the dependence of TREX recruitment on RNA . However , the elongating RNA chain is not necessary for the increase in TREX occupancy along the gene ( Figure 2D ) . Taken together , we propose the model that TREX is recruited to the transcription machinery by interaction of THO with RNA and the S2/S5 diphosphorylated CTD ( Figure 7 ) . The 5′ to 3′ increase in S2 phosphorylation mediates a corresponding increase in TREX occupancy ( Figure 7 ) . Previously , we showed that the Prp19 splicing complex ( Prp19C ) is not necessary for initial recruitment of TREX to the 5′ end of genes , but rather ensures TREX occupancy along the gene unit [33] . This suggests that Prp19C functions to stabilize the interaction between TREX and the transcription machinery . Our studies above now complement this finding by demonstrating a direct interaction between TREX and the S2/S5 diphosphorylated CTD ( see above ) . Consistent with earlier observations [44] TREX leaves the gene at the polyadenylation site suggesting its dissociation from RNAPII before transcription termination ( Figures 1 and 7 ) . The dissociation of TREX could be brought about by the decrease in S2 phosphorylation at the polyadenylation site . However , TREX could also be dissociated by termination factors that bind to the CTD when Y1 phosphorylation decreases and polyadenylation factors are recruited [11] , [44] ( Figure 7 ) . The latter scenario seems especially likely since the meta gene occupancy profiles of TREX components more closely resemble that of Y1P than S2P ( Figure 3A ) but THO does not bind to the Y1 phosphorylated CTD ( Figure 5 ) . In addition , TREX dissociation could be enhanced by loss of Prp19C function and/or cleavage of the mRNA during 3′ end formation . Thus , multiple processes may ensure the timely dissociation of the proteins necessary for mRNP formation ( Figure 7 ) . Sub2 and Yra1 might leave the transcription machinery as part of the mRNP ( Figure 7 ) , whereas THO might either bind to the mRNP or be directly recycled for a new round of transcription ( not shown ) . Although highly speculative , the interaction of THO with the phospho-CTD could be conserved in metazoans . A direct interaction of TREX with the transcription machinery might be the basis for the association of TREX with naturally intronless mRNAs via a sequence element termed CAR-E [51] . TREX could then recruit the splicing machinery to transcribed genes , consistent with the largely cotranscriptional splicing in higher eukaryotes . Interestingly in this context , a human transcription elongation factor , CA150 , binds directly to the phospho-CTD and to the splicing factor SF1 repressing transcription elongation [52] , [53] . Conversely , metazoan TREX could be recruited to the transcription machinery by the spliceosome . This seems especially likely if human TREX also interacts with Prp19C , which is recruited to the transcription machinery by direct interaction with the splicing factor U2AF65 , which in turn interacts directly with the phospho-CTD [54] . In any event , it will be interesting to see whether TREX recruitment to the mRNA also increases towards the 3′-end of the gene in higher eukaryotes and , if so , whether TREX will be important for the expression of long transcripts . Although the major components of TREX , THO , Sub2 and Yra1 , show a 3′ biased distribution , Gbp2 and Hrb1 , two members of the TREX complex , do not . The mRNA-binding proteins , Nab2 and Npl3 , also show equal occupancy across both long and short genes . We hypothesize that these mRNA-binding proteins are transferred to the mRNA during transcription elongation and thus leave the transcription site . Interestingly , Npl3 – which is not a TREX component – also binds to the S2 phosphorylated CTD [55] . Therefore , mRNA-binding proteins may be recruited to the site of transcription by interacting either directly or indirectly , i . e . via THO , with the phosphorylated CTD . During transcription elongation these mRNA-binding proteins are then transferred from the CTD to the mRNA packaging it into an mRNP . The occupancy of THO , Sub2 and Yra1 increases from the 5′ to the 3′ end of the gene . This makes this “core” TREX complex unique among the transcription elongation factors ( Figure S2 ) as well as known S2P-CTD-interacting proteins ( Figure S11 ) . Importantly , this 5′-3′ increase is physiologically important . Exploiting a mutant of THO2 , in which TREX is recruited to the gene but does not increase towards the 3′ end of the gene ( Tho2-TAP ) , we show that the 5′-3′ increase in TREX occupancy is important for the correct expression of long genes . THO , Sub2 and Yra1 might be needed at higher levels towards the 3′ end of genes , to keep the nascent mRNA in the vicinity of the CTD ( Figure 7 ) . This could be necessary to ensure efficient and correct processing and packaging of the mRNA , which is consistent with the finding that a continuous transcript is needed for mRNA processing [38] . A fully extended CTD is approximately 700 Å long , which corresponds to the length of a 2 . 5 kb long linear mRNA . Since the median length of an mRNA in S . c . is 1 . 436 nucleotides [56] , the CTD is in principle able to span the entire length of an average mRNA . However , it is unlikely that the CTD as well as the mRNA exist in a fully extended form in vivo . Thus , it remains to be elucidated , how mRNP formation is spatially organized . We propose that TREX promotes mRNP packaging through its bifunctional binding to the CTD and RNA by ensuring spatial proximity of the nascent mRNA to mRNP binding proteins , which are recruited to the CTD . In summary , we identify a direct interaction of TREX with the phospho-CTD as one molecular mechanism of TREX recruitment to transcribed genes ( Figure 7 ) . Thus , in addition to its many known functions the CTD code probably also coordinates transcription with mRNA export .
Yeast strains and plasmids are listed in Tables S1 and S2 , respectively . Chromatin immunoprecipitation ( ChIP ) assays were performed according to [57] and ChIP-chip experiments as in [7] , [11] . ChIP-chip of Rpb3-TAP , TAP-Tho2 , Tho2-TAP , Hpr1-TAP , Mft1-TAP , Gbp2-TAP , Hrb1-TAP , Nab2-TAP and TAP-Npl3 was performed using IgG coupled Sepharose beads ( GE ) . Antibodies directed against the protein were used for Yra1 [16] , Sub2 [16] , Y1P [11] and S2P [12] . Details and differences are described in the Supplementary Material . The ChIP-chip data has been deposited in ArrayExpress ( www . ebi . ac . uk/arrayexpress/ ) , accession number E-MTAB-1400 . Tho2-TAP and wild type cells were grown in SDC media , RNA was isolated and hybridized in dye-swap biological replicate to dual-channel 70-mer oligonucleotide arrays to obtain four measurements as previously described [58] . Up- or down-regulation of expression in the THO2-TAP strain was defined as a >1 . 7-fold change versus the average wild-type with a p-value of <0 . 05 . The average length , GC-content , expression level , RNAPII occupancy , convergence , divergence , +1 and +2 nucleosome positioning of the up- and down-regulated genes was calculated and the statistical significance determined using the Wilcoxon rank sum test . SAGA and TFIID dominated genes were analyzed for expression changes versus all promoters using the Wilcoxon rank sum test . The microarray gene expression data has been deposited in ArrayExpress ( www . ebi . ac . uk/arrayexpress/ ) , accession number E-MTAB-1892 . Tandem affinity purifications ( TAPs ) were essentially done as described previously [16] . HA-tagged Yra1 was detected with an anti-HA antibody ( Roche ) . Protein complexes used in the pulldown assays were purified from yeast until the TEV eluate and for THO and TREX followed by a second purification step using metal ion affinity chromatography . Details are given in the Supporting Information . Pulldown assays were performed as described previously with following modifications [59] . For each pulldown assay 15 µl of Streptavidin coupled magnetic beads ( Invitrogen ) were washed three times with HS buffer ( 1 M NaCl , 25 mM Tris/HCl , pH 8 . 0 , 5% Glycerol , 2 . 5 mM DTT , 0 . 025% NP-40 , 0 . 1% BSA ) . Beads were resuspended in 100 µl HS buffer and incubated with 10 µg of each peptide for 2 h at 4°C . Peptides sequences are listed in Table S3 . Peptides were ordered from PSL ( Heidelberg ) and PANAtecs ( Tübingen ) . Beads were then washed once with HS buffer and two times with LS buffer ( 100 mM NaCl , 25 mM Tris/HCl , pH 8 . 0 , 5% Glycerol , 2 . 5 mM DTT , 0 . 025% NP-40 , 0 . 1% BSA ) . For alkaline phosphatase ( AP ) treatment samples were washed two times with 1× fast digestion buffer ( Fermentas ) , incubated for 15 min at 37°C with 25 U FastAP ( Fermentas ) , washed 2× with LS buffer and resuspended in 100 µl LS buffer . To test CTD binding equal amounts of the different protein complexes ( typically 5–10 µl ) were incubated with the CTD-coupled beads for 90 min at 4°C . The non-bound fraction was collected . After 4 washing steps with 500 µl LS buffer beads were resuspended in 1× gel-loading buffer to elute the bound protein complexes . Non-bound and bound protein complexes were detected with an anti-CBP antibody ( Open Biosystems , CAB 1001 ) recognizing the remaining CBP-tag on the tagged proteins ( Hpr1 , Rix1 and Pcf11 , respectively ) . Raw and normalized data are available at ArrayExpress ( www . ebi . ac . uk/arrayexpress/ ) , accession numbers E-MTAB-1400 ( ChIP-chip data ) and E-MTAB-1892 ( microarray gene expression data ) . | Gene expression is a fundamental cellular process that translates the information stored in the DNA into proteins , the workhorses of the cell . Eukaryotic cells contain a nucleus , where the genetic information is stored and transcribed by RNA polymerase II into messenger ( m ) RNAs . These copies of the blueprint of life need to be exported to the cytoplasm for protein production . Interestingly , mRNA synthesis is coupled to nuclear mRNA export . The protein complex TREX mediates this coupling of transcription to mRNA export . To assess the recruitment mechanism of TREX to genes we analyzed the presence of TREX over the whole genome in budding yeast . We found that there is more TREX at the end than at the beginning of genes . TREX binds to a subunit of RNA polymerase II , phosphorylation of which increases over the gene mediating the increase in TREX . Importantly , this increase in TREX over genes is important for normal levels of long transcripts . Thus , we show for the first time that a gradual increase of a protein complex is important for correct expression of the genome . We propose that TREX functions to keep the mRNA in the vicinity of the transcription machinery for correct processing and mRNP formation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2013 | Recruitment of TREX to the Transcription Machinery by Its Direct Binding to the Phospho-CTD of RNA Polymerase II |
Across the nervous system , certain population spiking patterns are observed far more frequently than others . A hypothesis about this structure is that these collective activity patterns function as population codewords–collective modes–carrying information distinct from that of any single cell . We investigate this phenomenon in recordings of ∼150 retinal ganglion cells , the retina’s output . We develop a novel statistical model that decomposes the population response into modes; it predicts the distribution of spiking activity in the ganglion cell population with high accuracy . We found that the modes represent localized features of the visual stimulus that are distinct from the features represented by single neurons . Modes form clusters of activity states that are readily discriminated from one another . When we repeated the same visual stimulus , we found that the same mode was robustly elicited . These results suggest that retinal ganglion cells’ collective signaling is endowed with a form of error-correcting code–a principle that may hold in brain areas beyond retina .
Understanding the manner in which population of neurons encode information is fundamental to systems neuroscience . Recent years have seen rapid progress in experimental techniques for recording simultaneously from hundreds of neurons of more [1–5] , which provide us with excellent access to the activity states of relevant neural populations . However , what continues to make this problem challenging and mathematically complex is the fact that collective neural activity patterns have enormous dimensionality–for instance , if we only need to keep track of spiking or silence for each neuron , a population of N neurons still has 2N possible activity states [6 , 7] . While debate rages about the amplitude and significance of noise correlations [7–10] , it is well established that nearby neurons have overlap in their receptive fields or tuning curves which introduces signal correlation and redundancy . Notwithstanding the popularity of ideas about efficient coding and redundancy reduction [11 , 12] , direct measurement reveals extensive information redundancy among neurons in many brain regions [13–17] . One popular framework has been ring models or probabilistic population codes , in which a population of neurons with a spectrum of different tuning curves encode for scalar stimulus variables , such as neurons in V1 encoding the orientation of a bar of light [18–21] . Study of these models has revealed many important insights into population codes , such as the dramatic effect that weak pairwise correlations can have on the code of a large neural population as well as the sensitivity of these effects to the specific pattern of correlation among neurons [7 , 18 , 22–24] . However , the generalization of this framework to the high-dimensional stimuli that we often encounter in real life is not so obvious . An alternative approach is to formulate approximate models of the probability distribution over all possible neural activity patterns [25–27] and examine the structure of this entire probability landscape . An appealing hypothesis about the function of large populations of sensory neurons is that their combinatorially huge space of possible codewords could enable a form of error correction . Specifically , the high capacity N-dimensional coding space could be partitioned into subsets , and within each subset distinct words would be considered noise-corrupted versions of one another . The actual information conveyed by the channel is then the identity of the relevant subset , as opposed to the precise word within that subset . This qualitative picture is the intuitive basis of error-correcting codes [28]; it is natural to then ask whether the redundancy observed in local neural circuits is a signature of such codes being operative in the neural population . In the retina , each image point is encoded by many ganglion cells with overlapping receptive fields [29–31] . The population code of the retinal ganglion cells is understood as having parallel channels of visual information: each channel is formed by ganglion cells of a single functional/morphological type , whose receptive fields tile visual space with minimal overlap , and each such mosaic conveys qualitatively different visual information to the brain using receptive fields with different spatiotemporal selectivity [32] . The simplest idea is that different visual channels project to different targets in the central brain , thus subserving different visual behaviors . However , there are many ganglion cell types , at least 32 [31] , and only two major targets in the central brain: the LGN and the superior colliculus . Furthermore , most ganglion cell axons branch and project to multiple targets [33] . In fact , almost all of the ganglion cells send collaterals to the superior colliculus / optic tectum in many species [34–37] . One might argue that different ganglion cell types segregate to different lamina in thalamus [38] and colliculus [39] , thereby keeping the visual channels distinct . However , these channels recombine by synapsing into the same target neurons as early as layer 2/3 of the primary visual cortex [40] and the intermediate layers of the colliculus [41] . Therefore , visual centers just beyond the retina are able to combine information across the different parallel channels . Important for the scope of population codes , correlation is significant between nearest neighbor ganglion cells in each mosaic [42–44] as well as among ganglion cells of different functional type [13 , 30 , 45] . And combinations of firing among multiple ganglion cells have been shown to encode different visual features than those encoded by the constituent cells [46–48] . As a result , populations of hundreds of neurons from a combinatorial neural code as early as the level of the retinal ganglion cells . Since downstream areas infer visual information solely from the ganglion cell spike trains , this choice of readout must be derived from the statistical structure of the retinal population output . The question of optimal statistical structure over noisy , spiking sensory neurons has been investigated theoretically [49] . In conditions of high noise , an error-correcting organization featuring discrete clusters of population words optimizes information transfer . Is this the case in real neural populations ? In the present paper , we address this question empirically by fitting a statistical model to the retinal ganglion cell population representing a small visual patch , and show that a discrete population structure emerges from the model and naturally captures qualitatively important features of the visual stimulus . We obtained multi-electrode array data from ∼150 salamander retinal ganglion cells simultaneously responding to various visual stimuli , including natural movies . Our electrode array covered a small , dense patch of retina , so that the cells in our recording had largely overlapping receptive fields conveying information about a restricted region of visual space [1] . We then investigated a model in which the complex , fluctuating patterns of spikes across the population could be grouped into a modest number of “collective modes . ” The possibility of this dimensionality reduction is a consequence of redundancy in the retinal code . Our model is highly tractable to fit , and allows for exact sampling from the probability distribution over retinal population response patterns . We found that this model provided an excellent description of the frequency of occurrence of activity patterns in the neural population , and we further investigated its implications for information encoding . After fitting the model , we inferred the temporal sequence of mode occurrence underlying the data . Population modes carried a detailed representation of the visual stimulus . Moreover , our model grouped spiking patterns together into discriminable modes in a manner reminiscent of error-correcting codes , and the modes were indeed far more reliable over repeated stimulus presentations than were individual spiking patterns . Thus , the organization of the activity states into collective modes may constitute an important aspect of the population neural code of retinal ganglion cells . We also note that nothing in our analysis methods was specific to the retina , and so this approach can be readily applied to population codes in other brain regions as well .
We recorded spike trains from 152 retinal ganglion cells in response to a non-repeated natural movie stimulus ( see Methods ) . In our recordings , the retinal population response was characterized by rapidly fluctuating groups of ∼10 neurons spiking simultaneously , interspersed with periods of relative quiescence ( Fig 1A and 1B ) . If such patterns involve specific subsets of neurons firing together more regularly than expected by chance , they may represent distinct population code words that capture collective information not represented by single-cell responses alone . In this paper , we refer to these patterns as “collective modes” of the retinal ganglion cell population , and we investigate a model that explicitly captures their structure . A natural way to incorporate such structure into a statistical model is by assuming the existence of a discrete set of hidden states representing the set of collective modes in the population , which we denote by the symbol α . Modes were therefore described by a latent variable of our model that changed across time , α ( t ) . The observed pattern of population spiking at a given time then depends on the instantaneous mode present at that time . Since the observed spiking pattern depends upon unobserved stimulus- and noise-driven fluctuations , the spiking pattern {σi ( t ) } will be generated probabilistically by the mode α ( t ) . To model this relationship quantitatively , we introduced for each mode a probability distribution over spiking patterns , Qα ( {σi} ) . This is the conditional distribution of the spike pattern given the mode , and is termed the mode’s emission distribution . The mean of this distribution assigns to each cell , i , a mode-dependent firing probability miα = E[σi|α] , where σi = 1 if cell i spiked in a time bin , and σi = 0 otherwise ( see Methods ) . By associating to each mode a unique pattern of firing probability , with different subsets of cells likely to be active in different modes , many spiking patterns can be combined in a flexible way . Therefore , a latent variable model can , in principle , capture arbitrary patterns of high-order correlation among cells , even without incorporating complex correlations into the emission distributions Qα ( {σi} ) . However , we found that model performance was improved by the addition of weak mode-dependent correlations ( see below , and Methods ) . After learning the model parameters ( see Methods ) , we could then invert the model to infer the probability of each mode as a function of time ( Fig 1C ) . Our techniques therefore allowed us to transform the observed population spike raster ( Fig 1A ) into a simpler representation , the temporal sequence of modes ( Fig 1C ) . Formally , we modeled population spiking activity with a hidden Markov model [50] ( Fig 2 ) . In order to better understand the full structure of this model , it is instructive to introduce features of the model sequentially . In the limit in which there are no temporal dependencies , this form reduces to a mixture model , in which the static probability distribution is described as a weighted sum over a set of emission distributions ( Fig 2A ) . If there are no mode-dependent correlations among cells , then emission distributions are simply a product over each mode-dependent response probability ( Fig 2A , i ) . However , we can capture some of the effects of neural correlation given a particular mode by adding pairwise correlations having a tree structure to the emission distribution ( Fig 2A , ii ) . These pairwise correlations are characterized by a joint response probability , pα ( σi , σj ) . We can represent this emission distribution with a graph having nodes for each cell and edges for each joint probability ( Fig 2B ) . This graph has the structure of a tree , meaning that there are no loops formed by the edges of the graph . Our full model included temporal correlations within ganglion cell spike trains through a non-uniform probability to transition from one mode to another mode in the subsequent time bin ( Fig 2B ) . This probability of a transition between modes in adjacent time bins , P ( α ( t ) |α ( t−1 ) ) , is termed the transition matrix . The model was completed by specifying the emission distributions , Qα ( {σi} ) . We found that the simplest choice of independent emission probabilities predicted too-small correlations between cells , necessitating the inclusion of correlations into the emission probabilities . In order to incorporate weak correlations while still maintaining tractability of the model , we parameterized the emission probability distributions by constraining joint distributions on pairs of neurons chosen to form a tree ( Fig 2B ) . With this choice , correlation coefficients between neuron pairs fall off exponentially with the number of tree links separating the two neurons ( see Supporting Information File ) . Most neuron pairs were thus only weakly correlated within a given mode , and the overall correlation structure was captured by the fluctuating hidden mode . Intuitively , one can think of the visual stimulus as determining which mode is activated , in which case the tree structure represents noise correlation while the presence of multiple emission distributions induces signal correlation . For some purposes , we would like our model to estimate a static ( time-independent ) probability distribution . To this end , we calculated the weights , {wα} , that solved the detailed balance equation ( Fig 2C ) and then used the mixture model formed with the fitted emission distributions ( Fig 2A ) . A latent variable model provides great flexibility in capturing arbitrary dependency structures , but the costs of this flexibility are ( i ) the possibility of overfitting to training data , and ( ii ) of needing an intractably large number of hidden states to accurately model the data . As an extreme example , experimental data from N neurons could be perfectly reproduced by a model with 2N hidden states , one for each possible word . We therefore controlled overfitting by selecting the number of modes by a cross-validation procedure ( Fig 2D ) . For the natural movie recording presented in Figs 1–8 , 70 modes optimized the cross-validated likelihood . We note that there were fewer modes than cells ( N = 152 cells in this data set ) , suggesting that the complexity of the model would remain relatively low even for large populations . We next analyzed the fitted parameters of the model in order to understand the structure of the response probability distribution and its decomposition into modes . The structure and distribution of the modes may be described at a high level by three quantities per mode: the overall probability weight , the centroid location , and the size . There was one mode with significantly higher probability than the others; this corresponded to the mode with the highest emission probability for the all-silent state . The remaining modes had roughly similar probability to one another ( Fig 3A ) . In addition to its overall probability wα , each mode has a location in response space given by its mean spike probability vector , miα = E[σi|α] . To visualize the distribution of these N-dimensional centroid vectors , we projected them into two dimensions by applying multidimensional scaling ( MDS ) . This projection preserves the location of the zero point , and approximately preserves Euclidean distance between points . Therefore , radial distance from zero in the MDS plane closely corresponds to the overall activity level of the mode . Modes were found dispersed throughout the MDS plane , roughly tiling most of this space . Notably , the high-activity modes appeared to pack less densely than the low-activity modes ( Fig 3B ) . Finally , we sought to examine the width of each mode , which we quantified by the entropy of the emission distribution , Sα ( see Methods ) . This measure increased with the modes’ average population spike count ⟨k⟩ , suggesting that the modes occupied an expanding region of response space at higher activity levels ( Fig 3C ) . Notice that this property is complementary to the observation that high-activity modes pack the MDS space less densely . The mode identity code therefore became coarser with increasing spike count ⟨k⟩ . However , the tremendous increase in the number of possible activity patterns at higher ⟨k⟩ potentially compensate for this coarse-binning ( Fig 3C , dashed line ) . Indeed , the total amount of response capacity increased faster than the entropy per mode ( Fig 3C , inset ) . Therefore , provided sufficiently non-overlapping mode distributions , this high capacity of activity patterns at higher ⟨k⟩ could be partitioned into a larger number of discriminable modes than at lower ⟨k⟩ . The increasing size of these regions could then support noise suppression , as in error-correcting codes . A schematic depiction of the structure implied by examining the parameters of our model is shown in Fig 3D . The probability distribution resembles a “mountain” with an overall peak given by the all-silent state and an overall slope that corresponds to a decrease in the probability as the spike count increases due to the sparseness of neural activity . Extending out from the central peak are localized lumps of probability—peaks and ridges—that correspond to each mode . Each local lump decreases in amplitude and increases in area as the overall activity level increases . Because the volume of response space increases with activity level ( represented by rings around the central peak with increasing circumference ) , the number of modes increases with activity level . Another important element of our HMM is the matrix of transition probabilities , P ( α | β ) , which describes the probability of finding mode α at time step t given the presence of mode β at time step t-1 . This matrix was dominated by the diagonal elements ( Fig 4A ) , which introduced a persistence of each mode across several time bins ( 20 ms ) . Because the temporal kernel of ganglion cell receptive fields typically has a width of ~100 ms [30] , it is expected that the activity state of the ganglion cell population exhibits persistence on this time scale . The median probability of remaining in the same mode on a subsequent time step was 0 . 70 . The average dwell time spent in the same mode ranged from 20 ms to 77 ms . We also found weaker , distributed off-diagonal transition elements ( Fig 4B ) . These off-diagonal elements reflect temporal correlations in the stimulus and are expected to depend strongly on the choice of stimulus ensemble . In order to assess the structure and significance of the off-diagonal transition elements , we calculated the transition entropy , Htrans ( β ) , which measures how many modes can be accessed when starting from mode β ( defined in Methods ) . This transition entropy had a value of ~2 bits , which implies that the number of accessible states was 2Htrans ≈ 22 = 4 ( Fig 4C ) . If we removed the diagonal transition element and renormalized the probabilities , we found a much higher number of accessible states , ~24 = 16 ( Fig 4C ) , consistent with the fact that the diagonal transition element always dominated . The transition entropy was roughly constant across different modes , and there were no modes with exceptionally low values . These results are consistent with a picture in which each mode eventually transitions to a broad set of other possible modes , and hence there were no strong ‘sequences’ of modes preferentially observed in our data . Finally , we can gain more intuition about the structure of our model by investigating how individual ganglion cells are organized into collective modes . One important measure of this organization is the number of individual cells that participate in each collective mode . Another complementary measure is the number of modes containing a given cell . Because each cell has a mode-dependent firing probability , miα , that varies continuously from zero to one , both of these measures are subject to an arbitrary choice of how large miα must be in order for a cell to “participate” in that mode . Thus , we calculated these measures as a function of the threshold criterion , θ=miα/m¯i , where m¯i is the average firing probability for cell i across all the modes ( Fig 5A ) . At θ = 0 , these values matched the total number of cells and modes , respectively , and they decayed monotonically to zero for large θ . At a threshold of θ = 3 , the number of cells participating in a given mode showed a highly skewed distribution with an average of 13 . Another way of viewing these statistics is to note that the majority of the modes had 10 or fewer cells participating , and less than 30% had more than 20 cells participating . Thus , collective modes had a broad distribution of overall firing , and there was only a small fraction of modes with more than 20 out 152 cells participating . In contrast , we can see that the average number of modes containing a given cell had a roughly normal distribution around an average of 6 . Thus , individual ganglion cells can belong to multiple modes , and this partitioning is relatively even across the different modes . We found that this model closely reproduced the statistics of the data . Model performance was assessed by training on half the data , and evaluating goodness-of-fit on the other half using the log-likelihood measure . The log-likelihood is given by L=∑wfemp ( w ) lnPmodel ( w ) , where w indexes each unique binary spike word in the population , femp ( w ) is the empirical frequency of that word in the test data , and Pmodel ( w ) is the probability predicted by the model . For the full HMM , which includes temporal structure , we took Pmodel ( w ) to be the stationary distribution ( see Fig 2 ) . The overall cross-validated likelihood of our model compared favorably to previous state-of-the-art models ( Fig 6 ) , including the K-pairwise maximum entropy model of [25] , which incorporates constraints on the mean spike rates , pairwise correlations , and spike-count distribution , and the reliable interactions ( RI ) model [51] . Because our HMM also reproduces some of the temporal structure of the data , we expect that the full likelihood of the HMM to be even better than indicated here . Our model was as good as the RI model ( with a word-count threshold parameter nRI = 2 counts ) in reproducing the probabilities of words appearing at least two times in the data ( Fig 6B ) . We note that our HMM effectively reached the upper bound on possible performance for this metric , because the RI model’s performance on many of these words reached sampling error , by construction . Our model has the advantage over the RI model of being a well-defined probability distribution , whereas the RI model is a non-normalized pseudo-likelihood . For example , many words occurring only once in the data were assigned “probabilities” greater than one by the RI model . The HMM is therefore more amenable to interpretation and statistical sampling for simulation . As a detailed assessment of goodness-of-fit , we compared our model’s predictions to the actual values ( measured from a held-out test set ) of many statistical quantities , again using the stationary distribution of the HMM . The low-order moments were well-reproduced ( Fig 7A–7C ) : the coefficient of determination between model-predicted and true pairwise correlation coefficients was r2 = 0 . 95 for pairwise correlations and 0 . 80 for triplet correlations . The incorporation of tree-structured correlations into the emission distributions improved these values relative to a model with only independent emission probabilities; in the latter case the pairwise correlations had r2 = 0 . 92 , and the triplet correlations had r2 = 0 . 74 . Moreover , the model accurately predicted probabilities of individual population words ( Fig 7D and 7E ) and the probability distribution of population spike count , k=∑iσi ( Fig 7F ) . The latter is a sensitive measure of the degree of high-order correlation [25] . The empirical frequency distribution of population words was highly skewed , and approximated a power law distribution ( see Supporting Information File ) . Indeed , in this particular dataset , words occurring once each accounted for 90% of the unique observed words , and 38% of the probability mass . The model-predicted probability of such low-count words , and , of course , of the many unobserved words , is poorly constrained by the data . The consequent underestimation of such words’ probability ( Fig 7E ) translated into a slight downward bias in the predicted second- and third-order moments ( Fig 7B and 7C ) . However , there was no explicit constraint on these moments in fitting the model ( in contrast to the maximum entropy paradigm [6 , 25 , 53] ) and the slight inaccuracy in their prediction was compensated by good performance on the frequently occurring words ( Fig 7D ) . Furthermore , a parametric bootstrap analysis showed that the degree of mismatch between predicted probability and empirical frequency in the low-count words was largely consistent with sampling error in a highly skewed distribution ( Fig 7E , gray band ) . We have demonstrated that a hidden Markov model , based on the assumption that neural population responses may be grouped into a collection of discrete modes , constituted an excellent statistical description of our data . Nevertheless , the success of the model is not in itself evidence that the population responses form discrete clusters , since our model contains no constraint preventing the recovered modes from overlapping to an arbitrary degree . Moreover , it is not a priori obvious that clustering structure , if present , is due to any nontrivial processing within the retina beyond the pattern of correlations induced by overlapping receptive fields . To check that the modes identified by our model indeed formed discriminable clusters , we applied linear discriminant analysis ( LDA ) . We first identified the set of unique population words that were mapped into each mode by the model ( Fig 8A and 8B , top ) . We carried out Fisher’s LDA to project the set of words assigned to each pair of modes onto the one dimension that best separated them ( Fig 8A and 8B , bottom ) [54] . We then measured the discriminability of modes by d′ ( the distance between mode means along the LDA axis , scaled by the standard deviation ) and identified for each mode the distance to its nearest neighbor , i . e . the mode with smallest d′ . Such neighboring pairs typically involved the activation of non-overlapping subpopulations of cells . Furthermore , the nearest-neighbor d′ was at least 1 . 5 standard deviations for all modes , and above 2 standard deviations for many modes ( Fig 8C ) . The degree of discriminability also increased slightly with ⟨k⟩ . This is the worst case , nearest-neighbor discriminability: d′ between arbitrary mode pairs was typically higher: 1 . 5 − 4 . 5 standard deviations ( Fig 8D ) . Since the LDA procedure seeks out the one-dimensional projection of a 152-dimensional space that maximizes d′ , there is some possibility that this high degree of nearest-neighbor discriminability could be obtained by chance . We therefore repeated the same analysis after randomly shuffling the assignment of time bins to modes . These shuffled modes had robustly smaller d′ ( Fig 8C and 8D ) . We concluded from the above analyses that the modes captured by our model indeed corresponded to discriminable response clusters , especially for patterns with many spikes across the population . We next checked whether the degree of clustering identified by our model was a nontrivial feature of the data by comparing to simplified models . First , the specific patterns of synchronous spiking characterizing our modes may naively be expected to arise from receptive field overlap . We therefore constructed a linear-nonlinear ( LN ) model of the population that reproduced this effect of common stimulus drive . In order to do so , we analyzed a data set in which the retina was driven by a binary white noise stimulus , so that the receptive fields could be estimated by spike-triggered average analysis [55] . This dataset consisted of 155 ganglion cells . We fit an LN model to each neuron in the recording ( Methods ) and then simulated data by sampling from the LN model , driven by the actual white noise sequence used in our experiment and assuming conditional independence across neurons . We then fit our model to both the real data and the simulated LN data , and evaluated the model’s likelihood in 10-fold cross-validation to assess the optimal number of modes ( Fig 9 ) . There was a substantial difference between the two models , with 50 modes being selected as optimally describing the real data ( Fig 9 , blue ) but only 6 for the LN data ( Fig 9 , green ) . Finally , we verified that correlations were necessary to give rise to mode structure by randomly shuffling each spike train in the data in order to eliminate all correlation between all cells . The resulting shuffled data was best fit by just one mode ( Fig 9 , red ) , which is unsurprising since our model with one mode should , in principle , capture uncorrelated data exactly . However , this analysis does demonstrate that the mode structure is a specific consequence of correlated spiking among cells , and not a statistical artifact of sparse firing . Our model was fit in an unsupervised way , without explicitly accounting for the stimulus . Nevertheless , the time-varying mode sequence “drives” spiking responses in a manner analogous to the unknown stimulus . It is thus natural to ask whether the modes have any meaningful relationship to the stimulus , and , if so , which features of the visual input are represented . We trained the model on the white-noise checkerboard data introduced above , which was well described by 50 modes . We then estimated the receptive fields of all cells by computing the spike-triggered average stimulus [55] , and classified cells into ON- and OFF-type . We next evaluated each mode’s receptive field by taking the mode-triggered stimulus average , i . e . by averaging all stimuli preceding a time bin in which the mode was active . We split all receptive fields ( cell and modes ) into a spatial and temporal part by finding the best-fitting separable approximation , and compared each mode’s spatial receptive field to those of the cells most active within the mode . Next , we fit the best 2D Difference of Gaussian function to the spatial profile of each mode’s receptive field ( see Supporting Information File for details ) . We used the parameters of these fitting functions to divide the mode receptive fields into four qualitative types: 1 ) intersection , 2 ) union , 3 ) oriented dipole , and 4 ) independent ( Fig 10 ) . Half the modes resembled the intersection of many individual ganglion cells’ receptive fields ( 24 modes; Fig 10A ) . These modes had small , compact fields , often concentrated over only 1−4 checks . The receptive fields of the cells with highest mode-dependent firing probability miα tended to be larger than the receptive field of the mode itself , and overlapped extensively . This arrangement represents a straightforward mechanism by which the retinal population code can achieve greater spatial acuity than individual cells , by encoding highly localized stimuli through the simultaneous spiking of several cells each containing the stimulus in their receptive field . The intersection-type modes predominantly had an OFF response , and the cells overlapping the center of these were OFF-cells . This may reflect the high asymmetry of OFF- to ON-cells in our recording ( 122 OFF cells and 33 ON cells ) , which is characteristic for the tiger salamander retina [30] . Similar features were found to be encoded by the simultaneous spiking patterns of groups of 2–4 retinal ganglion cells [46] . We view our finding of intersection-type receptive fields as an extension of these previous results to the case of ~150 cells . However , we also find other qualitatively different spatial features not reported in [46] ( see below ) . In addition , the approach of studying the features encoded by specific , multi-neuronal spiking patterns [46 , 47] suffers from the difficulty that combinations of spiking and silence among increasingly many neurons become prohibitively rare to observe . In contrast , our current approach maps many neural activity patterns onto the same collective mode , and consequently our collective modes were all observed with significant frequency ( Fig 3A ) . Thus , our current approach generalizes better to large neural populations . Other modes had extended receptive fields which were larger than many of the underlying individual cells’ receptive fields . We call these union-type modes , and they occurred with both OFF- and ON-polarity ( 8 modes , 6-OFF type , 2 ON-type; Fig 10B ) . Extracting such population responses could be useful for detecting complex spatiotemporal stimulus features that require pooling over several spatial subunits . These modes might also embody a form of position invariance for the occurrence of a similar stimulus feature over a wider area . We quantified the size of intersection and union mode receptive fields by examining the receptive field radius , defined as the semi-major axis length of the 95% confidence ellipse of the best-fit 2D single Gaussian ( see Supporting Information File ) . The median radius of the union modes’ spatial receptive fields was 165 μm , whereas it was 121 μm for intersection modes , and 138 μm for independent modes ( see below ) . For cells , the median radius was 143 μm . We compared modes’ receptive field radius to the individual cells contributing to the mode by computing an average cell radius , which we denote rα ( null ) , for each mode as the weighted average of each cell’s receptive field radius , weighted by the cell’s mode-dependent firing probability ( see Supporting Information File ) . For the union-type modes , the mode receptive field was 16% larger than the average cell receptive field , as a median value , and intersection-type modes were 19% smaller than cells . Both of these differences were statistically significant ( p < 0 . 01 , Wilcoxon signed-rank test ) . Intriguingly , several modes had oriented dipole-type receptive fields ( 3 modes; Fig 10C ) . In these modes , OFF-cells and ON-cells both had high mode-dependent spiking probability . The receptive fields of these OFF and ON subpopulations , however , formed distinct spatial regions with minimal overlap . This arrangement resulted in a mode receptive field that had separate ON- and OFF-polarity subfields . These modes are therefore well-suited to detecting an oriented contrast edge . Interestingly , this orientation-selective receptive field is reminiscent of a V1 simple cell [56] . However , we note with caution that the amphibian pallium has been poorly studied , and the responses of tectal neurons–the predominant downstream targets of the retina–have not traditionally been reported to be tuned to static , oriented stimuli [57] ( but see [58] for a recent demonstration of orientation selectivity in the mouse superior colliculus ) . Nevertheless , our results demonstrate that important visual primitives emerge as statistically robust features of the retinal population code . We find this result particularly intriguing , because the stimulus ensemble was spatiotemporal white noise–i . e . , having no spatial features with orientation that were statistically over-represented . This suggests that retinal circuitry itself may be biased to extract oriented spatial features from the visual scene . Lastly , several of the modes had receptive fields that were not statistically different than expected from their constituent ganglion cells ( 14 modes , not shown in Fig 10; see Supporting Information File for more details ) . In addition , 1 of our 50 modes had a spatial profile that was too noisy to analyze further . The temporal profiles of the mode-triggered averages qualitatively resembled those of individual cells , with most having a monophasic and some a biphasic time course . Mode temporal profiles were slower than those of individual cells , which is to be expected due to the autocorrelation in the mode response . The results reported in the previous section imply that each mode corresponds to a discriminable cluster of response words . Noise in the retinal response will produce a number of distinct words , all evoked by the same input stimulus . If distinct words , related by noise , preferentially fall within one mode’s cluster , then the mode identity will be robust to noise–i . e . , the mode representation would constitute a form of error-correcting code . To test whether this was the case , we evaluated the reproducibility of modes across stimulus repeats . We evaluated the model’s fit to several different stimulus classes featuring identical , repeated segments: a natural movie ( 170 cells , 100 modes , 73 repeats ) , a dark bar extended across the recording area , randomly moving in the transverse direction ( 140 cells , 30 modes , 108 repeats ) , and the white noise checkerboard stimulus analyzed above ( 155 cells , 50 modes , 69 repeats ) . All three stimuli featured a long period of non-repeated stimulation interspersed with the shorter repeats . We fit the model to the non-repeated data and evaluated reproducibility on the repeated portion . The same modes were frequently activated on different stimulus repeats , with especially robust reproducibility apparent at high population activity level k . An example portion of the bar stimulus is shown in Fig 11A , where each color indicates a mapping of the neural activity pattern onto the same mode . Generally , mode reproducibility tended to degrade during ( i ) periods of quiescence ( time bins with no more than two cells spiking ) , and ( ii ) transitions between modes . Both results are expected a priori . During periods of quiescence , too few cells are spiking to embody much robustness to noise , and during transitions between modes , there must always be one point in time in which the likelihood of the earlier and later modes is exactly equal . To quantity the overall degree of reliability of modes , we generated a binary vector for each mode , which was 1 when the mode was active at a given time bin and 0 otherwise . We then calculated , for each mode , the fraction of the mode’s entropy that was informative about the stimulus ( see Methods ) . This “information efficiency” measure had a median of 0 . 57 for the natural movie and 0 . 60 for the bar stimulus; moreover , it was above 0 . 80 for some modes . Information efficiency was lower for the less-structured white noise , with a median of 0 . 44 , but still comparable to the other stimuli for many modes ( Fig 11B ) . Moreover , the information efficiency of modes exceeded that of individual cells . For the natural movie , the median efficiency of single cells was 0 . 41 ( c . f . 0 . 57 for modes ) . The most efficient single cell reached 0 . 68 , while the best mode had an efficiency of 0 . 81 . Since the natural movie stimulus is most representative of the retinal input under ecological conditions , and it evoked the largest number of modes , we focused further analysis on this dataset . We sought to compare the reproducibility of individual population spiking words to the reproducibility of modes . Since most individual words occur very rarely , and not at all on some repeats , the above information measure is poorly estimated for words . We instead quantified the reproducibility of words ( modes ) as the fraction of repeats on which a word ( mode ) was active at a particular time bin , given that the same word ( mode ) occurred within 80 ms on a different repeat . Individual words were highly variable , with reproducibility decreasing with k and dropping to its minimal possible level by k = 5 ( Fig 11C ) . In fact , high-activity words often occurred only once each in the entire dataset . Nevertheless , mode reproducibility was substantially higher and increased slightly with activity level ( Fig 11C ) . We also quantified noise in the underlying population response by the average Hamming distance between pairs of words occurring at the same time bin within the stimulus , but on different repeats . By this measure as well , population responses were quite noisy , with Hamming distance exceeding spike count k and increasing with activity level ( Fig 11D ) . This result is consistent with the variability in the spike count of individual single ganglion cells [59] and provides additional insight into the low reproducibility of high-activity words . Certain trivial effects may explain the high level of mode reproducibility . We sought to control for these by carrying out two manipulations comparing the mode assignment produced by our model to randomized modes preserving only limited structure ( Fig 11E ) . Details of the construction of both controls are reported in Methods . First , there is a “similarity effect”: the model assigns to the same mode words which are similar to one another or that occur close in time ( due to the temporal correlation incorporated into the model ) . This grouping by similarity and temporal proximity may automatically reduce noise . We constructed a control which randomized the location of each mode in response space–its mean–while approximately preserving its entropy , as well as preserving the overall form of the HMM ( including temporal correlations ) . To ensure that the temporal correlations were optimal relative to the newly constructed modes , we re-fit the full HMM while fixing the randomized mode means . We will refer to this construction as the “shuffled means” manipulation . It groups words by similarity to the randomized means and by temporal proximity induced by the learned temporal correlations , and thereby reproduces the similarity effect . This control produced modes that were significantly less informative than the true modes identified by our model ( Fig 11E ) , with a median information efficiency of 0 . 37 . The fact that this randomized control still carried some information likely reflects the underlying reproducibility of single cells , since some of the random modes ended up being highly selective for certain individual cells . Indeed , the median single-cell efficiency was 0 . 41 , quite close to the value for the shuffled means control . Another possible explanation for the high degree of reproducibility is that the mode representation compresses tens-of-thousands of words into a much smaller number of modes . This large reduction of entropy may be expected to enhance reproducibility by chance . To control for this “compression effect” , we constructed modes in which each word was assigned to a random , unique mode , while matching the overall number and probability distribution of modes to that in the full model . We refer to this control as the “random partition” . We found that modes defined by random partition had very poor information efficiency , with a median of 0 . 17 ( Fig 11E ) . The random partition modes in fact had a reproducibility near chance , here defined by randomly permuting the time bins within each stimulus repeat before calculating information efficiency on the original basins . This chance value had a median of 0 . 14 . The only randomized modes that were comparably informative to the true modes were those that contained the all-silent state and other very low spike count words . There is also a “word reliability effect”: although most individual words had extremely low reproducibility ( Fig 11D ) , some were more reliable across repeats ( in particular , the all-silent state–which is also overwhelmingly the most frequently occurring word ) . Therefore , it is conceivable that modes automatically inherit this reproducibility , even without detecting any special structure . Both randomized manipulations control for this reliability effect . Together , the poor information efficiency of both of these controls ( compared with that of the true modes ) suggests that there is an underlying structure to the data that is captured by the full model , but not by arbitrary mappings of activity patterns onto modes . Thus , collective modes form a representation of the visual stimulus that is highly robust to noise in the activity of individual neurons .
There are two distinct and complementary ways to create a probabilistic model of the retinal output ( or , indeed , of any sensory system ) . One approach attempts to explicitly capture the dependence of neural responses on the stimulus . Models of this class include standard linear-nonlinear ( LN ) models , generalized linear models ( GLMs ) [26] [60] , and stimulus-dependent maximum entropy models [61] . These models , which we will call “encoding” models , have two primary advantages: they may highlight the circuit-level computation by which the retina transforms its input , and they provide a principled basis for deriving decoders that transform the retinal output into a reconstruction of the stimulus [26 , 62] . Mathematically , all such models are parameterizations of the conditional distribution of responses on stimulus , P ( {σi ( t ) }|s ( t ) ) , where s ( t ) denotes the stimulus . Many versions of such models also have the benefit that they can be applied to arbitrary patterns of light , putting them at the highest level of generality . The drawback of this approach is that these models do not correctly predict ganglion cell spike trains in many , if not most , visual conditions . To give a few examples: 1 ) LN models ( and their generalizations including ganglion cell gain control ) do a good job predicting the firing rate for objects moving at constant velocity [63 , 64] , but fail for discontinuities of motion [65–67] or for wide field motion [68 , 69]; 2 ) LN models ( and their generalizations including refractory periods ) do a good job for random flicker that is spatially uniform or nearly uniform across the receptive field center [26 , 70 , 71] , yet fail for random flicker with small checkers [72]; 3 ) even the primate midget and parasol cells , which seem more closely linear than most other ganglion cell types , require models with multiple gain control stages to predict the firing rate for temporal-chromatic natural movies [73] . The picture that emerges from these studies is that there are multiple sources of non-linearity within the retinal circuit that are significant in some ( but not all ) visual conditions and that endow ganglion cells with light responses qualitatively different from the LN model . Examples include bipolar cells with rectification and synaptic depression in their axon terminal [74–76] as well as different types of amacrine cells that veto ganglion cell response when specific visual features are present [68 , 77–80] . Given the extreme diversity of amacrine cell types [81] as well as the fact that they can individually remold each bipolar-to-ganglion cell transfer function [82] , the prospect for deriving an input-output model that will succeed over the entire range of natural visual conditions currently seems quite remote . An alternative approach–the one adopted in the present work–is to directly model the structure of retinal responses , without reference to the stimulus . Mathematically , these models represent the distribution P ( {σi ( t ) } ) . Models of this form , which we call “activity” models , have their own advantages , complementary to those of encoding ( stimulus-dependent ) models . Most importantly , they correspond to the problem faced by the rest of the brain , which lacks direct access to either the stimulus or the internal structure of the retina . These models , rather than highlighting the circuit-level computation , instead emphasize the structure of the neural code produced by the system . Such activity models address the question of how the information conveyed to the rest of the brain is formatted by the retina , and therefore how it could be most naturally interpreted . Note that this question is quite distinct from the notion of stimulus decoding–in our view , the goal of the visual system is decidedly not to reconstruct the exact light pattern comprising the stimulus , which anyway already exists in the photoreceptors . Rather , the goal is to amplify behaviorally relevant stimulus features while suppressing irrelevant ones–e . g . , the operation of recognizing the same object in the environment under different viewing conditions [83]–which is ultimately a complex function of information encoded at the photoreceptor level . Activity models provide neuroscientists with data-driven hypotheses of what these features may be . Our results reported in the present paper suggest that the collective modes are one example of a relevant signal emphasized at the retinal ganglion cell level; there could be others as well . One advantage of activity models is that they can be formulated in circumstances in which there exists no satisfactory encoding model , such as under natural movie stimulation . The method is simply to measure simultaneously the activity states of the neural population over a stimulus ensemble , as is possible with techniques like multi-electrode recording or optical imaging . In taking this approach , we were able to identify many more collective modes in response to checkerboard flicker than could be identified with matched LN models of each neuron ( Fig 9 ) . This result demonstrates the benefit of direct experimental measurement of neural activity over inferences from encoding models , at least in the context of our study . A consequent obvious disadvantage of the approach based on activity models is that the distribution of states , P ( {σi ( t ) } ) , will depend strongly on the choice of stimulus ensemble . However , it should be pointed out that encoding models also suffer implicitly from this difficulty , as the form of model that succeeds for one stimulus ensemble may not succeed in others . This limitation is a consequence of the aforementioned circuit complexity arising from gain control , amacrine cell network interactions , and other effects . Finally , we note that even a perfect encoding model–which would necessarily be far more complex and sophisticated than the present state-of-the-art models–would require an analysis along the lines of the present work in order to address questions at the level of the output neural code that we have studied here . Our modeling work suggests that neural circuits downstream of the retina can extract meaningful information about the visual stimulus merely by clustering retinal activity patterns into a sequence of collective modes . Our model enabled us to extract population modes that were informative about the stimulus and robust to noise . Since the modes were defined without any information about the stimulus ( i . e . , its repeat structure ) , the reproducibility results of Fig 11 are highly nontrivial . Furthermore , these modes often encoded features of the stimulus that were not represented by any individual neuron . How much information is carried by the sequence of modes ? For our non-repeated natural movie stimulus , with 70 modes , the modes carry 2 . 0 bits/s per neuron of entropy . This is about 25% of the maximum-entropy upper bound on total response entropy estimated in [25] for a separate dataset of salamander ganglion cells responding to a natural movie . There are three possible explanations for this gap , which may be explored in future work . First , much of the population entropy is noise: estimating noise entropy for the dataset used in [25] , we find that almost 60% of the population entropy is noise entropy ( this is an upper bound , assuming neurons are conditionally independent given the stimulus ) . Since we found that modes are less noisy than population words , we expect a smaller fraction of the mode entropy to be noise . Second , the modes as we have defined them may only represent one channel of meaningful stimulus information . Further information may be represented as substructure of the modes , or as a distinct partitioning of the population activity . Such a multiplexed population code would then allow different downstream targets to readily extract different information streams from the same input . For example , given that we find substantial variation of population spike count within each mode , one hypothesis may be that mode identity represents a particular stimulus feature while population spike count encodes the contrast or intensity of that feature . Other examples of such multiplexed encoding are known to exist . In the retina , contrast adaptation leads to a multiplexed neural code: the visual information encoded by individual spikes is roughly invariant to the contrast , while the total firing rate encodes the absolute contrast on a longer time scale of ~10 sec [84 , 85] . In the visual cortex , information about different aspects of the stimulus–such as the spatial phase , contrast , and orientation of gratings–is represented at different time scales in the same cortical spike trains [86 , 87] . Finally , we may be vastly underestimating the total number of modes , as we are limited by finite sampling . With greater sampling–note that our recording times of ∼ 1 hr are orders of magnitude lower than what’s available to the organism under natural conditions–many more rare words would be expected to occur . These new words would form new modes , and the entropy of the probability distribution over modes would increase . In contrast , the aforementioned maximum entropy bound should be fairly stable against improved sampling , since it’s based on a maximum entropy model constrained on a set of consistent statistics ( indeed , as more structure is revealed , the true population entropy may decrease further below the maximum entropy bound ) . Likewise , sampling a larger fraction of the cells overlapping the patch of visual field would likely yield a larger number of more readily distinguished modes . In contrast , the entropy per cell reported in [25] continues to decrease with neuron number , apparently not saturating until 200 − 300 neurons . Therefore , we expect the fraction of total entropy captured by the modes to increase with sampling in both time and cells , provided that the observations we have made about the response distribution are generalizable to unobserved data . One approach to identifying error-robust sets of codewords was suggested by the application of maximum entropy ( MaxEnt ) models that constrained the pairwise correlations among retinal ganglion cells [6] . The pairwise interactions learned by fitting these models typically feature both positive and negative interaction strengths . In models of physical magnetic systems , such disordered interactions are known to lead to a probability distribution featuring many local peaks [88] . “Error-corrected” codewords have been identified in retinal data as local maxima of the probability landscape defined by MaxEnt models , and indeed were shown to encode reliable information [25] . Fitting maximum entropy models is computationally demanding , especially for datasets from a large number of cells . Moreover , the definition of local probability maxima depends on an arbitrary choice of metric distance between population states–such as the Hamming distance–which may not weigh cells in a way that facilitates optimal discrimination . We therefore adopted a more direct approach to extracting the robust collective representations of the retinal population code , by exploring a model of population neural activity that explicitly incorporates dependence of the observed spike patterns on a hidden variable identifying the population mode . The coding symbols identified in [25] persisted with an average timescale of 48 ms , consistent with the dwell time of modes that we discovered by our methods . Therefore , we predict a close relationship between the two constructions . Latent variable models have been successfully applied to other neural systems; several related approaches and findings have recently been reviewed [89] . Dimensionality reduction methods applicable to single-trial spiking data , the paradigm adopted in our work , include continuous state-space models in addition to HMMs . Hidden Markov models have been employed to capture population spiking in a variety of cortical structures , including frontal [90–92] , premotor [93] , gustatory [94] , and somatosensory areas [91] . A similar latent-variable-based approach , restricted Boltzmann machines , were applied to visual cortex in [95] . Relatedly , a clustering algorithm was employed to show that auditory cortical dynamics were organized into transitions between discrete states [96] . Our methodology extends the prior applications of HMM , which assumed conditional independence of different neurons within each hidden state , by introducing correlations into the emission distribution in a way that enables efficient inference . Moreover , that a discrete state-space model such as HMM should apply to the retina , as we have demonstrated , is perhaps more surprising than in cortex , where such a discrete structure may be expected as a consequence of recurrent attractor dynamics or global brain state fluctuations . In the present work , we have argued that such discreteness of the population response is desirable in early sensory stages for functional reasons , facilitating error-robust encoding of information . An important question in extending the present model to larger neural populations is how the number of modes scales with the number of neurons . For the dense , local patch of 100–200 retinal ganglion cells sampled in the experiments analyzed in this paper , we found that the optimal number of modes was less than the number of neurons . However , we expect this result to be a consequence of the high degree of redundancy among these cells , due to their mutual receptive field overlap . If cells are not so highly correlated , the number of modes may scale faster . For example , for our model to describe two completely non-overlapping and statistically independent retinal patches would require a number of modes equal to the product of the number of modes in each patch . To describe such data , future work could explore a hierarchical model tying together several HMMs into deeper layers , modeling structure at a variety of scales . More generally , the appropriate latent variable model for a given neural population will depend on the balance between redundancy and synergy . We expect our results to apply to predominantly redundant populations elsewhere in the brain . The general problem faced by the retina is to extract relevant features from a complex , high-dimensional input , and then to package those into output spike trains in a way that facilitates downstream recognition of the relevant features while suppressing irrelevant features ( e . g . , noise ) . This way of thinking about neural computation clearly generalizes beyond the retina , applying broadly across the nervous system . The techniques introduced in the present work allow such hidden relevant features to be explicitly inferred from output spike trains alone , and our application of these ideas to the retina forms a simple test case in which the input–the visual stimulus–is well understood . Future work could apply our model to analyze the collective mode structure of more complex , higher-order , brain areas where the relevant features of the input are unknown . For example , in a study that recorded from large populations of neurons in the inferotemporal cortex while the monkey was viewing different categories of objects , the activity states of the neural population clustered into categories that matched the object categories [97] . In both this study and in ours , the clustering of activity states was only observed in large populations . Measurement at the scale of hundreds of neurons may reveal this structure in other brain areas as well . In cortical areas involved in higher-level sensory perception or decision making , the partition of the input into “relevant” and “irrelevant” features could be dynamic , depending on behavioral demands . It would be quite interesting to apply our model to such areas to see how the mode structure changes with attentional state or task structure . Our approach also reduces complex population activity to a single , scalar , codeword–the mode identity . In brain areas where individual neurons demonstrate difficult-to-interpret mixed selectivity , it may be the case that the collective representation is more explicit than that of any single neuron [98] . Our model provides a straightforward method to test similar hypotheses and to explicitly identify the relevant population codewords in spiking neural populations across the nervous system .
This study was performed in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . The protocol was approved by the Institutional Animal Care and Use Committee ( IACUC ) of Princeton University ( Protocol Number: 1828 ) . We recorded responses from larval tiger salamander retinal ganglion cells using a custom 252-electrode multi-electrode array ( MEA ) . For all but the natural movie experiments , eyes were removed in darkness and the retina separated from the pigment epithelium . The natural movie experiments were conducted with the pigment epithelium intact and surgery performed in dim light . The isolated retina was then placed onto the array and superfused with oxygenated Ringer’s medium at room temperature . Offline spike sorting was performed with custom software . Detailed recording and analysis methods , including the spike-sorting algorithm , are described in [1] . Stimuli were projected onto the array from a CRT monitor at 60 Hz . We presented three stimulus classes , in different experiments: a natural movie , a binary white noise checkerboard , and a diffusively moving dark bar . The natural movie consisted of a 7–minute gray scale recording of leaves and branches blowing in the wind . We conducted natural movie experiments with two different designs: in the first , the movie was looped with a different pixel centered on the recording area for each repeat . Since the patch of retina recorded by the MEA subtended only a small portion of the stimulus , we were able to construct this stimulus in such a way that the retinal input was effectively non-repeated over the full recording session . In a second design , we generated a similar “non-repeated” movie , but alternated every 60 sec of non-repeated stimulation with an identical 60 sec clip . The binary white noise stimulus consisted of a 40 x 40 flickering checkerboard , randomly updated every frame . Each checker had a size of 55 x 55 μm when projected onto the retina . The drifting bar’s motion was a random walk subject to a restoring force to keep it , on average , centered on the recording area [99] . Both the checkerboard and drifting bar designs consisted of a long period of non-repeated stimulation interspersed with repeated segments at randomly-chosen time intervals ( repeat duration = 60 sec for the bar , 30 sec for the checkerboard ) . We binned spike trains into 20 ms time bins , producing a sequence of binary spike words σi ( t ) , where i = 1…N labels the neuron identity and t the time bin . We set σi ( t ) = 1 whenever at least one spike occurred within bin t , and 0 otherwise . Because our hidden Markov model ( HMM ) incorporates temporal correlations between adjacent time bins [50] , we fit the parameters of this model to the entire sequence of binary spike words , which we denote by ( {σi ( 0 ) , … , σi ( T ) } ) ( see Fig 2 ) : PHMM ( {σi ( 0 ) , … , σi ( T ) } ) =∑α ( t ) mode sequences∏t=1TQα ( t ) tree ( {σi ( t ) } ) P ( α ( t ) |α ( t−1 ) ) , where the summation is over all possible sequences of modes from time 0 to T . Because there are M possible modes in each time bin , this sum involves a total of MT+1 terms . Here , α ( t ) and {σi ( t ) } denote the mode index and binary spike word in a single time bin , t , respectively . The transition matrix was assumed to be stationary , i . e . independent of t: P ( α ( t ) |α ( t−1 ) ) ≡ P ( α|β ) . The emission probabilities Qαtree ( {σi} ) were defined by constraining joint probabilities between certain neuron pairs as follows: Qαtree ( {σi} ) =∏icellspα ( σi ) ∏〈i , j〉edgespα ( σi , σj ) pα ( σi ) pα ( σj ) , where the pairs indexing the second product , ⟨i , j⟩ , are chosen to form a tree ( Fig 2B ) . This tree constraint is necessary for the factorized form to give a correctly normalized probability distribution; the choice of tree topology , however , is a priori arbitrary and was learned during the fitting step . The motivation behind this choice was that these interaction terms help describe correlations within the neural population , yet have a form with parameters that can be readily fit to data . In the statistics literature , this emission distribution is known as the Chow-Liu tree [100] . In our model , the tree topology may differ for each mode α . The parameters of the full model are therefore: M , the number of modes; the M × M transition matrix P ( α | β ) ; the choice of tree topology for each mode; and the collection of joint probabilities on each mode’s tree , pα ( σi , σj ) . The latter may be parameterized by a mode-dependent spiking probability for each cell , miα = pα ( σi = 1 ) , and a pairwise correlation for each tree edge , Cijα = pα ( σi = 1 , σj = 1 ) , giving M ( 2N − 1 ) parameters describing the emission distributions . Finally , we introduced a single regularization parameter η , described below . After fitting the full model , we inferred the mode that was active at each time bin of the data by maximizing the posterior probability of the mode sequence , conditioned on all the observed spike train data from the beginning of the experiment up until time t: α* ( t ) = argmax PHMM ( α ( t ) | ( {σi ( 0 ) , … , σi ( T ) } ) ) . This maximization was implemented by the Viterbi algorithm [101] . For some purposes , we wanted to reduce the full HMM to a static ( time-independent ) probability distribution , which describes the probability of occurrence of a single binary spike word . This static probability distribution had a mixture form ( see Fig 2C ) : Pmix ( {σi} ) =∑αmodeswαQα ( {σi} ) . where we used the set of tree emission distributions fit for the full HMM along with mode-dependent weights , {wα} , defined by the detailed balance equation: wα=∑βmodesp ( α|β ) wβ . For fixed M , we inferred the parameters by maximum likelihood , using the Baum-Welch algorithm with an M-step modified to accommodate the specified form of Qα [50] . Full details are reported in the Supplement . To mitigate overfitting , we introduced a regularization parameter η ∈ [0 , 1] ( see Supporting Information File ) . The above maximum likelihood fitting was carried out for fixed M and η Increasing M and decreasing η both increase the complexity of the model , and lead to a strictly improved likelihood on the training data . To choose these parameters we carried out an n-fold cross-validation procedure; generating n non-overlapping splits of the data into training and test sets . M was then chosen to maximize the test set likelihood , averaged over the n folds ( Fig 2D ) . In practice , we chose n = 2 . Our results were little affected by the precise value of η , provided it was small but nonzero . We used η = 0 . 002 throughout the present work . To calculate the entropy of the emission distributions plotted in Fig 2 , we used an analytical formula obtainable from the above-factorized probability distribution: Sα=−∑icells∑σi[0 , 1]pα ( σi ) log2pα ( σi ) −∑〈ij〉edges∑σi , σj[0 , 1]pα ( σi , σj ) log2pα ( σi , σj ) pα ( σi ) pα ( σj ) . After fitting the full model ( see above ) , we were able to obtain all parameters necessary to evaluate this equation exactly . To calculate the maximum entropy possible for an emission distribution with a given average spike count , <k> , we started by observing that the maximum entropy would be obtained when all possible states were equally likely . For an integer spike count , this uniform emission distribution is given by: Quniformk ( {σ} ) =1C ( N , k ) = ( N−k ) ! k ! N ! . Thus , the maximum entropy is given by: Smax ( 〈k〉 ) =−∑{σ}Quniform〈k〉 ( {σ} ) log2Quniform〈k〉 ( {σ} ) =log2Γ ( N+1 ) Γ ( N−〈k〉+1 ) Γ ( 〈k〉+1 ) , where Γ ( N ) is the Gamma distribution . To characterize the statistics of how one mode transitions to another mode in the next time step , we calculated the transition entropy: Htrans ( β ) =−∑αmodesp ( α|β ) log2p ( α|β ) . This quantity has the interpretation that from mode β , transitions can be made to roughly 2Htrans different modes . Because self-transitions dominated , we also calculated the transition entropy using transition probabilities with only off-diagonal components: p˜ ( α|β ) =p ( α|β ) / ( 1−p ( α|α ) ) for α ≠ β . For each mode , we computed r ( t ) , the fraction of stimulus repeats on which the mode was active at time bin t , and its time average r¯=1T+1∑tr ( t ) . We then defined the output entropy of mode occurrence , Sout , and its noise entropy , Snoise , as follows: Sout=−r¯log2r¯− ( 1−r¯ ) log2 ( 1−r¯ ) Snoise=−1T+1∑t=0Tr ( t ) log2r ( t ) + ( 1−r ( t ) ) log2 ( 1−r ( t ) ) . The information efficiency was defined to be the mutual information divided by the output entropy: ( Sout−Snoise ) Sout . For each cell individually , we fit a linear-nonlinear ( LN ) model with a logistic nonlinearity: pi[s ( x¯ , t ) ]=11+exp ( αKi ( x¯ , t ) ∗s ( x¯ , t ) −θ ) , where pi[s ( x¯ , t ) ] is the probability of the cell i spiking within time bin t , given the stimulus s ( x¯ , t ) in that time bin , and spiking is modeled as independent ( Bernoulli ) across time bins . The stimulus s ( x¯ , t ) and the linear filter Ki ( x¯ , t ) contain both spatial ( pixels ) and temporal ( past time bins ) dimensions , with Ki*s denoting summation over the spatial dimensions and convolution over the temporal: K ( x¯ , t ) ∗s ( x¯ , t ) =∑x¯pixels∑τtimeK ( x¯ , τ ) s ( x¯ , t−τ ) , where τ indexes time bins and x¯ spatial pixels . To fit the model , we first set the linear filter Ki ( x¯ , t ) equal to the spike-triggered average in a binary white-noise checkerboard experiment . In our experiments , the receptive field centers typically occupied a small region of only 2–4 checks across , while the stimulus was 40x40 checks . We therefore set all checks outside of a 7x7 patch centered on the receptive field peak to zero , in order to suppress noise introduced in the spike-triggered average estimate . The remaining two parameters α and θ were then estimated by numerical maximum likelihood on the model . | Neurons in most parts of the nervous system represent and process information in a collective fashion , yet the nature of this collective code is poorly understood . An important constraint placed on any such collective processing comes from the fact that individual neurons’ signaling is prone to corruption by noise . The information theory and engineering literatures have studied error-correcting codes that allow individual noise-prone coding units to “check” each other , forming an overall representation that is robust to errors . In this paper , we have analyzed the population code of one of the best-studied neural systems , the retina , and found that it is structured in a manner analogous to error-correcting schemes . Indeed , we found that the complex activity patterns over ~150 retinal ganglion cells , the output neurons of the retina , could be mapped onto collective code words , and that these code words represented precise visual information while suppressing noise . In order to analyze this coding scheme , we introduced a novel quantitative model of the retinal output that predicted neural activity patterns more accurately than existing state-of-the-art approaches . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"medicine",
"and",
"health",
"sciences",
"action",
"potentials",
"markov",
"models",
"membrane",
"potential",
"ocular",
"anatomy",
"electrophysiology",
"neuroscience",
"probability",
"distribution",
"mathematics",
"ganglion",
"cells",
"thermodynamics",
"entropy",
"animal",
"cells",
"hidden",
"markov",
"models",
"probability",
"theory",
"physics",
"cellular",
"neuroscience",
"retina",
"cell",
"biology",
"anatomy",
"physiology",
"neurons",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"physical",
"sciences",
"ocular",
"system",
"afferent",
"neurons",
"retinal",
"ganglion",
"cells",
"neurophysiology"
] | 2016 | Error-Robust Modes of the Retinal Population Code |
Spores are an essential cell type required for long-term survival across diverse organisms in the tree of life and are a hallmark of fungal reproduction , persistence , and dispersal . Among human fungal pathogens , spores are presumed infectious particles , but relatively little is known about this robust cell type . Here we used the meningitis-causing fungus Cryptococcus neoformans to determine the roles of spore-resident proteins in spore biology . Using highly sensitive nanoscale liquid chromatography/mass spectrometry , we compared the proteomes of spores and vegetative cells ( yeast ) and identified eighteen proteins specifically enriched in spores . The genes encoding these proteins were deleted , and the resulting strains were evaluated for discernable phenotypes . We hypothesized that spore-enriched proteins would be preferentially involved in spore-specific processes such as dormancy , stress resistance , and germination . Surprisingly , however , the majority of the mutants harbored defects in sexual development , the process by which spores are formed . One mutant in the cohort was defective in the spore-specific process of germination , showing a delay specifically in the initiation of vegetative growth . Thus , by using this in-depth proteomics approach as a screening tool for cell type-specific proteins and combining it with molecular genetics , we successfully identified the first germination factor in C . neoformans . We also identified numerous proteins with previously unknown functions in both sexual development and spore composition . Our findings provide the first insights into the basic protein components of infectious spores and reveal unexpected molecular connections between infectious particle production and spore composition in a pathogenic eukaryote .
The formation of survival structures in response to adverse conditions is an essential tool used by diverse organisms across biology to propagate life on earth . Spores are a particularly successful cell type used by many microorganisms , including bacteria , fungi , and protozoa to survive unsuitable growth conditions and/or to disperse to new environments [1] . Among eukaryotes , some of the most environmentally resistant spores are those of fungi , and much of our current understanding of spores comes from studies in model fungi such as Saccharomyces cerevisiae and Aspergillus nidulans [2] . There are two general categories of fungal spores—sexual and asexual , and both forms occur across diverse fungal species via myriad developmental strategies . For example , in the budding yeast S . cerevisiae sexual spores are formed when yeast diploids are subject to nitrogen starvation and a nonfermentable carbon source , resulting in four haploid ascospores; S . cerevisiae does not produce asexual spores [3 , 4] . In contrast , the filamentous fungus Aspergillus nidulans produces both asexual and sexual spores via the development of multicellular fruiting structures with thousands of spores per structure [5] . In all cases , spores are adapted for general survivability . As a consequence , the basic characteristics of fungal spores are constant: First , mature spores are relatively metabolically quiescent , allowing them to remain dormant for long periods of time under sub-optimal growth conditions ( e . g . in the absence of nutrients ) [2] . Second , spores are resistant to environmental stresses , such as high temperatures , desiccation , and UV radiation , thus facilitating long-term survival and/or dispersal across great distances around the globe [1] . Third , upon encountering growth-promoting environments , spores can rapidly escape quiescence and germinate to resume vegetative growth [5 , 6] . As such , spores have evolved to facilitate survival of fungal species in diverse environments , contributing to nearly ubiquitous representation of fungi across all ecosystems on earth . Spore-producing fungi commonly generate spores with thick , protective coats and robust stress resistance , due to the accumulation of protective solutes ( e . g . mannitol and trehalose ) and the production of heat shock proteins and other factors that are important for both spore stability and dormancy [2] . Spores respond to different environmental signals to initiate germination , depending on their adapted niches . For example , spores of S . cerevisiae germinate readily in response to the presence of a fermentable carbon source [6] , whereas spores of Talaromyces macrosporus require nutrients and a rigorous external trigger of very high temperature or pressure [7 , 8] . These triggers generally result in responses such as water uptake , cell wall remodeling , and activation of nutrient metabolism and protein synthesis , leading to active fungal growth [5] . The transition from dormant particle to actively growing cell is particularly important because fungal survival cannot occur in the absence of the ability to germinate when ( and only when ) appropriate for vegetative growth . Environmental fungi are well adapted to their niches , and interestingly , these adaptations have led to a handful of fungi with the ability to cause life-threatening diseases in humans . Histoplasma capsulatum , Blastomyces dermatitidis , Aspergillus fumigatus , Coccidioides immitis , Sporothrix schenkii , Penicillium marneffei , and Cryptococcus neoformans all represent environmental fungi that can cause disease in humans , and the most common route of infection is through the inhalation of cells from environmental sources [9] . Spores ( sexual or asexual , depending on the fungus ) are the most likely infectious particles for all of these pathogens; however , very little is known about their basic spore biology , making the development of disease prevention and treatment strategies challenging . Among human fungal pathogens , the most common cause of fatal disease ( and a well-developed model for study ) is Cryptococcus neoformans , a primarily opportunistic pathogenic yeast , which causes meningoencephalitis [10] . People with AIDS are particularly susceptible , and there are an estimated one million cases and 600 , 000 deaths annually worldwide from cryptococcosis [11] . C . neoformans is ubiquitous in the environment , and inhalation of aerosolized spores and/or yeast is the most common route of infection of humans [12 , 13] . Under laboratory conditions , spores are produced through sexual development between haploid yeast of opposite mating types ( a and α ) or by α fruiting . In response to specific environmental conditions , cells form filaments and fruiting bodies ( basidia ) from which haploid , recombinant spores bud in chains [14 , 15] . Spores of C . neoformans exhibit the fundamental properties of most fungal spores such as stability in the absence of nutrients and resistance to a variety of environmental stresses , such as high temperature , desiccation , and oxidative stress [16] . These spores have also been shown to germinate efficiently and synchronously in response to nutrients , and they germinate and cause disease in a mouse inhalation model of infection [17 , 18] . These findings indicate that C . neoformans spores harbor intrinsic properties that facilitate survival in the environment , maintain spore viability and stability , and initiate germination in response to external signals , including those of a mammalian host . One approach to understanding how spore-specific properties and behaviors are conferred in C . neoformans and lead to disease is to identify molecular components that contribute to spore biology . We hypothesized that proteins specific to spores would be more likely to contribute to spore-specific properties than proteins in other cell types ( such as yeast ) . The yeast of C . neoformans are the vegetative growth form and are physically distinct from spores . They also do not exhibit the same environmental resistance , dormancy properties , or germination processes as spores . To test our hypothesis directly , we carried out a proteomic analysis of C . neoformans spores and yeast to identify proteins found preferentially in spores that could contribute to fundamental spore behaviors . Here , we present the overlapping and distinct proteomes of both spores and yeast . Using these datasets , we identified spore-enriched proteins , knocked out a cohort of genes encoding eighteen proteins identified only in spores , and assessed the resulting mutants for a wide array of phenotypes . While we anticipated that these spore-enriched proteins would act in spore-specific processes ( e . g . stability , resistance , and germination ) , we discovered instead that the majority of mutants showed defects in early sexual development and spore formation . Our data indicate that many spore-represented proteins are associated with pre-spore and spore formation events during sexual development rather than in conferring intrinsic spore-specific properties , suggesting that spore-resident proteins function in both multi-cellular development and subsequent progeny survival .
To determine differences in protein composition between spores and yeast , we carried out a proteomic comparison using gel fractionation/nanoscale liquid chromatography coupled to tandem mass spectrometry ( nanoLC-MS/MS ) . Proteins were extracted from spores and yeast independently in triplicate . All extracts were subjected to one dimensional sodium dodecyl sulfate polyacrylamide gel electrophoresis ( 1D SDS-PAGE ) analysis ( S1A and S1B Fig ) , trypsin digestion , and high performance liquid chromatographic fractionation/automated tandem mass spectrometry ( LC-MS/MS ) . The resulting mass spectra were evaluated against C . neoformans proteome databases for peptide and protein identification at a 1% false discovery rate . More than 2000 proteins out of the ~6500 predicted proteins of C . neoformans were identified for both spores and yeast , and there was ~80% overlap in protein composition between the two cell types ( Table 1 and Fig 1A and S1 Table ) . To assess the overall quality of our data , we evaluated the distribution of the molecular weights and isoelectric points , the predicted intracellular locations , and the general molecular functions of all the proteins encoded by the genome and compared those to the proteins in our datasets . The distribution patterns for the proteins identified by mass spectrometry were similar to the predicted proteome overall ( Fig 1B , S1C and S1D Fig ) . Transmembrane proteins were the only major exceptions: 39 . 4% of the C . neoformans genome is predicted to encode transmembrane domains , but only 21 . 7% of the proteins in our dataset were predicted to harbor transmembrane domains . This amount of bias was consistent with previous proteomic studies using MS , indicating that no additional bias was introduced during protein extraction or recovery in our experiment [19] . The proteins detected in our experiment also included all of the proteins from previously published proteomic studies of C . neoformans secreted and cell wall-bound proteins as well as immunodominant proteins ( S2 Table ) [20 , 21] . Total numbers of proteins identified were also similar to other MS analyses of C . neoformans ( Table 1 ) [22] . Overall , these findings indicate that the quality of the yeast and spore datasets is high and very likely to accurately reflect the proteomes of each cell type . To identify proteins likely to confer cell type-specific properties , we evaluated the spore and yeast proteomes for differences using two approaches . First , we assessed proteins in a binary manner , determining only their presence or absence in all datasets . We defined proteins that were detected in at least one spore sample , but never in a yeast sample as "spore-enriched" and vice versa . Using this method , we identified 374 spore-enriched proteins . Within this group of 374 proteins , 88 and 18 proteins were detected in two or three spore replicates , respectively , but not in any yeast replicates ( Table 1 and Fig 2 and S3 Table ) . Conversely , we detected 334 "yeast-enriched" proteins in at least one replicate , 77 in two , and 16 in all three yeast replicates , but never in any spore samples ( Table 1 and Fig 2 and S3 Table ) . To evaluate the likely functional properties of yeast- and spore-enriched proteins , we carried out functional annotation clustering analysis , using the Database for Annotation , Visualization and Integrated Discovery ( DAVID ) [23 , 24] . Enrichment analysis showed that when compared to all 2560 proteins that were identified at least once in yeast or spores , the 374 spore-enriched proteins were significantly overrepresented ( p<0 . 05 ) in a variety of specific GO clusters , such as regulation of transcription , DNA metabolic processes , chromosome organization , and mRNA metabolic processes ( Fig 2 ) . In contrast , analysis of the 334 yeast-enriched proteins revealed overrepresentation in iron-sulfur cluster assembly , mitochondria function , and RNA processing and degradation ( Fig 2 ) . These patterns of representation are consistent with the concept of the spore as a cell type poised to respond to changing environmental conditions and of the yeast as an actively growing vegetative cell type . In the second analysis , we used spectral counting to estimate relative abundances of each protein in spores and yeast ( rather than a binary readout of presence or absence in the sample ) . Spectral counting is particularly useful for estimating large differences in protein composition between samples and can provide more accurate estimates than stable isotope labeling methods [25 , 26] . Specifically , normalized peptide spectral match ( PSM ) values were used to calculate a spore-overrepresentation ratio ( r ) for each protein ( r = spore PSM value divided by yeast PSM value ) . We designated 156 spore-overrepresented proteins ( r>4 ) and 317 yeast-overrepresented proteins ( r<0 . 25 ) . Spore-overrepresented proteins show GO term enrichment similar to that of spore-enriched proteins , but enriched terms also included fungal cell wall , external encapsulating structure , and Golgi apparatus . Yeast-overrepresented proteins also showed GO term enrichment similar to yeast-enriched proteins but enriched terms also included specific biosynthetic and metabolic processes ( Table 2 ) . Thus , the addition of yeast- and spore-overrepresented protein categories resulted in the identification of additional differences in GO categories in biological processes and cellular components between yeast and spores . These data further distinguish differences between the yeast and spore cell types at a molecular level . Given the unique role of spores in microbial survival , we focused on proteins likely to be important for spore-specific functions . We hypothesized that spore-enriched proteins would be more likely than other proteins to be involved in spore-specific processes such as dormancy , stress resistance , and germination . Thus , we focused on the 18 proteins that showed a spore-enriched MS identification pattern in all three spore proteome replicates . To verify that the spore-enriched pattern represented in the mass spectrometry analysis was reflected in the levels of protein in vivo , we created two strains harboring fusions between spore-enriched proteins and a fluorescent protein , mCherry . In both cases , mCherry fluorescence was visible only in spores and not in yeast , consistent with the proteomic data ( S2 Fig ) . We assigned the genes encoding the 18 spore-enriched proteins gene names based on similarity to known genes in S . cerevisiae when possible and placed them in five groups based on predicted functions ( Table 3 ) . Group 1: Replication and Chromosome Biology , 2: Transcription and Splicing , 3: Cellular Transport , 4: Carbohydrate Metabolism , and 5: Proteins of Unknown Function . Seven genes encoding proteins with no similarity to previously named proteins were named Identified Spore Protein ( ISP1-7 ) . To test our hypothesis , we deleted the entire open reading frames of genes encoding these 18 spore-enriched proteins in haploid yeast of both mating types , and the resulting mutants were evaluated for discernable phenotypes . Among the eighteen genes , 3 appeared to be essential because we were repeatedly unable to recover transformants that did not harbor a wild type copy of the targeted gene . The homologs of these genes in S . cerevisiae , IRR1 , PRP31 , and PRP11 , are all known to be essential for viability [27–29] . Among the other fifteen genes that were deleted , all of the resulting mutants were viable , including top1Δ . This was surprising because TOP1 has been shown previously to be essential for viability in C . neoformans strain H99 ( serotype A ) [30] . TOP1 was readily deleted in the JEC20 and JEC21 strains ( serotype D ) used here . Because TOP1 is not essential for survival in many fungi [31 , 32] , we surmised that unknown differences between serotype backgrounds may account for this difference in phenotype within C . neoformans . None of the other genes had been characterized previously in C . neoformans . To assess general yeast growth of the 15 viable gene deletion strains , we evaluated growth of multiple independent knockout strains for each in both mating types at 30°C under nutrient-rich growth conditions ( YPD agar ) . Eleven mutants grew in a manner indistinguishable from that of wild type strains , three ( isp2Δ , rsc9Δ and top1Δ ) showed slightly slower growth , and only one ( isp1Δ ) showed a substantial growth defect ( Fig 3 ) . Growth at 25 and 37°C generally followed the same pattern ( S3 Fig ) . For all of the mutant strains , growth rates relative to wild type strains in YPD liquid culture at 30°C were also similar to growth on solid agar ( S4 Fig ) . Given the generally robust growth of the deletion strains on solid agar at multiple temperatures , we were confident that the strains could be assessed accurately in further phenotypic analyses and reveal phenotypes independent of basic vegetative growth . To assess the ability of mutants to produce spores , we assessed the behavior of viable deletion mutants in crosses ( a mutant × α mutant ) . In C . neoformans mating between a and α cells initiates a form of sexual development that is robust under laboratory conditions ( V8 agar at 25°C for 5–7 days ) and results in filamentation , fruiting body ( basidium ) development , and spore formation . We evaluated crosses between a and α strains of each mutant for defects in development delineated in details below . To summarize , we discovered that eight of the fifteen mutants displayed defects in sexual development . One mutant showed severe defects in cell fusion ( rsc9Δ ) , two were defective in filamentation ( isp1Δ and bch1Δ ) , three showed severe defects in spore formation ( ddi1Δ , dst1Δ , and top1Δ ) , and two yielded fewer spores than wild type strains ( emc3Δ and gre2Δ ) . One mutant ( isp2Δ ) showed more robust filamentation and increased spore biogenesis ( Fig 4 and Table 3 ) . The remaining mutants ( sfh5Δ , isp3Δ , isp4Δ , isp5Δ , isp6Δ , and isp7Δ ) did not show any discernible phenotypes in sexual development . For each mutant sexual development was evaluated microscopically to assess the formation of developmental structures . Crosses between mutant strains were initiated and evaluated after 24 hours for the presence of classic fusant structures and filaments . One mutant , rsc9Δ , showed a severe defect in the ability to undergo cell fusion and no fusants were observed ( Fig 5A ) . This inability to initiate sexual development resulted in an absence of subsequent filamentation ( Fig 5B ) . Two strains ( isp1Δ and bch1Δ ) formed fusants at frequencies indistinguishable from wild type strains ( Fig 5A ) , but both showed severe defects in filament formation along with rsc9Δ ( Fig 5B and S5 Fig ) . The remaining six mutant strains ( top1Δ , dst1Δ , ddi1Δ , emc3Δ , gre202Δ , and isp2Δ ) all formed robust filaments and were assessed microscopically for basidium and spore formation . All of the strains formed basidia in a manner indistinguishable from wild type . However , spore formation was severely affected in three mutants; crosses between ddi1Δ , dst1Δ , and top1Δ strains formed basidia , but only few or no visible spores , consistent with significant reductions in spore yields when purified using density gradient centrifugation ( Fig 5C ) [16] . Higher resolution microscopy of the filaments of these three mutants revealed apparent defects in nuclear migration ( S6 Fig ) In contrast , emc3Δ , gre202Δ , and isp2Δ crosses all formed robust spores . To quantitate the relative number of spores from each , we carried out spore isolations via gradient centrifugation and quantitated the number of spores formed per cross . We discovered that two strains , emc3Δ and gre202Δ , consistently yielded 2–4 fold fewer spores than wild type crosses , whereas the third strain , isp2Δ , yielded ~50% more spores than wild type ( Fig 5D ) . To evaluate the roles of identified spore proteins in conferring spore-specific properties , we purified spores from crosses and subjected them to a series of assessments of their basic morphological properties , ability to survive stressful environmental conditions , and ability to remain dormant under non-germinating conditions . Spores were recovered from crosses of all of the mutants that produced sufficient numbers of spores for analysis ( i . e . all mutant strains except rsc9Δ , bch1Δ , isp1Δ , top1Δ , dst1Δ , and ddi1Δ ) . In all cases spores from the mutant strains were indistinguishable from wild type spores with respect to basic morphology and surface composition as determined by both light and fluorescence microscopy ( S7A Fig ) . Spores were also evaluated for their ability to survive high temperature ( 50°C ) , oxidative stress ( 20 mM H2O2 ) , and longer-term viability ( growth after 8 weeks at 4°C in PBS ) . Spores from the mutant strains exhibited survival properties identical to those of wild type spores under these conditions ( S7B Fig ) . Finally , we assessed spores for overall stability ( ability to remain dormant ) by incubating them in PBS at 30°C for 120 hours and recording any changes in morphology and germination ability over that time . Again , no differences were detected between mutant and wild type spores ( S7B Fig ) . To identify roles that spore-enriched proteins might play in the spore-specific process of germination , mutant spores were grown under a variety of conditions and evaluated for the ability to germinate and form a colony . Spores were grown on YPD at 25°C and 30°C , under nutrient limiting conditions ( YP , SD , filament agar ) , and on YPD in the presence of cell wall stressors ( Congo Red , caffeine , or SDS ) or an osmotic stressor ( NaCl ) . All of the mutant spores grew into colonies under all conditions tested at a frequency indistinguishable from wild type spores ( S7C Fig ) . However , germinated spores of one mutant ( isp2Δ ) consistently produced smaller colonies relative to wild type spores ( Fig 6A , top ) . Because yeast of the isp2Δ strain grew at a slower rate than wild type yeast on agar plates ( Figs 3 and 6A , bottom ) , we surmised that the differences observed between isp2Δ and wild type spores during germination could be a consequence of differences in yeast growth rates subsequent to germination . In wild type strains , colonies from germinating spores are smaller than those from yeast due to the delay caused by germination itself ( 12h ) . To account for the time required for germination and identify differences in colony size specific to germination in the mutant strain , we adjusted the growth times for yeast and spores ( 51h vs . 63h ) to normalize the yeast and spore colony sizes for the wild type strain ( Fig 6A , WT yeast growth vs . WT spore germination ) . We then quantitated the difference in colony size between yeast and spores of the isp2Δ strain and identified a significant difference in size between yeast- and spore-derived colonies from the isp2Δ strain under these conditions ( ~50% decrease in colony size ) . These data suggested that the isp2Δ strain harbored an additional growth phenotype associated specifically with germination ( Fig 6B ) . To confirm this observation we also carried out germination assays in liquid culture . We had observed previously that the modest slow-growth phenotype of isp2Δ yeast was limited to growth on agar plates and did not occur in liquid culture ( S4 Fig ) . Thus , we carried out quantitative germination assays in liquid culture to evaluate the isp2Δ spores under conditions that would suppress any differences in yeast growth . Spores and yeast from stationary growth phase cultures were seeded into YPD liquid medium and grown at room temperature for 50h , and OD600 was measured every 3min . We observed a clear delay in the growth of isp2Δ spores relative to wild type spores , whereas there was no significant difference in growth between isp2Δ and wild type yeast ( Fig 6C and 6D ) . Quantified doubling times were nearly identical for wild type and isp2Δ yeast ( 7 . 3±0 . 5h vs . 7 . 6±0 . 9h ) ; however , it took 13 . 8±0 . 2h for wild type spore germination cultures to double the initial cell concentration and 15 . 7±0 . 6h for isp2Δ mutants , resulting in an ~2h delay for the mutant in achieving log phase growth , suggesting that ISP2 plays a role in spore germination . To exclude the possibility that deletion of ISP2 resulted in changes in expression of nearby genes , leading to the observed phenotype , we assessed levels of gene expression of ISP2 and its neighbors in both wild-type and mutant strains via qRT-PCR . As expected , levels of ISP2 transcript were easily detected in the wild type strain and undetectable in the isp2Δ strain . In contrast , transcript levels of the genes upstream and downstream of ISP2 in the isp2Δ strain were indistinguishable from the wild type strain , indicating that disruption of ISP2 did not affect neighboring loci ( S8 Fig ) . From these data we conclude that ISP2 is responsible for the germination delay and plays a specific role in spore germination . The apparent difference in germination rate between wild type and isp2Δ spores suggested several possibilities: a delay in the initiation of germination , a slower rate of germination ( i . e . a decrease in rate of differentiation from a spore into a yeast ) , or a delay in entering regular vegetative growth . To differentiate among these possibilities , we microscopically evaluated individual spores of both wild type and isp2Δ strains every four hours from 0h to 16h after the initiation of germination . We observed that the isp2Δ spores were identical to wild type spores in their abilities to initiate , sustain , and complete the process of germination ( 110 and 122 spores from wild type and isp2Δ strains were evaluated , respectively ) . By 12h the morphological transition from spore to yeast was complete and indistinguishable between wild type and isp2Δ spores ( Fig 6E ) . Based on these observations , it appears that the delay leading to smaller colonies during germination occurs after the morphological transition to yeast . These findings indicate that the isp2Δ mutant spores were delayed between the end of the morphological transition from spore to yeast and the beginning of active growth . This is consistent with the liquid germination assays in which there was a two-hour delay in isp2Δ mutant entry into log phase growth . As such , we can pinpoint a delay in growth of the isp2Δ mutant spores at the transition from a fully formed yeast to a vegetatively reproducing yeast , confirming a role for Isp2 in the spore-specific process of re-initiating vegetative growth in response to nutrients during germination . To determine the source of Isp2 activity during germination ( i . e . basidium-derived vs . spore-derived ) , we also carried out liquid germination assays with spores purified from crosses between wild type and isp2Δ strains ( WT × isp2Δ ) . This population of spores contains a 50:50 ratio of wild type to isp2Δ genotypes . We predicted that if the Isp2 protein necessary for wild type germination were produced by spores , then half of the spores from a wild type by isp2Δ cross would harbor the germination delay phenotype ( causing a shift in the liquid germination assay curve that would fall between those of WT × WT and isp2Δ × isp2Δ spores ) . In contrast , if the Isp2 necessary for normal germination were basidium-derived ( deposited in spores during spore biogenesis ) , then the presence of Isp2 in the mutant spores would lead to a germination curve identical to wild type spores . We observed that spores from crosses between wild type and isp2Δ strains germinated at a rate indistinguishable from wild type spores , indicating that Isp2 function is not impaired in this mixed-genotype population ( Fig 7A ) . These findings suggest that the Isp2 protein detected in spores is produced in the basidium and deposited into spores during spore formation , resulting in subsequent wild type germination in the absence of the ISP2 gene ( Fig 7B ) .
Proteomic analyses of spores have been used in an array of fungal species , and studies have identified proteins associated primarily with the biological processes of protein synthesis , protein folding and degradation , and metabolism and energy production [19 , 33] . In C . neoformans spores , we detected proteins involved in these same categories , but we found no significant differences in enrichment of these categories between yeast and spores . Instead , our proteomic comparisons between spores and yeast indicated that spore proteins are enriched in distinct processes , including regulation of transcription , DNA metabolic processes , chromosome organization , and mRNA metabolic processes . Comparing spores with vegetative growth forms provides the advantage of being able to ascribe spore-specific features , properties , or functions to any given class of proteins and improve resolution of spore-relevant pathways . For example , analyses of Aspergillus fumigatus conidia ( asexual spores ) and mycelia ( vegetative form ) revealed that conidia disproportionately harbor proteins associated with reactive oxygen intermediates ( ROI ) detoxification , pigment ( melanin ) biosynthesis , and conidial rodlet layer formation [34] . In contrast , in the related species Aspergillus nidulans , proteins related to ROI detoxification and some heat shock proteins were more abundant in mycelia than in conidiospores [35] . In C . neoformans , there were no differences between spores and yeast in any of these categories , suggesting that there are large differences in survival strategies across fungal species . Perhaps the different conditions leading to spore formation contribute to distinct proteomes among species or between asexual and sexual spores . Comprehensive proteomic studies to link the well-characterized morphological changes , underlying molecular events , and transcriptional networks that control sporulation in many fungi will be extremely useful in providing a global understanding of how the spore proteome is synthesized through the corresponding developmental process . In C . neoformans sexual development between cells of opposite mating types involves five distinct morphological events: mate detection , cell fusion , filamentation , basidium formation , and sporulation . We predicted that proteins overrepresented in spores would be likely to participate in subsequent spore processes such as dormancy , stress resistance , and germination . Instead , we discovered that nearly half of the mutants in spore-enriched proteins showed phenotypes in development prior to the formation of spores . None has been associated previously with fungal development or spore formation , and these proteins fell into highly diverse biological and functional categories . For example , we discovered that Rsc9 is required for haploid yeast to fuse with one another to initiate sexual development . In S . cerevisiae and Schizosaccharomyces pombe Rsc9 is one component of the RSC chromatin remodeling complex , and it is required for viability in both [36 , 37] . In S . cerevisiae RSC9 is involved in genome-wide transcriptional response to a stress-induced signaling cascade [38] , and in S . pombe RSC9 is down-regulated when cells are subjected to nitrogen limitation [39] . Although in C . neoformans RSC9 is not required for vegetative viability , the slower growth of rsc9Δ mutants indicates its involvement in vegetative growth [40] . One possibility is that the dramatic defect observed during sexual development is due to a role for RSC9 in a regulatory response to nutritional stresses encountered during development . Of the two genes we found required for early filamentous growth , ISP1 and BCH1 , the latter likely participates in active intracellular trafficking during filamentation . Bch1 is a member of the ancient family of Chs5p-Arf1p-binding Proteins ( ChAPs ) that are conserved in fungi and required for export of specialized cargo from the Golgi [41] . In S . cerevisiae one important cargo , chitin synthase III ( Chs3 ) , is required for mating projection formation and chitosan production in ascospore walls . More importantly , filamentous fungal growth requires the endocytic system that transports secretive vesicles and early endosomes , through the cytoskeleton in a variety of fungi [42] . Thus , it is possible that Bch1 is required for the transport of specific cargos necessary for filamentation during sexual development of C . neoformans , leading to abnormal vacuole distribution and/or accumulation and disruptions in nuclear movement in bch1Δ strains ( S5A Fig ) . The other filamentation gene , ISP1 , contains several conserved NAD ( P ) binding sites and is annotated as a putative short chain dehydrogenase , but no significant homologs have been identified in other organisms . In S . cerevisiae , mitochondrial/metabolic functions are required for the transition to a filamentous form under certain conditions of nutrient stress [43] . One possibility in C . neoformans is that Isp1 contributes to overall energy production through a development-specific redox reaction during filamentation that is important to drive filament formation and proper nuclear migration ( S5A Fig ) . Overall , the discovery of Rsc9 , Bch1 , and Isp1 as important players in early sexual development was surprising and suggests that proteins with high representation in spores could be "left over" from previous steps in development and passively carried into newly produced spores ( but not necessarily function in spore processes ) . Alternatively , these proteins could serve multiple roles in development and reflect a highly efficient , intercalated process in which proteins serve critical roles at multiple stages . In this case , sexual development proteins could function not only during fusion and filamentation but also later during spore biogenesis and be actively deposited into spores for spore-specific processes . For example , it is known that histone modifications and packaging of chromatin take place during spore formation in other organisms [3]; thus , it is plausible that Rsc9 plays roles in both signaling during mating and chromatin condensation during spore biogenesis in C . neoformans . In response to unknown signals , filamentation during sexual development ends , and a basidium forms on the terminal filament . None of the mutants in genes encoding spore-enriched proteins showed phenotypes specific to basidium formation . However , the majority of mutant strains with detectable phenotypes ( 5 in total ) showed defects during spore biogenesis . Three genes TOP1 , DST1 , and DDI1 were essential for spore formation , and mutants formed basidia that were largely devoid of spores and appeared "bald . " This phenotype has been observed previously for only two other genes in C . neoformans , SPO11 and UBC5 , which are conserved meiosis genes in fungi [44] . The absence of spores in spo11Δ and ubc5Δ strains suggests that meiosis and spore formation are tightly coupled ( as has been observed in other fungi ) . This would be consistent with our finding that mutants in TOP1 , a critical topoisomerase , do not form spores likely as a result of improper decatenation of meiotic products . Ddi1 in S . cerevisiae is a DNA damage-inducible SNARE binding protein that has been found to be involved in cell-cycle control and repression of protein secretion in vegetative cells of S . cerevisiae ( but does not play a significant role in sporulation ) [45 , 46] . Ddi1 in C . neoformans shares all of its conserved domains at high similarity with its S . cerevisiae homolog , but it is clearly required for spore formation . One possibility is that the formation of hundreds of spores per basidium in C . neoformans requires Ddi1 partitioning of limited vesicle resources in a manner not required for the four ascospores of S . cerevisiae . DST1 encodes the conserved general transcription elongation factor TFIIS , and is required for the balanced expression of genes encoding ribosomal components under transcriptional stress [47] . Deletion of DST1 in S . cerevisiae leads to a slight defect in sporulation [46]; however , in C . neoformans DST1 is indispensable for efficient spore production , perhaps reflecting higher demands for ribosomal components needed by the basidium to produce spore components . Deletion of genes encoding two spore-enriched proteins , Emc3 and Gre202 , led to no visible defect in sexual development or spore formation . However , gradient centrifugation purification of spores from emc3Δ and gre202Δ crosses consistently and repeatedly revealed decreases in spore yields of 2–4 fold ( Fig 5D ) . In other fungi Emc3 is a member of a conserved endoplasmic reticulum ( ER ) membrane protein complex , which contributes to efficient protein folding in ER and whose deletion induces the unfolded protein response ( UPR ) [48] . Gre202 is an NADPH-dependent methylglyoxal reductase , potentially capable of reducing a very wide spectrum of compounds [49] . It remains to be determined how these proteins contribute to spore yields and other spore-related processes . We designated the proteins found only in spores and not in yeast in our datasets as "spore-enriched" because they could be either truly specific to spores or simply highly overrepresented in spores relative to yeast . This possibility was borne out by the discoveries of Irr1 , Prp11 , and Prp31 in the spore proteome data ( and not in the yeast data ) . All three of these proteins are essential for viability in other organisms , and our failed attempts to knock out the genes encoding them in C . neoformans were consistent with these proteins playing essential roles in C . neoformans as well . Why these proteins are overrepresented in spores relative to yeast is not entirely clear; however , in S . cerevisiae , Irr1 is a subunit of the essential cohesin complex , which is needed for sister chromatid cohesion during mitosis and meiosis [50] , and Prp11 and Prp31 are necessary for pre-mRNA processing , which can contribute to meiosis-specific splicing of certain messages for the control of gene expression during sporulation [51 , 52] . It is plausible that given the demands for meiosis and repeated mitosis in the C . neoformans basidium prior to spore biogenesis , these proteins are at high levels in the basidium and are carried into developing spores . Alternatively , Prp11 and Prp31 may be deposited in spores to facilitate splicing of transcripts necessary for germination . One of the 18 mutant strains displayed a spore-specific phenotype; the isp2Δ shows a delay in spore germination at the step in which vegetative growth is initiated . The protein sequence of Isp2 is 374 amino acids , and shows no similarity to any other known or predicted proteins or protein domains in any organism ( with the exception of another weakly related protein in C . neoformans ( CNM00430 ) ) . As such , no clear structures or functions can be gleaned from sequence information; however , the Protein Homology/analogY Recognition Engine V 2 . 0 ( Phyre2 ) algorithm [53] suggests the possibility of a trihelix DNA binding . Therefore , Isp2 might function as a transcription factor during germination , but this remains to be determined . Discerning the function of Isp2 in the process of germination will be important in future studies in part because of the paucity of information available regarding fungal germination . Germination of spores is a critical process required for propagation of fungal species , and among fungal pathogens , germination is essential for causing disease . However , there are long-standing questions about how fungal spores germinate . Relative to other cellular processes , we know very little about the molecular mechanisms governing spore germination , and it is not yet clear whether germination is a specific process with specific machinery or if it is a somewhat modified exit from a form of stationary phase growth into log phase growth . Few spore proteins have been identified to be involved specifically in spore germination in the fungal kingdom . In S . cerevisiae , members of the ras/mitogen-activated protein kinase pathway and the transcription factor Ume6 have been found to be important for spore germination , but they also play important roles during vegetative growth and meiosis during sporulation [6 , 54] . In addition , large-scale genetic screens in S . cerevisiae for mutants with defects in spore germination have not identified germination-specific components [46 , 55] . In contrast , our data for Isp2 suggest very strongly that in C . neoformans germination is distinct from vegetative growth and likely requires at least some germination-specific machinery . The activity of Isp2 also points to a specific germination program because Isp2 does not function in a spore-autonomous manner . Rather , akin to the "maternally deposited" transcripts and proteins of metazoans during embryogenesis , Isp2 appears to be packaged into spores during spore biogenesis . It is this "maternal" protein that governs an efficient transition to vegetative growth during late germination; spores that do not harbor an intact ISP2 gene behave like wild type spores during germination as long as they were derived from crosses between wild type and mutant strains ( WT × isp2Δ ) . As such , Isp2 protein and/or transcript must be produced prior to spore formation and packaged into spores . Based on these findings , we propose that Isp2 protein is produced in basidia prior to and/or concurrent with the biogenesis of spores and is deposited specifically into the spores . This is consistent with the overrepresentation of Isp2 protein in spores relative to yeast in our proteomic data . We posit that Isp2 then contributes to efficient spore germination at a specific , late stage of differentiation—after the morphological transition of the small , ovoid spore into a larger , round yeast and before active replication during vegetative growth ( growth initiation phase ) ( Fig 7B ) . This is consistent with our direct observations of changes in spore morphology and quantitative assessments of spore germination rates and yeast growth . With the discovery of Isp2 , we anticipate that additional germination-specific machinery exists and can be identified using similar proteomic approaches . High sensitivity mass spectrometry ( nanoLC-MS/MS ) in combination with molecular genetics and quantitative phenotype assays is a powerful tool for assessing protein functions in rare cell types . Ultimately , stage-specific proteomics during germination will enable a high-resolution view of the germination process . The identification and characterization of a germination-specific , developmentally-deposited , spore-resident protein is an unprecedented finding in fungal spores with the potential to provide opportunities for novel pathway discovery . Isp2 and other germination-specific proteins could be promising targets for the development of inhibitors to be used in the prevention of fungal growth and spore-mediated disease .
All strains used were of the serotype D background ( Cryptococcus neoformans var . neoformans strains JEC20 and JEC21 ) , and their genotypes are listed in S5 Table [56 , 57] . All were handled using standard techniques and media as described previously [58 , 59] . Spores were purified from crosses after 6 days on V8 agar using density gradient purification as described previously [16] . Yeast were grown on V8 agar plates for 6 days prior to harvest . Spores or yeast were suspended in lysis buffer ( 50mM Tris-HCl pH 7 . 5 , 1% SDS , 10mM EDTA ) and sonicated 5 times for 12 seconds each at a power output of 2 and 100% duty cycle on ice using Branson Sonifier 250 ( Emerson Industrial Automation , USA ) . Resulting lysates were extracted with 1:1 phenol-chloroform , precipitated with ethanol , and washed with ethanol and acetone . The pellet was air-dried and dissolved in SDS-containing sample buffer [60] . Proteins were separated in a 12% PAGE Bis-Tris gel and visualized using Coomassie Brilliant Blue-staining [61] . Proteins were recovered and prepared for MS as follows: Each protein-containing PAGE gel lane was excised cut into 5 bands and minced . Gel samples were washed with 100mM NH4HCO3 , incubated at ambient temperature , and then dried . The samples were reduced by resuspension in 100mM NH4HCO3 with 10mM dithiothreitol and incubation at 56°C for 1h . Samples were then treated with 100mM NH4HCO3 with 55mM iodoacetamide , incubated for 45 minutes in the dark at ambient temperature , washed twice with 100mM NH4HCO3 , incubated in acetonitrile for 5 minutes , and dried . Samples were digested with 12 . 5ng/μL trypsin in 100mM NH4HCO3 for 30min . Peptides were extracted by overnight incubation in 100mM NH4HCO3 , followed by washing with 50mM NH4HCO3 , 50% acetonitrile and 5% formic acid , and 100% acetonitrile . Extracted peptides were dried and resuspended in 15μL 0 . 2% formic acid . Peptides were loaded onto a 75μm inner diameter column packed with 5μm Magic C18 particles ( Michrom ) and eluted with increasing acetonitrile over 60min . Eluted peptides were ionized by electrospray and analyzed by an LTQ Orbitrap Velos , Velos pro , or an LTQ mass spectrometer . For the MS1 survey scan , 1×106 ions were analyzed by the Orbitrap or 4×104 ions were analyzed by the ion trap . From this scan the 10 most intense features were selected for MS/MS analysis with a 30-60s dynamic exclusion . Fragmentation data was produced by either HCD with analysis in the Orbitrap with 2×105 ion target value , or with CAD with analysis in the ion-trap with 1×104 target value . Peptide analysis was performed using the COMPASS software suite . Spectra were converted into text files and searched against the Cryptococcus neoformans proteomic database from NCBI , using the OMSSA search algorithm . Precursor average mass was searched with a tolerance set to 3 . 5Da , monoisotopic fragment ions were searched with tolerance set to 0 . 5Da or 0 . 01Da for ion-trap and orbitrap spectra , respectively . Peptides were filtered to 1% FDR based on E-value alone or E-value and ppm mass error for Ion-trap and Orbitrap data , respectively . Peptides were grouped into proteins and filtered to 1% FDR using established rules . The mass spectrometry data from this publication have been submitted to the Chorus project database ( https://chorusproject . org/pages ) and assigned the project ID 751 . Molecular weights and isoelectric points of proteins were predicted using Compute pI/Mw Tool from the ExPASy server [62] . Transmembrane helices in proteins were predicted using TMHMM version 2 . 0 [63] . Cellular localizations were predicted using WoLF PSORT [64] and molecular functions by Blast2GO [65] . Functional annotation clustering analyses were performed using the Database for Annotation , Visualization and Integrated Discovery ( DAVID ) [23 , 24] and all proteins that were identified at least once in yeast or spores ( Table 1 and S1 Table ) were used as the background pool for comparison . Peptide spectral match ( PSM ) numbers associated with protein identification were used to estimate relative protein abundance . Data were normalized to the global average number of peptides detected per protein per experiment excluding the top and bottom fifth percentile data [66] . PSMs in each experiment were scaled to produce a global average across all experiments . To estimate the relative abundance of each spore protein compared to yeast , the PSM numbers for each protein from each replicate were summed and a ratio ( r ) of spore PSM to yeast PSM for each protein was determined . Deletion constructs were created using fusion PCR [67] and the primers listed in S5 Table . For each gene the 5'- region was amplified with primers P1 and P2 , the 3'-region was amplified with P5 and P6 , and the cassette of selection marker was amplified with P3 and P4 . PCR fusion using P1 and P6 was used to create the final full-length deletion cassette . The NATR and NEOR deletion cassettes were transformed into JEC20 and JEC21 by biolistic transformation , grown on rich medium containing 1M sorbitol , and selected on medium containing 200μg/mL G418 or 200μg/mL nourseothricin [68] . The URA5 deletion cassettes were transformed into JEC34 and JEC43 also by biolistic transformation , grown on minimum medium containing 1M sorbitol for selection . Resulting colonies were screened using PCR for correct insertion of the knockout construct ( P7 and P8 ) and absence of target ORF ( P9 and P10 ) , and further assessed via Southern blotting for single integration of the construct to identify multiple independent knockouts in both mating types [61] . At least three independent a and α strains were recovered for each deleted gene . All yeast growth phenotype tests were carried out with at least four independent strains per gene . All sexual development and spore phenotype tests were carried out in at least three pairs of crosses between independent a and α strains per gene . All phenotypes were observed in all independent deletion strains tested for each gene of interest . Recombinant constructs were created using fusion PCR [67] and primers were listed in S5 Table . For each gene , the ORF region immediately upstream of the stop codon was amplified with primers P1 and P2 , the downstream region was amplified with P5 and P6 , and the cassette of mCherry sequence and selection marker was amplified with P3 and P4 . PCR fusion using P1 and P6 was used to create the final full-length transformation cassette . The mCherry-URA5 replacement cassettes were transformed into JEC34 and JEC43 also by biolistic transformation , grown on minimum medium containing 1M sorbitol for selection . Resulting colonies were screened using PCR for correct insertion of the construct and further assessed via Southern blotting for single integration of the construct to identify multiple independent knockouts in both mating types [68] . At least three independent a and α strains were recovered for each gene . Spores were purified from at least three pairs of crosses between independent a and α strains per gene . Consistent expressions of recombinant mCherry proteins were observed in all independent strains tested for each gene of interest . Deletion strains were grown to stationary phase in liquid YPD culture overnight at 30°C with shaking and then used as seed cultures to assess their growth phenotypes in fresh liquid YPD culture during log phase or on solid YPD plate as described previously [69] . Briefly , the seed cultures were diluted to an OD600 of 0 . 05 in fresh liquid YPD medium and their growth at 30°C with shaking was monitored by taking samples for absorbance measurement at a wavelength of 600nm every two hours in a time course of around twenty-four hours . The growth curves were then plotted and doubling times were calculated based on their growth rates during log phase . The same diluted cultures were also spotted on YPD plates at 10-fold serial dilutions and grown for 3–4 days at room temperature , 30°C , or 37°C before being visually examined and photographed . Spores from each mutant were stained with 0 . 05mM calcofluor white MR2 ( Sigma ) or 1:50 diluted Concanavalin A ( ConA , Vector labs ) , visualized microscopically , and photographed . Spore germination was assessed on rich solid YPD medium by spotting 4μl 10-fold serial dilutions from a starting concentration of 103/μl and incubating for 3–4 days at room temperature , 30°C , or 37°C before being photographed using a SCION CFW-1309M Grayscale Digital Camera . Germination was also tested under a variety of stressful conditions at 30°C , including nutrient-limiting media ( YP no dextrose , Synthetic medium no dextrose , and filament agar ) and YPD medium with cell wall stressors ( 1mg/ml Congo Red , 0 . 1mg/ml caffeine , and 0 . 005% SDS ) or osmotic stressors ( 1M NaCl ) . To test their heat stress resistance , spores were incubated at 50°C for 5 or 10 minutes before spotting on YPD plate and germinating at 30°C as described above . Oxidative stress resistance was tested by incubating in 20mM hydrogen peroxide for 5 minutes prior to washing and germination [16] . Wild type , isp2Δ , and complemented crosses were initiated on V8 plates as described earlier and incubated for 48h or 72h before RNA extraction . Then RNA was extracted using a hot phenol method [61] and cleaned up using RNeasy Mini Kit ( Qiagen ) . First-strand cDNA synthesis was carried out using Superscript III reverse transcriptase with 5μg total RNA and oligo ( dT ) 12-18 according to manufacture’s manual ( Invitrogen ) . Quantitative realtime PCR ( qRT-PCR ) was performed using the Bio-Rad CFX96 real-time system with a C1000 thermal cycler ( Bio-Rad ) . Each PCR reaction was in triplicate and used 5ul diluted cDNAs , SYBR Green ( Sigma ) , and oligo pairs listed in S5 Table . The expression level of each gene is normalized to the internal reference gene GPD1 and relative to wild type . All values were generated by Bio-Rad CFX manager software v . 2 . 0 . | Spores are a critical cell type that allow long-term survival of diverse organisms from bacteria to fungi to plants . Among fungi , spores are often formed when growth conditions are poor; spores can then disperse to more favorable environments and reinitiate growth . Spores of some environmental fungi can cause fatal disease in humans . Here we used the meningitis-causing yeast Cryptococcus neoformans to determine the roles of spore-enriched proteins in spore biology . Using a combined proteomics-genetics approach , we identified eighteen spore-enriched proteins , knocked out the genes encoding each of them , and assessed the resulting strains for phenotypes in a broad array of assays . We predicted that mutant strains would be likely to show defects in spore-specific processes , but instead , we discovered that the majority harbored defects in sexual development , the process by which spores are formed . Only one mutant exhibited a defect in a spore-specific process ( germination ) . Our data reveal that many spore-represented proteins are associated with pre-spore developmental processes , rather than intrinsic spore-specific properties or processes . These findings indicate a previously unknown molecular connection between the developmental process that results in spore biogenesis and the composition of infectious spores . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Protein Composition of Infectious Spores Reveals Novel Sexual Development and Germination Factors in Cryptococcus |
Aedes albopictus is a vector of increasing public health concern due to its rapid global range expansion and ability to transmit Dengue virus , Chikungunya virus and a wide range of additional arboviruses . Traditional vector control strategies have been largely ineffective against Ae . albopictus and novel approaches are urgently needed . Photoperiodic diapause is a crucial ecological adaptation in a wide range of temperate insects . Therefore , targeting the molecular regulation of photoperiodic diapause or diapause-associated physiological processes could provide the basis of novel approaches to vector control . We investigated the global transcriptional profiles of diapause induction in Ae . albopictus by performing paired-end RNA-Seq of biologically replicated libraries . We sequenced RNA from whole bodies of adult females reared under diapause-inducing and non-diapause-inducing photoperiods either with or without a blood meal . We constructed a comprehensive transcriptome assembly that incorporated previous assemblies and represents over 14 , 000 annotated dipteran gene models . Mapping of sequence reads to the transcriptome identified differential expression of 2 , 251 genes in response to diapause-inducing short-day photoperiods . In non-blood-fed females , potential regulatory elements of diapause induction were transcriptionally up-regulated , including two of the canonical circadian clock genes , timeless and cryptochrome 1 . In blood-fed females , genes in metabolic pathways related to energy production and offspring provisioning were differentially expressed under diapause-inducing conditions , including the oxidative phosphorylation pathway and lipid metabolism genes . This study is the first to utilize powerful RNA-Seq technologies to elucidate the transcriptional basis of diapause induction in any insect . We identified candidate genes and pathways regulating diapause induction , including a conserved set of genes that are differentially expressed as part of the diapause program in a diverse group of insects . These genes provide candidates whose diapause-associated function can be further interrogated using functional genomics approaches in Ae . albopictus and other insects .
Dengue virus ( DENV ) and Chikungunya virus ( CHIKV ) are ( re ) emerging arboviruses transmitted primarily by the mosquitoes Aedes aegypti and Aedes albopictus . Estimates of annual DENV infections range from 50–390 million [1 , 2] . Although levels of CHIKV infection are much lower , a 2005–2006 CHIKV outbreak on La Réunion island in the Indian Ocean infected over 266 , 000 individuals [3] , and was followed by an outbreak in northern Italy in 2007 [4] . This latter event generated considerable concern because it represents the first temperate outbreak of CHIKV , which had previously been restricted to tropical areas . CHIKV has recently spread to the Caribbean islands and South America [5 , 6] , and local transmission of both DENV and CHIKV has recently occurred in peninsular Florida [7 , 8] and Europe [9] . Although Ae . aegypti has historically been considered the primary vector of both DENV and CHIKV , Ae . albopictus has been implicated as the primary vector in at least five DENV outbreaks between 2001–2010 , including one in the temperate location of Croatia [9 , 10] , and Ae . albopictus was also the primary vector of the CHIKV outbreak on La Réunion , which coincided with a CHIKV mutation from Alanine to Valine at position 226 of the E1 viral envelope protein which confers increased transmission efficiency of CHIKV by Ae . albopictus [11 , 12] . Ae . albopictus is an aggressive human biting mosquito that has spread from its native Southeast Asian range to all continents except Antarctica over the last 30 years [13] . This mosquito is a competent vector of 22 other arboviruses in addition to DENV and CHIKV [14] . Because vaccines and drug treatments are not available for DENV , CHIKV and most other arboviruses , vector control has been the most effective strategy for controlling these diseases . However , traditional vector control approaches such as insecticides and source reduction have been largely ineffective against Ae . albopictus [15] , in part because this ecological generalist occupies such a wide range of container types as larval habitats [15 , 16] . Thus , novel approaches to suppressing this vector are urgently needed . Photoperiodic diapause is a pre-programmed developmental arrest in response to the token environmental stimulus of photoperiod . Photoperiodic diapause is a crucial adaptation to survival during the unfavorable conditions of winter in a wide variety of temperate insects , including many important vectors of human disease [17–19] . Because diapause is essential for overwinter survival in temperate habitats , identifying novel targets for genetic or chemical disruption of diapause or diapause-associated physiological processes could provide new tools to augment traditional vector control approaches . Genetic approaches to vector control are becoming increasingly feasible [20–22] , and the diapause response represents an attractive target for genetic control strategies because diapause-disrupting genetic constructs could be effectively spread by released males during early-spring and mid-summer when there is no requirement for diapause , but then would have a lethal effect when winter arrives . Also , the diapause response involves the modulation of a wide variety of fundamental physiological processes . Therefore , determining the molecular regulation of these processes can provide additional targets for novel vector control strategies . Kostal [23] defines five eco-physiological phases of the diapause program . First , the diapause induction phase is characterized by perception of the environmental token stimuli well in advance of the adverse seasonal conditions . Diapause induction is followed by a preparation stage when direct development continues but certain physiological processes occur to help the organism prepare for the actual diapause ( arrest ) stage . Next , during diapause initiation , direct development ceases and metabolic rates are reduced . During the actual diapause phase , the state of diapause is maintained , even under conditions favorable for growth and reproduction . After a certain period of time or in response to chilling or other unknown factors [23] , diapause is terminated and direct development can be resumed . Recent studies utilizing RNA sequencing ( RNA-Seq ) in a range of species have begun to elucidate global transcriptional profiles of diapause preparation [24 , 25] , initiation , maintenance [26 , 27] and termination [27 , 28] . These studies have significantly increased understanding of the molecular basis of this crucial ecological adaptation . Results of these studies emphasize that diapause is a dynamic physiological and metabolic process [29] . For example , in the preparation phase organisms need to accumulate extra nutrients to survive the long months of diapause through the winter [29] . Lipid metabolism stands out as a common molecular theme across species and at different stages of diapause , indicating that energy conservation and utilization before , during and after diapause are essential for the organisms’ survival [29] . Despite increasing knowledge of the molecular regulation of diapause preparation , maintenance and termination , the molecular mechanisms regulating the more “upstream” stage of diapause induction remain much less well understood . The molecular mechanisms of diapause induction have been well studied in Bombyx mori [30–32] , and to a lesser extent in the mosquito Culex pipiens [33–35] . These studies confirm the fundamental importance of hormonal regulation during diapause induction [36 , 37] . However , the mechanism by which organisms measure and interpret photoperiod remains completely unresolved and controversial . Some researchers have argued that the circadian clock provides the mechanistic basis of photoperiodic time measurement [38–40] , while others have argued that components of the circadian clock , specifically timeless , might function as a component of an “hourglass” interval timer that can measure photoperiod independent of its role in the circadian clock [41 , 42] . Ae . albopictus is an outstanding model to study the molecular underpinnings of photoperiodic diapause . Ae . albopictus undergoes a well characterized photoperiodic diapause , which can be easily and consistently induced in the laboratory by short-day photoperiods [43–45] . In the diapause response of Ae . albopictus , the photosensitive pupal or adult female perceives the signal of short day length and subsequently produces offspring in which the pharate larva enters diapause inside the chorion of the egg . In Ae . albopictus , diapause eggs are both more resistant to cold temperatures [46] and desiccation [47 , 48] than non-diapause eggs . Diapause eggs are also larger and contain ca . 30% more total lipids than non-diapause eggs [49] . Genomic resources for the study of diapause include previously established extensive gene expression profiles of Ae . albopictus across multiple life stages in the diapause program [24–26 , 50] . In addition , the published genome sequence of Ae . aegypti [51] , a closely related species , provides a valuable genomic resource . In this study , we utilized an RNA-Seq approach to examine global transcriptional profiles of diapause induction in the adult Ae . albopictus females that had either received a blood meal or had not received a blood meal . Previous experiments with Ae . albopictus have demonstrated that the “photoperiodic switch” triggering the transition from producing non-diapause eggs to diapause eggs can occur before a blood meal [52] . Thus we hypothesized that genes in molecular pathways related to photoperiod interpretation and the early stages of hormonal regulation would exhibit differential expression under diapause-inducing vs . non-diapause-inducing photoperiods in females without a blood meal . We also hypothesized that genes in molecular pathways related to nutritional provisioning of diapause eggs would exhibit differential expression in blood-fed females under diapause conditions . We first constructed a comprehensive transcriptome for Ae . albopictus across multiple life stages by combining sequences from the current study of the adult stage with those from previous studies of pre-adult stages . We then analyzed differential gene expression by performing read mapping back to the composite transcriptome . Our analysis of differential gene expression proceeded in the following three steps: A ) we validated key transcriptional responses to a blood meal under both diapause-inducing and non-diapause-inducing conditions , B ) we identified potential regulatory elements during diapause induction by analyzing differential gene expression in females reared under diapause-inducing vs . non-diapause-inducing photoperiods without a blood meal , C ) we identified genes and pathways related to maternal provisioning of diapause eggs by analyzing differential gene expression in females reared under diapause-inducing vs . non-diapause-inducing photoperiods with a blood meal . Our results emphasize that diapause is an adaptive metabolic plasticity that involves dramatic changes in transcriptional activity and reinforces our previous hypothesis that a conserved set of genes has contributed to the evolution of diapause in divergent insect lineages .
A laboratory colony of Ae . albopictus was established from over 200 individuals collected as larvae from more than 10 used tires located at a recycling center in Manassas , Virginia . The colony was maintained for two generations under a non-diapause-inducing long-day ( LD ) photoperiod of 16L:8D at 21°C and approximately 80% relative humidity as described previously [53 , 54] . An overview of the experimental design is presented in Fig 1 . Details of the general workflow for producing tissue and RNA samples under diapause and non-diapause conditions can be found in Poelchau et al . [44] . Briefly , the F3 laboratory generation larvae were reared under LD conditions described above . Upon pupation , females were transferred into replicate adult cages ( = biological replicates ) of approximately 60 females per cage . Eight adult cages were established under a diapause-inducing short-day ( SD ) photoperiod ( 8L:16D ) and eight adult cages were established under a non-diapause-inducing LD photoperiod ( 16L:8D ) ( 16 total cages ) . Additionally , three adult mass-swarm cages containing 30 male and 30 female mosquitoes were established under both SD and LD photoperiods to measure diapause incidence ( six total mass-swarm cages ) . This experiment was designed to examine the effects of diapause-inducing short-day photoperiods on gene expression of females both with and without a blood meal ( Fig 1 ) . Therefore , adult female cages were maintained under SD and LD conditions for 11 days which is sufficient to produce an unambiguous diapause ( SD ) vs . non-diapause ( LD ) signal [52] . After 11 days , within each photoperiod treatment ( SD and LD ) , four replicate cages received a blood meal and four replicate cages did not receive a blood meal . Females were blood fed on a human host to repletion between Zeitgeber times ( ZT ) 3–4h . The Georgetown University Institutional Review Board ( IRB ) determined that mosquito blood feeding did not qualify as human subjects research and thus did not require IRB approval . However , the blood feeding protocol was approved by the Georgetown University Occupational Health and Safety Committee . Shortly after blood feeding , females were CO2 anesthetized and only females with a swollen abdomen and a visible blood meal were retained for the blood meal treatment . The non-blood-fed cages were maintained in parallel to the blood-fed cages . Twenty-six to twenty-eight hours after blood feeding , at ZT 6–8h , female whole bodies from both blood-feeding treatments under both LD and SD photoperiods were snap-frozen in liquid nitrogen and stored at -80°C for RNA extraction . Twenty-six to twenty-eight hours after a blood meal at 21°C is expected to correspond to slightly before the peak of the transcriptional response to a blood meal [55] , when the proteolytic activity in the midgut is still increasing [56] . Adult whole bodies were chosen for RNA extraction instead of specific tissues in order to obtain a global transcriptional profile of diapause induction because very little is known about the molecular physiology of diapause induction in adult females of species that undergo embryonic diapause ( but see [30 , 31] ) , and literally nothing has been published on this topic in mosquitoes [19] . Prior to RNA extraction , the blood bolus was dissected out from the female body in RNAlater ( Sigma Aldrich , St . Louis , Missouri ) , and the female bodies were stored in RNAlater at 4°C for approx . 24 hours . Next , the female whole bodies were ground in TRI Reagent ( Sigma Aldrich , St . Louis , Missouri ) and RNA was extracted according to manufacturer’s instructions . Residual DNA in the RNA samples was removed by Turbo-DNA free ( Ambion , Austin , Texas ) . Integrity assessment of total RNA was performed by fluorometry on an RNA chip ( Bioanalyzer 2100 , Agilent Technologies , Santa Clara , California ) . Four biological replicates from each of four experimental treatments ( i . e . , SD blood fed and non-blood fed , LD blood fed and non-blood fed ) were submitted for sequencing , resulting in 16 sequenced RNA libraries ( see Sequencing below ) . Diapause incidence measurements were performed following methods described in [57] . Briefly , females from the three mass-swarm cages maintained under SD and the three mass-swarm cages under LD were allowed to blood feed to repletion on a human host two to three days after eclosion . Females were blood fed a second time , 6 days later , to collect eggs over multiple gonotrophic cycles . Four days after the first blood meal , a small brown jar half-filled with approximately 50mL of deionized water and lined with unbleached paper towels was placed into each cage to stimulate oviposition . Six days after the first blood meal , paper towels with eggs were removed , maintained under SD conditions for 48–72h , and then gently air-dried . Egg collection was performed every two to three days for two weeks . Egg papers were stored at approximately 80% relative humidity under SD for at least seven days before they were exposed to a hatching stimulus . Eggs are not photosensitive and the uniform treatment under SD at the egg stage does not induce diapause in Ae . albopictus [43] . Eggs ranging from one to two weeks of age were stimulated to hatch by submersion in water . The number of hatched larvae was recorded and the egg papers were re-dried . This procedure was repeated 7 days later and then the eggs were bleached [58] to record the number of embryonated but unhatched ( = diapause ) eggs . Diapause incidence ( DI ) was calculated as DI = ( number of embryonated unhatched eggs ) / ( number of hatched eggs + number of embryonated unhatched eggs ) [52 , 57] . Illumina paired-end mRNA-Seq library construction was performed by the Institute for Genome Sciences at the University of Maryland according to the TruSeq RNA sample preparation kit ( Version 2 ) ( Illumina Inc . , San Diego , California ) . The 16 libraries were individually bar-coded [59] according to manufacturer’s instructions and equally split for paired-end sequencing on two flow-cell lanes of an Illumina HiSeq 2000 sequencer ( average insert size = 203 bp; read length = 101 bp ) . Raw reads are available in NCBI’s short read archive under BioProject accession PRJNA268379 . The procedures for assembly and annotation of the transcriptome described below , as well as the procedures for read mapping to quantify differential gene expression , have been described in detail in previous publications from this laboratory [25 , 50] . Raw reads were first screened using ssaha2 [60] and the UniVec database ( accessed July 7th , 2013 ) to remove vector sequences , adapters , linkers , and primers commonly used in cloning cDNA or genomic DNA as well as Ae . albopictus rRNA sequences ( GenBank accession L22060 . 1 ) . The cutoff for sequence removal was 95% identity and an alignment score of 18 . In addition , Illumina sequencing multiplexing adapters were identified using ssaha2 , with 100% percent identity and an alignment score of 18 as the cutoff for removal . In all cases , contaminated reads were removed along with their read mates in a pair . After ssaha2 screening , cleaned reads were further filtered using SolexaQA V2 . 2 to retain contiguous reads longer than 50 bp with phred quality scores higher than 30 . We performed one round of digital normalization [61] to reduce the number of redundant reads and the computational requirements for assembly . This method has previously been shown to drastically reduce the memory necessary for assembly while maintaining the assembly quality [50] . Digital normalization was performed with the default parameters , including a kmer coverage cutoff of 20 and a kmer size of 20 . De novo assembly was performed using Trinity [62] ( released February 25th , 2013 ) with default parameters . A minimum kmer coverage of 2 was used to reduce memory requirements with hundreds of millions of read pairs . In order to produce a comprehensive transcriptome assembly for Ae . albopictus , the contigs from the current experiment ( adult ) were combined with previous contigs from de novo assemblies of pre-adult stages , including mature oocytes [24] , developing embryos [25] and pharate larvae [26 , 50] using a modified Scaffolding Translation Mapping approach [63] ( see Table 1 ) . To eliminate contig redundancy within each life stage , before assembling a comprehensive transcriptome , de novo assembled contigs were clustered within each life stage using CD-HIT-EST [64 , 65] . Redundant contigs ( percent identity > = 99% ) in each cluster were eliminated , and the longest representative contigs were retained . Next , reference-based re-assembly and annotation of the contigs were performed simultaneously as described previously in two publications from this laboratory [25 , 50] and explained below . The parameters used in the reference-based re-assembly followed our previous publications [25 , 50] , and were chosen to be conservative to maximize our confidence in the analyses based on the transcriptome assembly . The assembly is available at http://www . albopictusexpression . org/ ? q=data . A non-redundant dipteran protein reference set was generated by downloading orthologous protein sequences from Ae . aegypti , Culex quinquefasciatus , Anopheles gambiae and Drosophila melanogaster from OrthoDB [66] , Version 7 ( accessed on September 13th , 2013 ) , with one single ortholog retained per ortholog group in the order specified above ( i . e . , order of relatedness to Ae . albopictus ) . The final reference set contained 19 , 272 protein sequences , and represented a wide range of evolutionary diversity within Diptera with little redundancy . The merged contig set from adult and pre-adult stages was first aligned to the dipteran protein set by BLASTX [67] . Alignments with e-value ≤ 1e-6 were retained for subsequent analysis . Contigs that aligned to the same reference and with more than 95% identity at the overlapping regions were reassembled by CAP3 [68] . To verify the annotation of the re-assembled contigs , they were again aligned to the dipteran protein set by BLASTX and only the contigs that matched the original annotation ( e-value ≤ 1e-6 ) were retained . Only contigs with more than 70% identity to the matching reference were retained in the final assembly . Chimeric contigs were identified as having secondary alignments longer than 50 bp outside of the primary alignment with > 80% percent identity of the primary alignment . Chimeric contigs were discarded . Contigs not retained in the protein reference-based re-assembly were subsequently used in a genome reference-based re-assembly using genomic scaffolds from Ae . aegypti [51] as a reference ( accessed from VectorBase on September 30th , 2013 with SCAFFOLDS_AaegL1 for alignment and TRANSCRIPTS_AaegL1 . 4 for annotation ) . To reduce the computational requirements for alignment against the Ae . aegypti genome , contigs were first matched to the Ae . aegypti genomic scaffolds using BLASTN [67] ( e-value ≤ 1e-6 ) to find the best matching scaffolds for each contig . Next , the contigs were aligned to their best matching scaffolds using EXONERATE [69] ( parameters were—model est2genome—softmasktarget TRUE—bestn 1—dnahspdropoff 0 ) . To be conservative , only the top 95% of contigs with the best alignments ( percent identity > 71 . 78% ) were retained in subsequent analysis . Contigs that aligned to the same reference with > 95% identity of the overlapping regions were re-assembled by CAP3 [68] . To verify the annotation , re-assembled contigs were re-aligned to the Ae . aegypti genomic scaffolds by EXONERATE with the same parameters as the first alignment and only the contigs that matched the original annotation were retained . Similar to the first EXONERATE alignment , the top 95% of contigs with the best alignments ( percent identity > 66 . 67% ) were retained in the final assembly . The lower percent identity of the top 95% of re-assembled and re-aligned contigs reflects the longer contig length of this set . Similar to the protein reference-based re-assembly , chimeric contigs were identified as contigs having secondary alignments outside of the primary alignment and were discarded . Contigs that aligned within 1 kb up- or down-stream of annotated gene models with at least 90% of their length covered by the alignment were identified as potential untranslated regions ( UTRs ) for those genes . Contigs that did not align within 1 kb up- or down-stream of annotated gene models were labeled as unannotated genomic contigs . As a result of the procedures described above , the final transcriptome assembly contained contigs annotated based on the dipteran protein reference set , gene models in the Ae . aegypti genome , potential UTRs and unannotated Ae . aegypti genomic regions . The final transcriptome assembly described above was used as a reference for read mapping to quantify levels of gene expression . Cleaned paired-end reads from all 16 libraries were mapped to the annotated full assembly using RSEM ( [70] , Version 1 . 2 . 4 ) to calculate read counts at the unigene level , accounting for redundancy because of allelic variation and/or alternative splicing . Reads counts were then processed in edgeR [71] in the R software environment ( www . r-project . org ) . First , the read counts were TMM normalized in edgeR to account for differences in library sizes and the total numbers of mRNAs sequenced across samples [71] . Genes with log counts per million smaller than one in at least four libraries were discarded as too rare for the differential expression analysis ( see [72] ) . Differentially expressed ( DE ) genes were identified as having an absolute value of log2 fold-change greater than 0 . 5 with a Benjamini-Hochberg ( FDR ) -corrected p-value less than 0 . 05 [25 , 26] . Previous studies using the same population , sequencing facility , transcriptome assembly , read mapping and normalization methods indicate high correspondence ( Pearson’s r = 0 . 92 , 20 genes ) between gene expression levels measured by RNA-Seq and qRT-PCR [25] . Distance matrices of gene expression across the libraries ( R function dist ) were visualized using multi-dimensional scaling ( R function cmdscale ) after raw read counts were transformed for linear modeling via the function voom [73 , 74] in limma [75 , 76] . To identify differential expression of functionally related groups of genes we tested for KEGG pathways [77 , 78] that were enriched for DE genes using GOSeq [79] , which corrects for the selection bias of DE genes caused by transcript length . KEGG pathway assignments were downloaded from http://www . genome . jp/kegg/ . We used KEGG pathway assignments from Ae . aegypti , which is in the same subgenus ( Stegomyia ) as Ae . albopictus and the most closely related species with well-documented KEGG pathway information . KEGG pathways were considered significantly enriched ( i . e . , over-represented ) , if there were five or more DE genes in the group with a FDR corrected p-value of over-representation less than 0 . 05 [25 , 26] . In addition to enriched KEGG pathways , we also analyzed additional pathways based on preliminary data analysis and/or a priori expectations concerning the molecular physiology of diapause induction . These pathways included fatty acid metabolism , DNA replication and cell cycle regulation . For DNA replication and cell cycle regulation , additional genes not included in the KEGG pathways but documented on the Interactive Fly ( http://www . sdbonline . org/sites/fly/aimain/1aahome . htm ) database were added to the analysis . Expression levels of all DE genes in enriched pathways were standardized as Z-scores and visualized as heat maps generated using hierarchical clustering ( R function hclust ) .
Diapause incidence ranged from 81 . 4% to 94 . 8% among three biological replicates of the females reared under diapause-inducing ( SD ) conditions and from 1 . 5% to 3 . 3% among three biological replicates of females reared under non-diapause-inducing ( LD ) conditions ( S1 Table ) . 472 , 006 , 080 read pairs were obtained from the Illumina HiSeq 2000 platform of which 182 , 036 , 362 read pairs and 108 , 891 , 552 single-end reads ( 472 , 964 , 276 total reads ) were retained after read cleaning . After digital normalization , 285 , 061 , 839 reads were utilized for de novo assembly via Trinity . De novo assembly of reads from the adult stage via Trinity produced 155 , 321 contigs with a median contig length of 497 bp ( mean = 993 bp; Table 1 ) . These contigs were combined with 539 , 506 contigs from the pre-adult stages of Ae . albopictus with a median contig length of 569 bp ( mean = 972 bp ) . The resulting protein reference-based re-assembly with contigs merged using CAP3 produced 95 , 863 contigs with a median length of 1 , 331 bp ( mean = 1 , 865 bp; Table 1 ) . The median contig length for the genome reference-based re-assembly was 533 bp ( mean = 924 bp ) . The annotated full assembly including both protein reference-based and genome reference-based contigs had a median contig length of 824 bp ( mean = 1 , 399 bp; Table 1 ) . Percent identity and coverage for the assemblies at various stages are presented in S1 Fig . Most contigs in the final transcriptome assembly were highly similar to the orthologous sequences in the dipteran protein reference set ( S1A Fig , median percent identity = 92 . 86% ) and the Ae . aegypti genome ( S1A Fig , median percent identity = 80 . 25% ) . Contig coverage from the protein reference-based re-assembly was intermediate ( S1B Fig , median coverage = 50 . 74% ) , due to the inclusion of UTRs in the contigs but not in the reference . Contig coverage from the genome reference-based re-assembly was high ( S1B Fig , median coverage = 71 . 54% ) , indicating that most contigs were nearly fully utilized in the alignments to the genomic scaffolds of Ae . aegypti . Most gene models from the protein reference-based re-assembly represented in our transcriptome were nearly full-length transcripts ( S1C Fig , median coverage = 73 . 24% ) . Genome reference-based re-assembly had low reference coverage ( S1C Fig ) because the genomic references are genomic scaffolds that are usually several hundred kilo-bases long . 14 , 077 non-redundant gene models were represented in the annotated full assembly , with 11 , 394 gene models in the protein reference-based re-assembly and another 8 , 636 gene models in the genome reference-based re-assembly . Of the 11 , 394 annotations based on the non-redundant Dipteran protein reference set , 10 , 296 were from Ae . aegypti proteins . 5 , 953 gene models were redundant between the two re-assemblies . The total of 14 , 077 annotated transcripts accounted for 80 . 9% of the annotated gene models in Ae . aegypti ( [51]; 17 , 391 protein coding genes from the Liverpool strain AaegL1 . 4 from VectorBase . org ) . Of 472 , 964 , 276 quality-filtered reads obtained from 16 libraries , we excluded unpaired single reads resulting in 182 , 036 , 362 retained read pairs ( ~364 million total reads ) . 89 . 44% of these read pairs mapped to the annotated full transcriptome assembly . Overall , differential gene expression profiles from biologically replicated libraries clustered tightly within each experimental treatment ( Fig 2 ) . One biological replicate from the SD blood-fed treatment exhibited increased variation relative to the other samples but was included in all downstream analyses to be conservative . Under non-diapause-inducing LD conditions , 920 genes were significantly up-regulated in response to a blood meal ( LD blood fed vs . LD non-blood fed ) , and 849 genes were significantly down-regulated ( Fig 3A ) . Under diapause-inducing SD conditions , 1 , 566 genes were significantly up-regulated in response to a blood meal ( SD blood fed vs . SD non-blood fed ) , and 1 , 408 genes were significantly down-regulated ( Fig 3A ) . There are 665 genes that were significantly up-regulated in response to a blood meal both under SD and LD conditions , and 603 genes were significantly down-regulated in response to a blood meal both under SD and LD conditions ( S2 Table ) . In non-blood-fed females , 1 , 293 genes were significantly up-regulated under diapause-inducing SD conditions ( SD non-blood fed vs . LD non-blood fed ) , and 524 genes were significantly down-regulated ( Fig 3B ) . In blood-fed females , 766 genes were significantly up-regulated under diapause-inducing SD conditions ( SD blood fed vs . LD blood fed ) , and 111 genes were significantly down-regulated ( Fig 3B ) . There are 406 genes that were significantly up-regulated under SD vs . LD conditions in both non-blood-fed and blood-fed females , and 37 genes were significantly down-regulated under SD vs . LD conditions in both non-blood-fed and blood-fed females ( S2 Table ) . Potential uniquely expressed genes under SD or LD with zero read count in the other photoperiodic treatment were rare in our transcriptome . This category included 184 genes of which > 97% had no more than 10 reads in any individual library . As a result , these genes were not included in the differential expression analysis . Information for all unigenes used in the differential expression analysis is presented in S2 Table . Vitellogenin-A1 precursor ( PVG1 ) and two trypsin genes were significantly up-regulated in response to a blood meal under both diapause-inducing and non-diapause-inducing conditions ( Table 2 ) , reflecting transcriptional up-regulation of vitellogenesis and blood digestion [55 , 80] . PVG1 up-regulation in response to a blood meal was greater under diapause than non-diapause conditions ( blood feeding×photoperiod interaction , p = 0 . 03 ) . Genes involved in detoxification , such as glutathione S-transferases ( i . e . , glutathione transferases ) and thioredoxin peroxidases were also up-regulated in response to a blood meal ( S2 Table ) , consistent with previous studies [81–83] . In addition , many stress response genes were differentially expressed in response to a blood meal ( S2 Table ) . Many cytochrome P450s were also differentially expressed in response to a blood meal ( S2 Table ) . Of special interest are CYP302A1 and the homolog of Spook , both of which were up-regulated in response to a blood meal under both SD and LD photoperiods ( Table 2 ) . CYP302A1 encodes the ecdysteroid 22-hydroxylase , a protein that catalyzes one of the final reactions in the synthesis of 20-hydroxyecdysone ( 20-E ) and Spook is one of the Halloween genes implicated in synthesizing 20-E . The up-regulation of these genes in response to a blood meal is consistent with the well-established role of 20-E in stimulating vitellogenesis in response to a blood meal [55] . Finally , CYP314A1 , which encodes ecdysone 20-monooxygenase , an enzyme catalyzing the final step in conversion of ecdysone to 20-E , was up-regulated only under diapause-inducing conditions in females with a blood meal ( Table 2 ) . In addition , all except one of the genes encoding JH-inducible proteins were down-regulated in blood-fed females ( S2 Table ) , consistent with decreasing juvenile hormone ( JH ) titers after a blood meal [84] . Four KEGG pathways , three of which are related to amino acid metabolism , were enriched for differentially expressed genes under SD vs . LD conditions in females without a blood meal ( Table 3 ) . The differential expression patterns for enriched amino acid metabolism pathways under SD vs . LD conditions are summarized in a heat map ( S2 Fig ) . The gene encoding one minor enzyme synthesizing glycine , threonine dehydrogenase ( AAEL003443 , S2 Table ) , was down-regulated under diapause-inducing conditions in non-blood-fed females , but genes encoding major enzymes synthesizing glycine in mammals [85] were all up-regulated under diapause-inducing conditions in non-blood-fed but not in blood-fed females ( S2 Table ) , including serine hydroxymethyltransferase ( AAEL002510 ) , sarcosine dehydrogenase ( AAEL014936 ) and alanine:glyoxylate aminotransferase ( AAEL000640 and AAEL012464 ) . In addition to amino acid metabolism pathways , the KEGG pathway for global metabolism was enriched for DE genes in non-blood-fed females under SD vs . LD conditions ( Table 3 ) . Similarly , all DE genes that are positive cell-cycle regulators were under-expressed under SD conditions in non-blood-fed females , and one negative cell cycle regulator , the growth arrest and DNA damage , or GADD45 , was over-expressed ( Fig 4A ) . Although not detected by KEGG pathway enrichment analysis , all DE genes involved in DNA replication were down-regulated under SD conditions in non-blood-fed females ( Fig 4B ) . Consistent down-regulation in these two KEGG pathways related to cell proliferation indicates that under diapause-inducing SD conditions non-blood-fed females down-regulate cell proliferation . The clock genes timeless and cryptochrome 1 were up-regulated under SD conditions in non-blood-fed females ( Table 4 ) but period ( AAEL008141 ) and clock ( AAEL012562 ) were not differentially expressed ( S2 Table ) . Similarly , phosphoenolpyruvate carboxykinase ( pepck ) was also up-regulated under SD conditions only in non-blood-fed females ( significant blood feeding×photoperiod interaction , Table 4 ) . Delta ( 9 ) -desaturase and delta ( 9 ) -desaturase 2 were up-regulated under SD conditions in non-blood-fed females ( Table 4 ) . Out of eight differentially expressed genes encoding putative JH-inducible proteins under SD conditions , seven genes were up-regulated in non-blood-fed females ( Table 4 ) . The oxidative phosphorylation pathway was not significantly enriched under SD conditions in non-blood-fed females ( but see below ) , but all 7 DE genes were up-regulated ( Fig 4C ) . In contrast to non-blood-fed females under SD conditions , in blood-fed females under SD conditions the oxidative phosphorylation KEGG pathway was significantly enriched ( Table 3 ) , with all DE genes up-regulated ( Fig 4C ) . The proportion of up-regulated genes ( 29/89 ) in the oxidative phosphorylation pathway in blood-fed females was significantly higher than that ( 7/89 ) in non-blood-fed females ( Fisher’s exact test , p-value = 1 . 93e-05 ) . Similar to non-blood-fed females under SD conditions , in blood-fed females under SD conditions the KEGG pathway for global metabolism was enriched for DE genes ( Table 3 ) . The proportion of up-regulated genes in the pathway under SD conditions in non-blood-fed females ( 108/665 ) did not differ significantly from that in blood-fed females ( 89/665 ) ( Fisher’s exact test , p-value = 0 . 16 ) , but more genes involved in metabolism were down-regulated under SD conditions in non-blood-fed females ( 31/665 ) than in blood-fed females ( 5/665 ) ( S3 Fig; Fisher’s exact test , p-value = 1 . 00e-05 ) . Fatty acid synthase , fatty acid desaturase and delta ( 9 ) -desaturase 2 were up-regulated under diapause-inducing conditions in blood-fed females but not in non-blood-fed females ( Table 4 ) . Similar to non-blood-fed females under SD conditions , two amino acid metabolism pathways were significantly enriched under SD conditions in blood-fed females ( Table 3 ) . In the valine , leucine and isoleucine degradation pathway , branched-chain amino acid ( BCAA ) aminotransferase ( AAEL007909 , S2 Table ) was up-regulated under diapause-inducing conditions both in non-blood-fed and blood-fed females . In the beta-alanine metabolism pathway , one enzyme involved in synthesizing beta-alanine from uracil in insects , dihydropyrimidine dehydrogenase [86] , was up-regulated both in non-blood-fed and blood-fed females under SD conditions ( AAEL014199 and AAEL010204 , S2 Table ) . Out of eight differentially expressed genes encoding putative JH-inducible proteins under SD conditions , four genes were up-regulated in blood-fed females ( Table 4 ) .
We combined 155 , 321 contigs obtained in this study from the adult life stage with 539 , 506 contigs obtained previously from pre-adult stages to produce a composite transcriptome assembly ( Table 1 ) . The number of gene models identified from the current comprehensive assembly including the adult stage increased to 14 , 077 from 13 , 261 gene models identified based on the pre-adult stages [50] . These 14 , 077 gene models represent approx . 81% of all annotated gene models in Ae . aegypti ( AaegL1 . 4 ) . The annotated comprehensive assembly is also moderately improved relative to previous assemblies [25 , 50] in terms of median contig length , contig coverage and gene model coverage ( Table 1 and S1 Fig ) . Based on comparison with the Ae . aegypti genome , this comprehensive transcriptome assembly likely represents the majority of the coding regions in the Ae . albopictus genome and therefore provides a powerful resource to effectively investigate global transcriptional components of diapause induction . Similar to our results , a previous RNA-Seq analysis of the transcriptional response to a blood meal in the closely related mosquito Ae . aegypti found extensive differential gene expression [89] . However , because adult females in this previous study [89] were maintained at 28°C rather than 21°C and RNA was extracted at 5 hours rather than 26–28 hours post blood meal , our results are not directly comparable . Nevertheless , extensive additional information on the transcriptional response to a blood meal in Ae . aegypti under LD conditions allows us to validate the transcriptional response to blood feeding in Ae . albopictus . For example , vitellogenin synthesis and blood digestion are key physiological components of the transcriptional response to a blood meal [55] . Both vitellogenin synthesis and typsin activity reach their peak at approx . 24 hours post blood meal in Ae . aegypti maintained at 27°C [80 , 90] . Populations used in this study were maintained at 21°C in order to optimally stimulate a robust diapause response , and thus the peaks of vitellogenin synthesis and blood digestion likely would have occured after 24 hours pbm . Therefore , the time point of sampling in this study , 26–28 hours pbm , is expected to correspond to near the peak of vitellogenin synthesis and blood digestion , consistent with the highly elevated transcriptional profiles of PVG1 and trypsins in response to a blood meal ( Table 2 ) . 20-hydroxyecdysone ( 20-E ) is an essential hormone stimulating vitellogenesis after females take a blood meal [55] . We found that three genes encoding enzymes in the 20-E synthesis pathway [91] were up-regulated in response to a blood meal ( Table 2 ) . Juvenile hormone also plays a crucial role in regulating the reproduction of adult mosquitoes [55] . In Ae . aegypti , after females take a blood meal , JH titers decrease until the end of a gonotrophic cycle [84] , in antiphase with 20-E [55] . Consistent with this pattern , our results show that 13 out of the 14 DE genes encoding putative JH-inducible proteins were down-regulated in response to a blood meal under either diapause-inducing or non-diapause-inducing conditions ( S2 Table ) . Cytochrome P450 ( CYP ) monooxygenases are mainly involved in hormone synthesis and insecticide resistance [92] . All three 20-E synthesizing CYP enzymes noted above ( Table 2 ) were up-regulated in response to a blood meal , consistent with the role of 20-E in promoting vitellogenesis . Additionally , glutathione S-transferases and thioredoxin peroxidases are mostly up-regulated in response to a blood meal ( S2 Table ) , potentially as a response to oxidative stress since these enzymes remove intracellular reactive oxygen species . It is also possible that the glutathione transferases might be involved in heme detoxification [93] . Finally , six out of seven genes involved in response to water stress were down-regulated in response to a blood meal . These results are generally consistent with a previous study in the Ae . aegypti midgut [83] and in the Ae . albopictus malpighian tubules [94] . These genes are likely related to osmotic stress and the intake of toxic substances ( i . e . , heme ) or microbes associated with blood feeding . Overall , the transcriptional responses to a blood meal detected in this study are consistent with previous studies and support the conclusion that transcriptome sequencing of whole bodies captured the major physiological benchmarks of the response to a blood meal . Furthermore , the overall level of differential expression in response to a blood meal was similar under SD and LD conditions . Under SD conditions , 11 . 2% more genes were up-regulated than down-regulated in response to a blood meal . Under LD conditions , 8 . 4% more genes were up-regulated than down-regulated . We hypothesized that some of the “upstream” transcriptional components of diapause induction would occur before adult female Ae . albopictus obtain access to a blood meal . Consistent with hypothesis , we found that timeless ( tim ) and cryptochrome 1 ( cry1 ) , two essential components of the circadian clock pathway in insects [39] , were up-regulated under diapause-inducing conditions in non-blood-fed females ( Table 4 ) . Because we controlled for circadian effects on gene expression by harvesting female whole bodies at the same Zeitgeber time , the increased expression of tim and cry1 under SD relative to LD conditions is interpreted as a response to diapause-inducing photoperiods . For almost 80 years , researchers have hypothesized that the circadian clock constitutes the underlying molecular mechanism for photoperiodic time measurement [95] . However , the causative link between these two biological timing systems remains unresolved and controversial [38 , 42] . In both Anopheles gambiae and Ae . aegypti , under LD , tim decreases at ZT 6–8h [96 , 97] relative to earlier and later peaks in the 24-hour cycle . In contrast , cry1 increases at ZT 6–8h in An . gambiae [97] , consistent with the role of CRY1 in the light-dependent degradation of TIM [39] . Among dipteran species , tim is required for diapause induction of Chymomyza costata [98 , 99] . At ZT 6–8h , the transcript level of tim is up-regulated both in the photosensitive larval brain of Sarcophaga crassipalpis under diapause-inducing conditions [100] and in the diapausing Wyeomyia smithii fourth instar larvae [101] . In Drosophila triauraria , additive allelic differences in tim and cry1 between diapause and non-diapause strains are positively associated with diapause incidence [102] . As a fundamental physiological timekeeper , the circadian clock system is responsible for the rhythmic expression patterns of thousands of genes throughout the 24-hour daily cycle [97 , 103] . Therefore , abnormal expression patterns in the circadian clock system are expected to cause considerable disruptions in the expression of genes important for a wide range of physiological functions other than diapause . Hence , it has been proposed that tim may be functionally involved in measuring photoperiodic ( seasonal ) time independent of its role in the circadian clock [42 , 104] . Our results are consistent with this hypothesis . In addition to tim , period ( per ) and clock ( clk ) are core components of the transcriptional negative feedback loop that drives the oscillatory behavior of the circadian clock [39] . Differential expression of tim but not per and clk in response to diapause-inducing short day lengths suggests that short-day photoperiods do not cause fundamental changes to the oscillatory behavior of the circadian clock . Because CRY1 is responsible for the light-sensitive degradation of TIM , we hypothesize that products of the CRY1-mediated breakdown of TIM could serve as a component of an “interval” photoperiodic timer in Ae . albopictus , independent of the circadian clock pathway . Two amino acid metabolism KEGG pathways were significantly enriched for differentially expressed genes in non-blood-fed females under diapause-inducing conditions: A ) alanine , aspartate and glutamate metabolism , and B ) glycine , serine and threonine metabolism ( Table 3 ) . Alanine levels increase during diapause initiation in Teleogryllus emma [105] , B . mori [106 , 107] , and Ostrinia furnacalis [108] . Alanine levels also increase during diapause in S . crassipalpis [109] , and both before and during diapause in Antheraea pernyi [110] . In A . pernyi , alanine levels also decrease as diapause terminates [110] . Despite this widespread association of alanine with the diapause program in a broad range of insects , the biological significance has not been elucidated . In this study , alanine aminotransferase ( AAEL009872 ) and alanine-glyoxylate aminotransferase ( AAEL000640 ) were up-regulated under diapause-inducing conditions in NB females ( S2 Table ) , suggesting increased metabolism of alanine . We hypothesize that alanine may be provisioned to the diapause offspring from the mother and could serve as a cryoprotectant as has been proposed for the diapause eggs of B . mori [107] . Inspection of genes in the glycine , serine and threonine metabolism KEGG pathway indicates that the gene encoding a minor enzyme synthesizing glycine , threonine dehydrogenase , was down-regulated under diapause conditions in non-blood-fed females . However , genes encoding major enzymes synthesizing glycine in mammals [85] were all up-regulated , including serine hydroxymethyltransferase ( AAEL002510 ) , sarcosine dehydrogenase ( AAEL014936 ) and alanine:glyoxylate aminotransferase ( AAEL000640 ) ( S2 Table ) . Diapause-destined larvae of Helicoverpa armigera accumulate more glycine [111] . In Leptinotarsa decemlineata , glycine-rich transcripts are up-regulated during diapause initiation phase [112] . Glycine has been implicated to regulate protein synthesis in vertebrates , and it could also regulate growth and development by serving as an indicator of nutrient levels [85] . In non-blood-fed Ae . albopictus adults during diapause induction , both positive cell cycle regulators and DNA replication transcripts were down-regulated ( Fig 4A and 4B ) , particularly the positive cell cycle regulator proliferating cell nuclear antigen ( pcna ) . Transcriptional suppression of the cell cycle is a common molecular hallmark of the diapause program during the developmental arrest stage of diapause , as illustrated in S . crassipalpis [113 , 114] , Helicoverpa armigera [115] and C . costata [116] . Furthermore , pcna is down-regulated during diapause both in S . crassipalpis [113 , 114] , and C . costata [116] . However , cell cycle transcripts are up-regulated after diapause termination in Rhagoletis pomonella [28] . The pcna transcript is up-regulated after diapause termination both in S . crassipalpis [113] and R . pomonella [28] , in synchrony with other changes in cell cycle regulation . In the current study we examined the adult stage which represents diapause induction rather than developmental arrest . Thus , our results are likely not relevant to cessation of the development during diapause . Rather , we hypothesize that cell proliferation is down-regulated under diapause-inducing conditions before females take a blood meal to allocate energy to alternative metabolic pathways . This interpretation is consistent with results from the oxidative phosphorylation pathway discussed below . These results are also consistent with a previous study showing alteration of the cell cycle during diapause preparation in early Ae . albopictus embryos [25] and emphasize that diapause induction involves the alteration of fundamental cellular processes far in advance of developmental arrest . The pepck transcript was up-regulated under diapause-inducing condition in non-blood-fed but not blood-fed females ( significant interaction , Table 4 ) . The up-regulation of pepck is similar to previous studies in mature oocytes , developing embryos and pharate larvae under diapause conditions in Ae . albopictus [24–26] . The pepck transcript is also up-regulated under diapause conditions in several other insect species , including S . crassipalpis [27] , and R . pomonella [28] , as well as in the dauer phenotype ( the counterpart of diapause in nematodes ) of Caenorhabditis elegans [117] . In W . smitthii , up-regulation of pepck is associated with diapause termination [118] . The pepck transcript is involved in response to cold and desiccation in Belgica antarctica [119] and in response to hormone stimulation in Drosophila [120] . It is also down-regulated by nectarine supplemented diet which increases longevity in Drosophila [121] . The observation that pepck was differentially expressed under diapause-inducing conditions only in non-blood-fed females suggests pepck could be a regulatory component of pre-diapause metabolism , potentially triggering a cascade of metabolic responses after the females take a blood meal . In light of its association with the diapause program across multiple stages in Ae . albopictus , with diapause induction or termination phases in other insects , and with stress resistance and hormonal response , it is increasingly evident that pepck is a central component of diapause metabolism in a wide range of organisms . Juvenile hormone has been implicated in the regulation of larval diapause in a variety of species [36] and the absence of JH has been demonstrated to initiate adult reproductive diapause in Cx . pipiens [33–35] . However , a role of JH in pharate larval diapause has not been noted in previous studies . For Ae . albopictus , in non-blood-fed females all seven differentially expressed genes encoding putative JH-inducible proteins were up-regulated under SD conditions ( Table 4 ) . Before a blood meal , JH induces the primary follicles to enter a resting stage , and also renders the fat body competent for vitellogenin synthesis after a blood meal [55] . Increased JH-induced signaling under diapause conditions in non-blood-fed females likely enhances the fat body’s competence for vitellogenin synthesis after a blood meal , thereby increasing vitellogenesis for augmented nutrient provisioning to offspring destined to undergo diapause . In blood-fed females under SD conditions , all four differentially expressed genes encoding putative JH-inducible proteins were up-regulated ( Table 4 ) . Under non-diapause conditions , JH levels are expected to decrease after a blood meal [84] . Therefore , the up-regulation of JH-induced signaling under diapause conditions in blood-fed females implies altered reproductive endocrinology during diapause induction . Energy metabolism is crucial for the survival of diapause insects through the winter . Levels of nutrient reserves during diapause directly affect overwinter survival , as well as post-diapause development and reproduction [29] . Photoperiodic diapause is determined maternally in Ae . albopictus , and diapause offspring ( pharate larvae inside the egg ) cannot obtain additional resources . As a result , maternal provisioning of diapause eggs is expected to have a large impact on offspring fitness . In fact , previous studies have established that diapause eggs of Ae . albopictus are larger and contain more total lipids than non-diapause eggs [49] . Consistent with these considerations , energy production ( oxidative phosphorylation ) and overall metabolism were elevated under SD conditions in blood-fed females ( Fig 4C and S3 Fig ) . Overall , our study suggests that after taking a blood meal , females exposed to SD signals enhance energy production through the oxidative phosphorylation pathway , presumably to meet the energetic requirements for generating more nutrients to provision the offspring destined to undergo diapause as described below . Analysis of individual genes up-regulated in blood-fed females under SD conditions provides further insight into the molecular basis of provisioning of diapause eggs . For example , up-regulation of vitellogenin synthesis gene PVG1 in response to a blood meal is greater under SD than LD conditions ( significant interaction , Table 2 ) . Additionally , fatty acid synthase ( fas ) was up-regulated under SD conditions only in blood-fed females ( significant interaction , Table 4 ) , indicating that blood-fed females synthesize more fatty acids under diapause-inducing than non-diapause-inducing conditions . This is consistent with previous results stated above that diapause eggs contain more total lipids compared to non-diapause eggs [49] . In Cx . pipiens , fas is elevated in diapause-destined females that overwinter at the adult stage , consistent with our results and the general importance of lipids as nutrient reserves during diapause [29] . Three genes encoding fatty acid desaturases were also up-regulated under SD conditions ( significant interaction in fatty acid desaturase , Table 4 ) , indicating that synthesis of desaturated fatty acids ( UFAs ) was increased in females exposed to diapause-inducing conditions . UFAs are proposed to enhance cold tolerance during diapause by preserving membrane permeability under low temperatures [122] . Our results are consistent with the previous studies that fatty acid desaturation is enhanced under diapause conditions [29 , 123] and cold acclimation [124] . These results imply that maternal provisioning of UFAs to the offspring contribute to the increased cold tolerance of diapause relative to non-diapause eggs in Ae . albopictus [46] . Two amino acid metabolism pathways were significantly enriched in blood-fed females under diapause-inducing conditions: valine , leucine and isoleucine degradation , as well as beta-alanine metabolism ( Table 3 ) . Michaud and Denlinger [109] reported increased leucine during the pupal diapause of S . crassipalpis . In mammals , leucine stimulates protein synthesis [125] , but no research has been performed regarding the effect of leucine on diapause in invertebrates . However , in the light of enhanced vitellogenesis in females exposed to SD with a blood meal , increased leucine might stimulate more protein synthesis after a blood meal to provision the diapause offspring . Beta-alanine has been implicated to recycle the photoreceptor neurotransmitter histamine in the photoreceptor cells of Drosophila [126] . One gene involved in synthesizing beta-alanine from uracil in insects , dihydropyrimidine dehydrogenase [86] , was up-regulated both in non-blood-fed and blood-fed females under diapause conditions ( S2 Table ) . The role of beta-alanine has not been examined in terms of diapause response , but increased beta-alanine under SD conditions might be used for differentially measuring photoperiod via its ability to recycle histamine , the photoreceptor neurotransmitter in insects . Diapause is an adaptive developmental plasticity of crucial ecological importance . Our results show that thousands of genes are differentially expressed under diapause-inducing conditions ( Fig 3B ) , but only approximately 1% of all genes are potentially uniquely expressed under diapause-inducing or non-diapause-inducing conditions . Therefore our study indicates that the transcriptional basis of diapause induction is primarily a quantitative rather than qualitative response , with changes involving mostly levels of transcription rather than specific genes that are uniquely expressed under either diapause or non-diapause conditions . This conclusion implies that a wide range of fundamental physiological pathways modified as part of the diapause response may also provide novel targets for genetic or chemical disruption under non-diapause conditions . We have identified novel putative regulatory elements of diapause induction ( i . e . , tim and cry1 ) , and our study confirms previous hallmarks of insect diapause at the transcriptional level , such as cell cycle regulation , pepck and lipid metabolism . Diapause appears to have evolved independently in several lineages within both Culicidae [19] and Diptera [127] . Our study supports a previous hypothesis [25] that despite hundreds of millions of years of evolution among dipteran species , a conserved set of genes has been repeatedly targeted by selection during the evolution of diapause in independent lineages , including pepck , pcna and fas . These genes provide targets for functional studies aimed at developing novel control strategies designed to disrupt the photoperiodic diapause response , a crucial ecological adaptation in a wide range of pest and vector species . Below is a list of the genes mentioned in the text and their Ensembl IDs , in the order of their appearance: vitellogenin-A1 precursor , AAEL010434; serine protease I , AAEL007432; late trypsin , AAEL013284; glutathione S-transferases/glutathione transferases , AAEL000092 , AAEL001061 , AAEL001090 , AAEL004229 , AAEL007947 , AAEL007955 , AAEL007964 , AAEL010157 , AAEL010582 , AAEL010591 , AAEL011741 , AAEL011752 , AAEL011934 , CPIJ018630; thioredoxin peroxidases , AAEL002309 , AAEL004112 , AAEL014548; CYP302A1 , AAEL015655; Spook , AAEL009762; CYP314A1 , AAEL010946; threonine dehydrogenase , AAEL003443; serine hydroxymethyltransferase , AAEL002510; sarcosine dehydrogenase , AAEL014936; alanine:glyoxylate aminotransferase , AAEL000640 , AAEL012464; growth arrest and DNA damage , or GADD45 , AAEL006883; timeless , AAEL006411; cryptochrome 1 , AAEL004146; period , AAEL008141; clock , AAEL002049; phosphoenolpyruvate carboxykinase , AAEL000006 , AAEL000080; delta ( 9 ) -desaturase , AAEL007213; delta ( 9 ) -desaturase 2 , AAEL004573; fatty acid synthase , AAEL001194; fatty acid desaturase , AAEL007516; branched-chain amino acid aminotransferase , AAEL007909; dihydropyrimidine dehydrogenase , AAEL014199 , AAEL010204; alanine aminotransferase , AAEL009872; proliferating cell nuclear antigen , AAEL012545 . | The mosquito , Aedes albopictus , is an aggressive human biter capable of transmitting Dengue virus , Chikungunya virus and at least 22 additional viruses that cause human illness . Over the last 30 years , this mosquito has spread from its native Asian range to all continents except Antarctica . Efforts to control this mosquito have met with limited success . Photoperiodic diapause refers to the ability of insects to measure day length ( photoperiod ) as a cue for initiating developmental arrest ( dormancy ) in order to survive unfavorable seasonal conditions such as winter . Photoperiodic diapause is a crucial ecological adaption that enables Ae . albopictus and other medically important mosquitoes to inhabit temperate environments and spread across broad geographic ranges . Here , we identify genes that exhibit changes in expression levels ( up-regulation or down-regulation ) in association with the induction of photoperiodic diapause in Ae . albopictus . Some of these genes , based on their known biological function in other organisms , are implicated in regulating photoperiodic diapause and represent exciting targets for novel vector control strategies based on genetic or chemical disruption of this important adaptation . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | Global Transcriptional Dynamics of Diapause Induction in Non-Blood-Fed and Blood-Fed Aedes albopictus |
We have established two mouse models of central nervous system ( CNS ) demyelination that differ from most other available models of multiple sclerosis ( MS ) in that they represent a mixture of viral and immune triggers . In the first model , ocular infection of different strains of mice with a recombinant HSV-1 that expresses murine IL-2 constitutively ( HSV-IL-2 ) causes CNS demyelination . In the second model , depletion of macrophages causes CNS demyelination in mice that are ocularly infected with wild-type ( WT ) HSV-1 . In the present study , we found that the demyelination in macrophage-intact mice infected with HSV-IL-2 was blocked by depletion of FoxP3-expressing cells , while concurrent depletion of macrophages restored demyelination . In contrast , demyelination was blocked in the macrophage-depleted mice infected with wild-type HSV-1 following depletion of FoxP3-expressing cells . In macrophage-depleted HSV-IL-2-infected mice , demyelination was associated with the activity of both CD4+ and CD8+ T cells , whereas in macrophage-depleted mice infected with WT HSV-1 , demyelination was associated with CD4+ T cells . Macrophage depletion or infection with HSV-IL-2 caused an imbalance of T cells and TH1 responses as well as alterations in IL-12p35 and IL-12p40 but not other members of the IL-12 family or their receptors . Demyelination was blocked by adoptive transfer of macrophages that were infected with HSV-IL-12p70 or HSV-IL-12p40 but not by HSV-IL-12p35 . These results indicate that suppression of IL-12p70 formation by IL-2 or following macrophage depletion causes T-cell autoreactivity leading to CNS demyelination in HSV-1-infected mice .
Multiple sclerosis ( MS ) is due to degradation of the myelin sheath [1] and visual disorders due to demyelination of the optic nerve is the early sign of individuals diagnosed with MS [2 , 3] . Thus , optic neuritis can be used as an early factor for detection of MS . Both genetic and environmental factors are implicated in development of optic neuritis and MS [4–8] . Considerable evidence supports the concept that dysregulation of IL-2 plays a critical role in the development of MS [9–18] . We therefore developed a model of MS in which we combined altered expression of IL-2 with an environmental signal , HSV-1 infection . In this model , ocular infection of mice with HSV-IL-2 recombinant virus caused demyelination in the brain , spinal cord , and optic nerve [19 , 20] . Ocular infection with parental , wild-type ( WT ) viruses , or with recombinant HSV-1 expressing either IFN-γ or IL-4 , did not induce CNS demyelination . Similar results were obtained following delivery of rIL-2 protein , IL-2 DNA or IL-2 synthetic peptides prior to infection with different strains of HSV-1 [21] . Thus , the HSV-IL-2 offers a new and different small animal model for MS that integrates an environmental ( viral ) signal [19 , 20 , 22–24] . In this HSV-IL-2 model , the production of IL-2 by HSV-IL-2 is similar to the increases in IL-2 that have been observed in MS and there was increased T-cell autoreactivity leading to the CNS demyelination . The second model arose from the finding that ocular infection of macrophage-depleted mice with WT HSV-1 leads to demyelination in the absence of an external source of IL-2 . CNS demyelination did not occur in macrophage-intact mice that were ocularly infected with WT HSV-1 in the absence of an external source of IL-2 [19 , 20 , 22–24] and CNS demyelination did not occur on depletion of T cells , B cells , dendritic cells ( DCs ) , or natural killer ( NK ) cells following ocular infection with WT HSV-1 [22] . The identification of these two closely related models provided the opportunity to use a comparative analysis approach to identify the mechanisms by which macrophages may contribute to , or modulate , demyelination in the context of ocular viral infection . Macrophages are mononuclear phagocytes that play critical roles in development , tissue homeostasis and the resolution of inflammation [25] . The wide variety of functions exhibited by macrophages include cytokine secretion and antigen presentation , and cytotoxicity as well as phagocytosis . Macrophage infiltrates are an integral component of the immune defense system . They are central to innate immunity and contribute to the intersection between innate and adaptive immunity . A number of factors are known to "activate" or engage macrophages in these activities , including viral infection . Following infection of naive mice with HSV-1 , macrophages are the major infiltrates of the eye [26–28] , and play a central role in both enhancement and blocking of inflammation in the eye [29–33] . In addition to their phagocytosis , antigen presentation and cytokine production [34 , 35] , macrophages are the major source of IL-12 production [36 , 37] , a cytokine reported to be involved in stimulation of both T cells and NK cells [38–40] . IL-12 also has been shown to enhance the TH1 response [41–43] . The results of our current study demonstrate that: 1 ) In the HSV-IL-2 model , demyelination was eliminated on depletion of FoxP3-expressing cells , when macrophages were present but not when macrophages were depleted . In contrast , the macrophage-depleted HSV-1 infected mice did not show demyelination when Foxp3-expressing cells were depleted . However , in the absence of macrophages FoxP3- T cells caused demyelination; 2 ) In both models , macrophages played a critical role in prevention of autoimmunity; 3 ) Either suppression of macrophages by IL-2 or their absence caused an imbalance of T cells and the development of autoaggressive T cells; and 4 ) Adoptive transfer of macrophages over-expressing IL-12p70 or IL-12p40 , but not IL-12p35 , blocked HSV-1 induced CNS demyelination in a dose-dependent manner . Collectively , our results suggest that macrophages play a major role in protection against HSV-1-induced CNS demyelination .
IL-2 is required for the induction of Foxp3 expression and the differentiation of Treg cells in the thymus [44] . We had found previously that the induction of CNS demyelination by WT HSV-1 in macrophage-depleted mice can be blocked by depletion of FoxP3-expressing cells [22] . To determine the contribution of FoxP3 , macrophage-intact FoxP3DTR mice were depleted of Foxp3-expressing cells and ocularly infected with HSV-IL-2 or parental virus . In this model , the depletion of Foxp3-expressing cells blocked the HSV-IL-2-induced demyelination in the optic nerve ( Fig 2A , FoxP3 depleted ) , brain ( Fig 2B , FoxP3 depleted ) and spinal cord ( Fig 2C , FoxP3 depleted ) of infected mice . The protocol used to deplete the FoxP3-expressing cells did not contribute to demyelination as no demyelination was observed in the optic nerve ( Fig 2D , FoxP3 depleted ) , brain ( Fig 2E , FoxP3 depleted ) and spinal cord ( Fig 2F , FoxP3 depleted ) of the macrophage-intact , FoxP3-depleted mice that were infected with parental virus . We then investigated whether FoxP3 can protect against HSV-IL-2-induced CNS demyelination in the absence of macrophages . FoxP3DTR mice were depleted of both macrophages and FoxP3 prior to infection with HSV-IL-2 or parental virus . Surprisingly , HSV-IL-2 infection in the context of concomitant depletion of both FoxP3-expressing cells and macrophages resulted in demyelination in the optic nerve ( Fig 2G , FoxP3 and macrophage depleted ) , brain ( Fig 2H , FoxP3 and macrophage depleted ) , and spinal cord ( Fig 2I , FoxP3 and macrophage depleted ) . Moreover , the level of demyelination was similar to that observed in FoxP3-expressing FoxP3DTR mice infected with HSV-IL-2 in the absence of macrophage depletion . Replication of HSV-IL-2 in the eye of depleted mice was similar to that of parental virus . This suggests that depletion of both macrophages and FoxP3 had no direct effect on virus replication in vivo . Thus , these results indicated a complex interaction , in which the FoxP3-expressing cells contribute to blockade of CNS demyelination in HSV-IL-2 mice in the presence of macrophages , but do not block the demyelination in the absence of macrophages . Previously , we found that both CD4+ and CD8+ T cells contribute to HSV-IL-2-induced CNS demyelination in macrophage intact mice [20 , 23] whereas in HSV-1-infected macrophage-depleted mice the demyelination can be blocked by CD4+ T cells alone [22] . To directly address whether T cells contribute to HSV-1-induced demyelination in the absence of macrophages , we depleted wt mice of macrophages and CD4+/CD8+ T cells or mock depleted prior to infection with HSV-IL-2 or parental virus . The HSV-IL-2 infected mice that were depleted of both macrophages and CD4+/CD8+ T cells did not show any signs of demyelination in the optic nerve ( Fig 3A , T cells and macrophage depleted ) , brain ( Fig 3B , T cells and macrophage depleted ) or spinal cord ( Fig 3C , T cells and macrophage depleted ) . Similarly , no demyelination was detected in mice infected with the parental virus that were depleted of both macrophages and CD4+/CD8+ T cells ( Fig 3D , 3E and 3F , parental virus ) . In contrast , as we reported previously [20] , demyelination was detected in mock depleted mice infected with the HSV-IL-2 ( Fig 3G , 3H and 3I , mock depleted ) but not in mice infected with parental virus ( Fig 3J , 3K and 3L , mock depleted ) . These results suggest that , in the absence of macrophages , depletion of T cells can block CNS demyelination after infection with HSV-IL-2 or WT HSV-1 . Collectively , our results supported the concept that macrophages play a major role in protection against HSV-1-induced CNS demyelination . To begin to identify the potential mechanisms involved , we compared the changes in the levels of mRNA in the brains of mice that were infected ocularly with HSV-IL-2 , HSV-IL-4 or parental virus . We determined the mRNA levels of the IL-12 subunit genes ( IL-23p19 , IL-27p28 , IL-35EBI3 , IL-12p35 , IL-12p40 ) , IL-12 receptor genes ( IL-12rβ1 , IL-12rβ2 , IL-23r , IL-27r , gp130 ) , and markers of immune cells ( CD4 , CD8 , FoxP3 , IFN-γ , as well as CD11b and F4/80 ) . In parallel , we determined the mRNA levels for the astrocyte marker gene , glial fibrillary acidic protein ( GFAP ) , and demyelination marker genes ( NSE , S-100 , MAG , MBP , PLP , MOG ) . On day 5 PI , RT-PCR was performed on total RNA from individual brain . The levels of each mRNA relative to the baseline seen in the uninfected mouse brain is shown in Fig 4 and a summary of the differences between HSV-IL-2 infected and parental virus-infected mice is provided in Table 1 . The levels of CD11b and F4/80 mRNAs were similar in the macrophage-intact mice infected with HSV-IL-2 , parental virus or HSV-IL-4 ( Fig 4A ) . In terms of the IL-12-associated mRNAs , no differences were detected in the brains of mice infected with HSV-IL-2 , HSV-IL-4 or parental virus groups in the mRNA levels of the IL-12 receptors ( IL-12rβ1 , IL-12rβ2 ) ; IL-23r , IL-27r , and gp130 ( Fig 4B ) ; or the IL-12 subunits IL-23p19 , IL-27p28 and IL-35EBI3 mRNAs ( Fig 4C ) . However , the levels of IL-12p35 mRNA , were significantly lower than baseline in the brains of mice infected with parental or HSV-IL-4 virus , but significantly higher than baseline in the IL-HSV-IL-2 infected mice ( Fig 4C ) . Conversely , the levels of IL-12p40 mRNA were significantly higher than baseline in the parental and HSV-IL-4 infected mice but significantly lower than baseline in the HSV-IL-2 infected mice ( Fig 4C ) . We therefore extended these experiments to analysis of the levels of IL-12p35 and IL-12p40 mRNAs in macrophage-depleted mice infected with WT HSV-1 . In these mice , the levels of IL-12p35 mRNA were significantly lower than in the macrophage-intact HSV-IL-2-infected mice model but significantly higher than in the macrophage-intact mice infected with parental virus or HSV-IL-4 ( Fig 4C ) . In terms of the mRNA levels of T cell-associated molecules , the levels of FoxP3 mRNA were similar in the macrophage-intact mice infected with HSV-IL-2 , parental virus or HSV-IL-4 ( Fig 4A ) . However , we found that in the brains of macrophage-intact HSV-IL-2-infected mice the levels of CD4 mRNA were significantly lower than in mice infected with parental virus or HSV-IL-4 ( Fig 4A ) whereas the levels of CD4 mRNA in the brains of macrophage-depleted HSV-IL-2-infected mice were similar to those seen in macrophage-depleted mice infected with parental virus or HSV-IL-4 ( Fig 4A ) . The levels of CD8 mRNA in the macrophage-intact HSV-IL-2-infected mice were significantly lower than those seen in macrophage-intact mice infected with parental virus or HSV-IL-4 ( Fig 4D ) . In the macrophage-depleted mice infected with HSV-IL-2 virus , the levels of CD8 mRNA were similar to the levels in the macrophage-intact mice , and were significantly lower than those seen in macrophage-depleted mice infected with parental virus or HSV-IL-4 ( Fig 4D ) . The levels of IFN-γ mRNA in the macrophage-intact HSV-IL-2 infected mice were similar to the levels in the macrophage-depleted HSV-IL-2 infected mice but significantly lower than parental virus or HSV-IL-4 infected groups , with HSV-IL-4-infected groups having the highest level of IFN-γ expression ( Fig 4D ) . These results suggest that the presence of IL-2 has a direct effect on the levels of IL-12p35 , IL-12p40 , CD4 , CD8 and IFN-γ mRNAs , while depletion of macrophages affects the levels of IL-12p35 , CD8 , and IFN-γ mRNAs in the brain on day 5 PI . Therefore , we used the same protocol to determine the levels of these mRNAs in the spinal cord and the brain on day 10 PI in macrophage-intact mice infected with HSV-IL-2 or parental virus and macrophage-depleted mice infected with parental virus . We found that at day 10 PI , the levels of IL-12p40 mRNA in the brains of the mice infected with HSV-IL-2 or parental virus were similar and were higher than in the macrophage-depleted mice infected with parental virus ( Fig 5A , brain ) . In contrast , the levels of IL-12p40 mRNA in the spinal cords of HSV-IL-2 infected mice were significantly lower than those in mice infected with WT parental virus and was similar to that of macrophage-depleted and infected mice ( Fig 5A , spinal cord ) . IL-12p35 mRNA expression was suppressed in HSV-IL-2 infected mice brain compared with parental infected or macrophage-depleted mice ( Fig 5B , brain ) , and similar patterns were observed in spinal cord of infected mice ( Fig 5B , spinal cord ) . CD4 ( Fig 5C ) , CD8 ( Fig 5D ) , and IFN-γ ( Fig 5E ) mRNAs levels were suppressed in HSV-IL-2 infected mice brain and spinal cords compared with parental-infected or macrophage-depleted mice . In addition , the patterns of CD4 ( Fig 5C ) , CD8 ( Fig 5D ) , and IFN-γ ( Fig 5E ) mRNAs expression were similar in brain versus spinal cord of infected mice . These results for day 10 PI suggest that HSV-IL-2 has a suppressive effects on IL-12p35 , IL-12p40 , CD4 , CD8 , and IFN-γ mRNAs expression and is similar to that of their expression of day 5 PI , while macrophage depletion only affected IL-12p40 mRNA expression level . In summary , our results showed similar mRNA expression profiles for IL-12p40 , CD4 , CD8 , IFN-γ , and GFAP but not IL-12p35 in brain and spinal cord of each group . The levels of GFAP mRNA were significantly lower in the macrophage-intact HSV-IL-2-infected mice than macrophage-intact mice infected with HSV-IL-4 or parental virus ( Fig 4E , GFAP ) . The levels of GFAP mRNA were similar in the macrophage-depleted mice HSV-IL-2 infected mice to that of the macrophage-depleted mice infected with HSV-IL-4 or parental virus ( Fig 4E ) . The levels of NSE , S-100 , MAG , MBP , PLP , and MOG mRNAs were similar in the brains of the HSV-IL-2 , HSV-IL-4 and parental virus infected mice ( Fig 4E ) . In the brain of infected mice , GFAP expression was the lowest for HSV-IL-2 infected mice followed by macrophage-depleted and infected mice compared with parental virus ( Fig 5F , brain ) , while GFAP expression was similar between groups in spinal cord of infected mice ( Fig 5F , spinal cord ) . The qRT-PCR studies described above suggested that an imbalance of IL-12p35 and IL-12p40 may contribute to the HSV-IL-2-induced CNS demyelination . We found previously that both HSV-IL-2-induced demyelination and the demyelination induced by WT HSV-1 in the absence of macrophages can be blocked by either IL-12p70 DNA or HSV-IL-12p70 recombinant virus [22–24] . These data raised the possibility that the IL-12p70 arm of the macrophage response plays a key role in mitigating CNS demyelination . They suggested a hypothetical model in which suppression of macrophage IL-12p35 and IL-12p40 signaling by IL-2 in the macrophage-competent HSV-IL-2 infected mice , and the lack of IL-12p70 due to macrophage depletion play a key role in the CNS demyelination in these models of MS . To test this hypothesis , we used an adoptive transfer strategy in which bone marrow ( BM ) -derived macrophages infected with different recombinant HSV-IL-12 viruses were transferred into recipients that were subsequently ocularly infected with HSV-IL-2 . The macrophages were infected with HSV-IL-12p35 , HSV-IL-12p40 , HSV-IL-12p70 , or parental virus , or mock-infected then were injected intravenously ( IV ) into female C57BL/6 mice . Two weeks after adoptive transfer of 1 X 106 cells , the recipient mice were infected ocularly with HSV-IL-2 . Fourteen days PI , the mice were sacrificed and the optic nerve , spinal cord and brain post-fixed and stained with LFB . The presence or absence of demyelination in each tissue was determined ( Table 2 ) . We found that all of the mice that received macrophages infected with HSV-IL-12p35 or WT HSV-1 , or macrophages that were mock infected , developed demyelination in the optic nerve , brain or spinal cord . In marked contrast , most of the mice that received macrophages infected with HSV-IL-12p70 or HSV-1L-12p40 were protected from demyelination in the optic nerve , brain and spinal cord . The mice that received macrophages infected with HSV-IL-12p70 showed better protection than the mice that received macrophages infected with HSV-IL-12p40 . In addition , the adoptive transfer of macrophages infected with HSV-IL-12p70 or HSV-1L-12p40 resulted in better protection in the optic nerve and brain of the recipient mice than in the spinal cord . We then repeated the experiment using a higher dose ( 1 X 107 ) macrophages infected with HSV-IL-12p70 or HSV-IL-12p40 and , as control , transfer of 1 X 107 DCs infected with HSV-IL-12p70 . Representative photomicrographs are shown in Fig 6 and a summary of the results is provided in Table 2 . No demyelination was detected in optic nerve , brain and spinal cord sections of mice that received 1 X 107 macrophages infected with HSV-IL-12p70 ( Fig 6 , Table 2 ) or HSV-IL-12p40 , whereas demyelination occurred in the mice that received HSV-IL-12p70-infected DCs or mock-infected macrophages prior to infection ( Fig 6 , Table 2 ) . Image analysis of the stained tissue sections suggested that the extent of demyelination differed amongst the experimental groups and the CNS tissue . In those mice that received 1 X 106 macrophages , the area of demyelination in the brains of mice was significantly larger in the mice that received HSV-IL-12p35-infected macrophages or mock-infected macrophages than the area of demyelination in the mice that received HSV-IL-12p40- or HSV-IL-12p70-infected macrophages ( Fig 7 , Brain ) . Moreover , the area of demyelination in the brains of mice that received HSV-IL-12p70-infected macrophages was lower than the area of demyelination in the brains of mice that received HSV-IL-12p40-infected macrophages ( Fig 7 , Brain ) . As described above , no demyelination was detected in brains of mice that received 1 X 107 HSV-IL-12p70-infected macrophages ( Fig 7 , Brain , Arrow: no demyelination ) . In those mice that received 1 X 106 macrophages , the area of demyelination was somewhat higher in the spinal cords of mice that received HSV-IL-12p35-infected macrophages than the area of demyelination in the spinal cord of mice that received mock-infected macrophages ( Fig 7 , Spinal cord ) . The level of demyelination in the spinal cord of mice that received HSV-IL-12p40-infected macrophages were similar to the level of demyelination in the spinal cord in mice that received HSV-IL-12p70-infected macrophages ( Fig 7 , Spinal cord ) . In contrast to the spinal cords , the area of demyelination in the optic nerves was somewhat lower in recipient mice that received HSV-IL-12p35-infected macrophages than the area of demyelination in the optic nerve of mice that received mock-infected macrophages ( Fig 7 , Optic nerve ) . However , the level of demyelination in the optic nerves of mice that received HSV-IL-12p40-infected virus were similar to the levels of demyelination in the optic nerves of mice that received HSV-IL-12p70-infected macrophages ( Fig 7 , Optic nerve ) . As described above , no demyelination was detected in spinal cord and optic nerve of mice that received 1 X 107 macrophages infected with HSV-IL-12p70 virus ( Fig 7 , Spinal cord , Optic nerve , Arrow: no demyelination ) .
We reported previously that CNS demyelination occurred following ocular infection of mice with HSV-IL-2 virus , while WT viruses , HSV-IFN-γ or HSV-IL-4 did not induce CNS demyelination [19 , 20 , 22–24] . In addition , severity of CNS demyelination in HSV-IL-2-infected mice was sex-dependent [20] . Similar results were reported for MS patients [45] and EAE model of MS [46] . Thus , in this study we used female mice for all our experiments . We also have shown that macrophages , but not B cells , DCs , NK cells or T cells , mediated self-tolerance and protection against autoimmunity following ocular infection with WT HSV-1 in a manner similar to that of HSV-IL-2 in HSV-1-infected mice [22–24] . Previously we have shown that macrophages play a significant role in blocking CND demyelination in mice infected ocularly with WT HSV-1 [22] . In the current study , we show an imbalance of IL-12p35 and IL-12p40 mRNA levels in the CNS of macrophage-intact HSV-IL-2-infected mice on days 5 and 10 PI that does not occur in parental virus-infected mice . A similar imbalance was observed in macrophage-depleted mice infected with parental virus . The results suggested that this effect was specific for these IL-12 subunits , as the transcription of other members of the IL-12 family that are involved in formation of IL-23 , IL-27 , and IL-35 were not affected . These results are is similar with our previous findings that IL-12p35–/–and IL-12p40–/–mice developed CNS pathology following ocular HSV-1 infection with WT viruses and that this demyelination did not occur when each knockout strain was reconstituted with its missing gene [22] . Moreover , this pathology was not detected in HSV-1-infected IL-23p19–/–mice or in EBI3–/–mice . Our previous data and the present study suggest that both p35 and p40 subunits of IL-12 are required for protection from CNS demyelination . Although the accumulation of macrophages around demyelination plaques suggests that they may play a pathologic role [47] , it is possible that their accumulation simply reflects their phagocytic function . Our current results demonstrate that macrophages , through their production of IL-12p70 , play a central role in protection from virus-induced demyelinating immunopathology . Injection of mice with IL-12p70 DNA prevented development of CNS demyelination in macrophage-depleted mice . However , IL-12p35 or IL-12p40 DNA alone , or together had no protective effect on prevention of CNS demyelination in WT macrophage-depleted mice , indicating that control of CNS demyelination is dependent on the IL-12p70 heterodimer [22] . Similarly , we have shown previously that demyelination induced by HSV-IL-2 can be blocked by either injection of IL-12p70 DNA or a recombinant HSV-1 expressing IL-12p70 [23 , 24] . In the current study , we have shown that transfer of macrophages infected with HSV-1 recombinant virus expressing IL-12p70 or IL-12p40 , but not IL-12p35 protected HSV-IL-2 infected mice from CNS demyelination in a dose-dependent manner . Higher demyelination in macrophages infected with IL-12p35 may be due to higher expression of IL-7 by microglia and macrophages . Previously it was reported that IL-12p35 , but not IL-12p40 , subunit of IL-12p70 is involved in the induction of IL-7 in microglia and macrophages [48] . Furthermore , increase of IL-7 expression were reported for individual with MS and in EAE model of MS [49 , 50] . In contrast , DCs infected with IL-12p70 or mock-infected macrophages did not block demyelination . Previously , we have shown that both macrophages and DCs can be infected with HSV-1 but the virus does not replicate and does not increase apoptosis or cell death in infected macrophages or DCs [51] . Thus , our results suggest that macrophages carrying IL-12p70 , but not DCs or macrophages without IL-12p70 , can compensate for the suppressive effects of IL-2 on the IL-12p70 components . In the current study , we found that the effects of depletion of FoxP3-expressing cells on demyelination was highly dependent on the experimental model . In HSV-IL-2-infected mice , depletion of FoxP3-expressing cells blocked demyelination in mice , whereas depletion of macrophages as well as FoxP3-expressing cells did not block demyelination . In macrophage-depleted parental HSV-1-infected mice , demyelination was blocked following depletion of FoxP3-expressing cells . Previously , we had found that in WT HSV-1-infected mice , the absence of CD25 also blocked demyelination in macrophage-depleted mice [22] but not in the HSV-IL-2 model of demyelination [23] . These results are consistent with the reports by other investigators that depletion of Treg cells can result in enhanced immune responses against some infectious agents [52] and that Treg cells can enhance tissue damage and autoimmunity [53–58] . The reports that IL-2 can expand and induce Treg cells in vivo [59] and in vitro [60] , are in line with our present study showing that depletion of FoxP3-expressing cells blocked CNS demyelination by HSV-IL-2 and required both IL-2 and viral infection . Thus , IL-2 can modulate effector and Treg cell function in the presence of HSV-1 infection . However , the absence of demyelination in the mice that were depleted of FoxP3-expressing cells infected with HSV-IL-2 was dependent on the presence of macrophages . In the mice that were depleted of FoxP3-expressing cells and macrophages and infected with HSV-IL-2 , the depletion of both CD4 and CD8 was required for blocking demyelination independent of CD25 . In contrast , as we reported previously demyelination in the absence of macrophages in mice infected with WT virus can be blocked by the absence of FoxP3 , CD4 , or CD25 [22] . In the HSV-IL-2 infected mice that were depleted of their macrophages but not in macrophage depleted mice infected with parental virus , the level of both CD4 and CD8 expression were reduced significantly . Thus , IL-2 signaling may be involved with contraction of T-cell responses in the HSV-IL-2 infected mice . Previously , it was shown that IL-2 signaling enhances susceptibility of T cells to apoptosis [61] . In addition , IL-2 impairs T follicular helper ( Tfh ) cells [62] , while enhancing induced Treg ( iTreg ) [60] . We found previously that between days 3 and 7 PI , HSV-IL-2-infected mice exhibit a mixed TH1 + TH2 response , whereas mice infected with HSV-IFN-γ exhibit a TH1 response [19] . Similarly , we have shown that mice infected with HSV-IL-2 had an imbalance of TH1/Tc1 cytokines as compared with WT HSV-1 or recombinant viruses expressing IL-4 or IFN-γ [19] . In the present study , the levels of IFN-γ were significantly reduced in the brain of HSV-IL-2 infected mice as well as in macrophage-depleted mice . Thus , these data suggest that suppression of IL-12p70 formation by combination of IL-2 and HSV-1 infection shifts the immune response from a TH1 response , which could promote T cell autoreactivity and induction of demyelination . Surprisingly , the levels of both CD4+ and CD8+ T cells in HSV-IL-2 infected mice were reduced as compared with parental and HSV-IL-4 viruses . This reduction of T cells in the CNS of HSV-IL-2 infected mice may have been accompanied by a skewing of the population from protective T cells to pathogenic T cells . With regards to macrophage-depleted and HSV-1 infected mice , we found that depletion of macrophages affected the CD8+ but not CD4+ T cells . Thus , this imbalance of T cells also may be responsible for generation of pathogenic CD4+ T cells in macrophage-depleted mice that were infected with WT HSV-1 . Previously , we reported that depletion of macrophages enhanced infection of GFAP+ astrocytes in the spinal cords of HSV-1 infected mice as compared to mock-depleted mice [22] . Image analyses of HSV-1-infected mice revealed a significantly higher GFAP burden in the spinal cord white matter and grey matter of macrophage-depleted vs . mock-depleted mice . In contrast , in the present study we observed a significant suppression of GFAP mRNA expression in the brains of HSV-IL-2-infected mice but not in the brains of other groups . The suppression of GFAP mRNA in the HSV-IL-2-infected mice on day 5 PI suggests that the astrocytes are not activated . However , by day 10 post infection GFAP mRNA expression was significantly lower in the brains of HSV-IL-2-infected and macrophage-depleted mice compared with parental virus , while its expression in the spinal cords of all group was similar but lower than on day 5 PI . The discrepancy , between our two models of demyelination with regards to the level of GFAP mRNA expression could be due the presence of IL-2 in our HSV-IL-2 model of demyelination . Similar to the results of our current study , varicella zoster virus ( VZV ) infection has been shown to downregulate GFAP mRNA expression in vitro [63] . Additionally , it was reported that loss of astrocytes occurs before that of CNS demyelination [64] . In this report , we demonstrate that GFAP expression was significantly affected by HSV-IL-2 infection or after depletion of macrophages and infection with WT HSV-1 . This suggests that a relationship exists between astrocytes and IL-2 or astrocytes and macrophages that control CNS pathology . Our results are supported by recent evidence that interruption of astrocyte function exacerbates pathogenesis of CNS diseases [65] . We propose that suppression of GFAP on astrocytes in the absence of macrophages or in the presence of IL-2 following infection with HSV-1 affect IL-12p70 expression thus leading to autoreactivity of T cells and thus CNS demyelination . However , this suppressive effect can be reversed by IL-12p70 or IL-12p40 . In contrast to this study , IL-2 treatment has been reported to increase GFAP expression and induce inflammation and macrophage infiltration [66] . The discrepancy between our results and this study is probably due to the HSV-1 infection in our study . Despite the presence of demyelinated plaques in the CNS of HSV-IL-2 infected mice , no significant change was observed in the mRNA levels of the demyelination marker genes ( NSE , S100 , MAG , MBP , PLP , MOG ) . These results suggest that the effects of HSV-IL-2 on demyelination are not executed at the transcriptional level . Any of these genes alone or in combination have been associated with degradation of myelin by activated T cells in the CNS of infected mice . Recently , we compared MOG35–55 , MBP35–47 , and PLP190–209 induced models of EAE with our HSV-IL-2-induced MS model [67] . CNS pathology in MOG treated mice was similar to that of HSV-IL-2 treated mice but both were different from MBP or PLP injected mice . The similarity of our HSV-IL-2 model of demyelination to the MOG-induced model of demyelination may suggest that HSV-IL-2 autoreactive T cells affect the MOG component of myelin . However , the contributions of other members of myelin , such as MBP and PLP alone or in combination , to CNS demyelination in our model cannot be ruled out . In summary , our results suggest that suppression of IL-12p70 causing the FoxP3+ T cells to become autoreactive leading to demyelination of the CNS in the infected mice . However , in contrast to our study , previous study found that the absence of the Foxp3+ T cells causing autoimmunity in both humans and mice [68–70] . We feel that these contrasting results most likely stem from the infection of HSV-1 in the presence of IL-2 over-expression or in the absence of macrophages in our two models of MS .
All animal procedures were performed in strict accordance with the Association for Research in Vision and Ophthalmology Statement for the Use of Animals in Ophthalmic and Vision Research and the NIH Guide for the Care and Use of Laboratory Animals ( ISBN 0-309-05377-3 ) . Animal research protocol was approved by the Institutional Animal Care and Use Committee of Cedars-Sinai Medical Center ( Protocols #2841 and 6134 ) . Plaque-purified HSV-IL-2 , HSV-IL-4 , dLAT2903 , HSV-IL-12p35 , HSV-IL-12p40 , and HSV-IL-12p70 were grown in rabbit skin ( RS ) cell monolayers in minimal essential medium ( MEM ) containing 5% fetal calf serum ( FCS ) as we described previously [19 , 20 , 22–24 , 71 , 72] . dLAT2903 is the parental virus for HSV-IL-2 , HSV-IL-4 , HSV-IL-12p35 , and HSV-IL-12p40 and is referred to as parental virus . Female C57BL/6 mice of 6 weeks of age were purchased from the Jackson Laboratory ( Bar Harbor , ME ) . C57BL/6-FoxP3DTR mice were a gift from Alexander Y . Rudensky ( Memorial Sloan Kettering Cancer Center , New York ) and were bred in the Animal Facility at the Cedars-Sinai Medical Center and we only used female C57BL/6-FoxP3DTR mice for this study . As we described previously [19 , 20 , 22–24 , 71 , 72] , mice were infected ocularly in both eyes with 2 x 105 PFU per eye for each virus . Each virus was resuspended in 2 μl of tissue culture media and administered as an eye drop . No corneal scarification was performed prior to infection . No behavioral changes were observed between infected animals . Liposome-encapsulation of dichloromethylene diphosphonate ( Cl2MDP ) was purchased ( ClodronateLiposomes . org , Netherland ) and depletions were carried out as we described previously [22 , 73] . Briefly , mice were injected twice with 100 μl of the mixture , once intraperitoneally ( i . p . ) and once subcutaneously ( s . c . ) , on days -5 , -2 , +1 , +4 , and +7 relative to ocular infection with HSV-1 . Optic nerves , brains , and spinal cords of experimental and control mice were removed at necropsy on day 14 PI , embedded in OCT ( Tissue-Tek , Sakura Finetek , Torrance , CA ) for cryosectioning , and stored at -80°C as we described previously [20] . Transverse sections of ONs , brains , and SCs , 10 μm thick ( spaced 50μm apart ) , were prepared using a Leica CM3050S cryostat , air-dried overnight , and fixed in acetone for 3 min at 25°C [74] . The presence or absence of demyelination in infected mice was evaluated using LFB ( Sigma-Aldrich ) staining of formalin-fixed sections of CNS as we described previously [19 , 20 , 22–24] . Demyelination in each section was confirmed by monitoring adjacent sections . Female C57BL/6-FoxP3DTR mice were depleted of FoxP3 by treatment with diphtheria toxin ( DT ) ( Sigma-Aldrich , Saint Louis , MO ) as described previously [22 , 75] . Briefly , the mice were administered DT at 72 and 24 h before ocular infection , followed by four additional treatments on days +1 , +3 , +5 , and +7 PI . Efficiency of FoxP3 depletion in spleens were monitored by flow cytometry analysis before ocular infection and 7 days after ocular infection . After three depletions , more than 97% of FoxP3+ T cells were depleted . Each mouse received an i . p . injection of 100 μg of purified GK1 . 5 ( anti-CD4 ) and 100 μg of 2 . 43 ( anti-CD8 ) monoclonal antibodies ( NCCC , Minneapolis , MN ) in 100 μl of PBS , -5 and -2 days before ocular infection as we described previously [23] . The injections were then repeated on days +1 , +4 , +7 , and +10 relative to ocular infection . Control mice were depleted using an irrelevant monoclonal antibody of the same isotype . The efficiency of CD4+ and CD8+ T-cell depletion was monitored by flow cytometry of splenocytes 24 h after the second depletion and before ocular infection . After the second depletion , more than 95% of CD4+ T cells and CD8+ T cells were depleted from spleen . Brain and spinal cord from individual mice were collected on day 5 or 10 PI , immersed in RNAlater RNA stabilization reagent ( Qiagen , Valencia , CA ) and stored at -80°C until processing as we described previously [73 , 76] . The mRNA expression levels of IL-23p19 , IL-27p28 , IL-35EBI3 , IL-12p35 , IL-12p40 , GFAP , NSE , S-100 , MAG , MBP , PLP , MOG , IL-12rβ1 , IL-12rβ2 , IL-23r , IL-27r , gp130 , CD4 , FoxP3 , CD11b , F4/80 , CD8 , and IFN-γ were determined using commercially available TaqMan Gene Expression assays ( Applied Biosystems ) with optimized primers as described below . In all experiments , GAPDH was used for normalization of transcripts . In this study we looked at the mRNA expression level and not protein expression . Primer probe sets consisted of two unlabeled PCR primers and the FAM dye-labeled TaqMan MGB probe formulated into a single mixture . All cellular amplicons included an intron-exon junction to eliminate signal from genomic DNA contamination . The assays used in this study were as follows: 1 ) IL-23 p19 , ABI assay I . D . Mm00518984_m1—amplicon length = 61 bp , 2 ) IL-27 p28 , ABI assay I . D . Mm00461164_m1—amplicon length = 69 bp , 3 ) IL-35 Ebi3 , ABI assay I . D . Mm00469294_m1—amplicon length = 123 bp , 4 ) IL-12 p35 , ABI assay I . D . Mm00434165_m1—amplicon length = 68 bp , 5 ) IL-12 p40 , ABI assay I . D . Mm01288990_m1—amplicon length = 105 bp , 6 ) GFAP , ABI assay I . D . Mm01253033_m1—amplicon length = 75 bp , 7 ) NSE , ABI assay I . D . Mm00469062_m1—amplicon length = 76 bp , 8 ) S-100 , Mm00485897_m1—amplicon length = 69 bp , 8 ) MAG , ABI assay I . D . Mm00487538_m1—amplicon length = 94 bp , 9 ) MBP , ABI assay I . D Mm01262035_m1—amplicon length = 83 bp , 10 ) PLP , ABI assay I . D . Mm00456892_m1—amplicon length = 67 bp , 11 ) MOG , ABI assay I . D . Mm00447824_m1—amplicon length = 93 bp , 12 ) IL-12rβ1 , ABI assay I . D . Mm00434189_m1—amplicon length = 60 bp , 13 ) IL-12rβ2 , ABI assay I . D . Mm00434200_m1—amplicon length = 74 bp , 14 ) IL-23r , ABI assay I . D . Mm00519943_m1—amplicon length = 72 bp , 15 ) IL-27r , ABI assay I . D . Mm00497259_m1—amplicon length = 69 bp , 16 ) gp130 , ABI assay I . D . Mm00439668_m1—amplicon length = 89 bp , 17 ) CD4 , ABI assay I . D . Mm00442754_m1—amplicon length = 78 bp , 18 ) FoxP3 , ABI assay I . D . Mm00475164_m1—amplicon length = 80 bp , 19 ) CD11b , ABI assay I . D . Mm00434455_m1—amplicon length = 69 bp , 20 ) F4/80 , ABI assay I . D . Mm00802529_m1—amplicon length = 92 bp , 21 ) CD8 , ABI assay I . D . Mm01182108_m1—amplicon length = 67 bp , 22 ) IFN-γ , ABI assay I . D . Mm00801778_m1—amplicon length = 101 bp , and 23 ) GAPDH , ABI assay I . D . Mm999999 . 15_G1 –amplicon length = 107 bp . Additionally , a custom-made primer and probe set was used for LAT as follows: forward primer , 5'-GGGTGGGCTCGTGTTACAG-3'; reverse primer , 5'-GGACGGGTAAGTAACAGAGTCTCTA-3'; and probe , 5'- FAM-ACACCAGCCCGTTCTTT-3'–Amplicon Length = 81 bp . Quantitative real-time PCR ( qRT-PCR ) was performed using an ABI ViiA 7 Sequence Detection System ( Applied Biosystems ) in 384-well plates as we described previously [73 , 76] . Six-week-old female C57BL/6 mice were used as a source of bone marrow ( BM ) for the generation of mouse DCs and macrophages in culture as we described previously [51] . Briefly , BM cells were isolated by flushing femurs and tibiae with PBS . Pelleted cells were resuspended briefly in water to lyse red blood cells and stabilized by adding complete medium ( RPMI 1640 , 10% fetal bovine serum , 100 U/ml penicillin , 100 μg/ml streptomycin , 2 mM L-glutamine ) . The cells were centrifuged and resuspended in complete medium supplemented with murine CSF1 ( 100 ng/ml; Peprotech , Rocky Hill , NJ ) to grow macrophages , while to grow DCs the media was supplemented with murine GM-CSF ( 100 ng/ml; Peprotech ) . The cells were plated in non-tissue culture plastic petri dishes ( 1 bone per 10 cm dish ) for 5 d at 37°C with CO2 . After 5 d , the media was removed , and the adherent cells recovered by incubating the cells for 5 min at 37°C with Versene ( Invitrogen , San Diego , CA ) . Cells were washed , counted , and plated onto tissue culture dishes for use the following day . Monolayers of macrophages were infected with 1 PFU/cell of dLAT2903 , HSV-IL-12p35 , HSV-IL-12p40 , or HSV-IL-12p70 , and monolayers of DCs were infected with 1 PFU/cell of HSV-IL-12p70 . One hour after infection at 37°C , virus was removed and the infected cells were washed three times and fresh media was added to each well . The monolayers at 24 h PI were harvested , washed , and counted for subsequent adoptive transfer experiments . Each recipient female C57BL/6 mouse was injected once intravenously ( i . v . ) with 1 X 106 or 1 X 107 infected macrophages in 100 μl of MEM . Similarly control mice received uninfected macrophages or infected DCs . Recipient mice were ocularly infected two weeks after transfer with HSV-IL-2 virus . The numbers of plaques and size of plaques on multiple LFB stained fields were evaluated in a blind fashion for each treatment group by inspection of serial sections of CNS tissues . The amount of myelin loss in the stained sections of brains , SCs and ONs was measured using NIH Image J software as described previously [67] . The areas of demyelination ( clear-white ) to normal tissue ( blue ) were quantified using 75 random sections from the brain and SCs or 30 sections from ONs of each animal . Demyelination in each section was confirmed by monitoring adjacent sections . The percentage of myelin loss was calculated by dividing the lesion size by the total area for each section . Level of demyelination in experimental and control groups were compared using Fisher’s exact tests . Student’s t test was performed for comparison of means of differences using Instat ( GraphPad , San Diego ) . Results were considered statistically significant when the "P" value was <0 . 05 . | Several mouse models of multiple sclerosis ( MS ) are now available . We have established two new mouse models . In the first model , ocular infection of different strains of mice with HSV-IL-2 recombinant virus causes CNS demyelination . In the second model , CNS demyelination was induced by different strains of wild type HSV-1 in the absence of macrophages . In the present study , we found differences in T-cell reactivity in the two models . However , both models exhibited an imbalance in IL-12p35 and IL-12p40 . The requirement for formation of the IL-12p70 dimer in prevention of demyelination was supported by adoptive transfer experiments . These results suggest a pathological role for macrophages in these models of virus-induced MS in which suppression of IL-12p70 formation by IL-2 or following macrophage depletion causes T-cell autoreactivity leading to CNS demyelination . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"multiple",
"sclerosis",
"nervous",
"system",
"neurodegenerative",
"diseases",
"immunology",
"neuroscience",
"demyelinating",
"disorders",
"clinical",
"medicine",
"optic",
"nerve",
"eye",
"diseases",
"adoptive",
"transfer",
"spinal",
"cord",
"white",
"blood",
"cells",
"animal",
"cells",
"t",
"cells",
"neuroanatomy",
"cell",
"biology",
"anatomy",
"central",
"nervous",
"system",
"clinical",
"immunology",
"neurology",
"ophthalmology",
"autoimmune",
"diseases",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"ocular",
"system",
"macrophages"
] | 2017 | Suppression of IL-12p70 formation by IL-2 or following macrophage depletion causes T-cell autoreactivity leading to CNS demyelination in HSV-1-infected mice |
Alveolar echinococcosis ( AE ) is a lethal zoonosis caused by the metacestode larva of the tapeworm Echinococcus multilocularis . The infection is characterized by tumour-like growth of the metacestode within the host liver , leading to extensive fibrosis and organ-failure . The molecular mechanisms of parasite organ tropism towards the liver and influences of liver cytokines and hormones on parasite development are little studied to date . We show that the E . multilocularis larval stage expresses three members of the fibroblast growth factor ( FGF ) receptor family with homology to human FGF receptors . Using the Xenopus expression system we demonstrate that all three Echinococcus FGF receptors are activated in response to human acidic and basic FGF , which are present in the liver . In all three cases , activation could be prevented by addition of the tyrosine kinase ( TK ) inhibitor BIBF 1120 , which is used to treat human cancer . At physiological concentrations , acidic and basic FGF significantly stimulated the formation of metacestode vesicles from parasite stem cells in vitro and supported metacestode growth . Furthermore , the parasite’s mitogen activated protein kinase signalling system was stimulated upon addition of human FGF . The survival of metacestode vesicles and parasite stem cells were drastically affected in vitro in the presence of BIBF 1120 . Our data indicate that mammalian FGF , which is present in the liver and upregulated during fibrosis , supports the establishment of the Echinococcus metacestode during AE by acting on an evolutionarily conserved parasite FGF signalling system . These data are valuable for understanding molecular mechanisms of organ tropism and host-parasite interaction in AE . Furthermore , our data indicate that the parasite’s FGF signalling systems are promising targets for the development of novel drugs against AE .
The flatworm parasite E . multilocularis ( fox-tapeworm ) is the causative agent of alveolar echinococcosis ( AE ) , one of the most dangerous zoonoses of the Northern hemisphere . Infections of intermediate hosts ( rodents , humans ) are initiated by oral uptake of infectious eggs which contain the parasite’s oncosphere larval stage [1 , 2] . After hatching in the host intestine and penetration of the intestinal wall the oncosphere gains access to the liver , where it undergoes a metamorphotic transition towards the metacestode larval stage [3] . The E . multilocularis metacestode consists of numerous vesicles which grow infiltratively , like a malignant tumor , into the liver tissue [1–3] . Due to the unrestricted growth of the metacestode , blood vessels and bile ducts of the liver tissue of the intermediate host are obstructed , eventually leading to organ failure [1] . Another hallmark of AE is extensive liver fibrosis which can lead to a complete disappearance of the liver parenchyma , and which most probably involves the activation of hepatic stellate cells during chronic infection [4 , 5] . Surgical removal of the parasite tissue , the only possible cure , is not feasible in the majority of patients leaving benzimidazole-based chemotherapy as the only treatment option . However , benzimidazoles act parasitostatic only and have to be given for prolonged periods of time ( often life-long ) , underscoring the need for novel treatment options against AE [1] . We previously established that E . multilocularis development and larval growth is exclusively driven by a population of somatic stem cells , the germinative cells , which are the only mitotically active cells of the parasite and which give rise to all differentiated cells [6] . Using in vitro cultivation systems for metacestode vesicles and germinative cells [7–10] , we also demonstrated that host insulin fosters parasite development by acting on evolutionarily conserved receptor kinases of the insulin receptor family that are expressed by the metacestode [11] . Evidence has also been obtained that host epidermal growth factor ( EGF ) stimulates Echinococcus germinative cell proliferation , most probably by acting on parasite receptor tyrosine kinases ( RTK ) of the EGF receptor family [12 , 13] . These studies indicate that the interaction of host-derived hormones and cytokines with corresponding receptors of evolutionarily conserved signalling pathways that are expressed by the parasite may play an important role in AE host-parasite interaction . Although the E . multilocularis genome project already indicated that the parasite expresses receptor TK of the fibroblast growth factor ( FGF ) receptor family in addition to insulin- and EGF-receptors [14] , no studies concerning the effects of host FGF and their possible interaction with parasite FGF receptors have been carried out to date . FGFs are an ancient group of polypeptide cytokines that are present in diploblastic animals , in deuterostomes and , among protostomes , only in ecdysozoa ( with some distantly related members in lophotrochozoa ) [15 , 16] . Humans express 22 different FGFs of which several ( FGF11 –FGF14 ) are not secreted and act independently of FGF receptors in an intracrine modus only [15] . The remaining FGFs act in a paracrine fashion and are typically released via N-terminal signal peptides . Notable exceptions are the prototypic FGF1 ( acidic FGF ) and FGF2 ( basic FGF ) which are ubiquitously expressed in human tissues , are the most active members of the FGF family , and are released in a signal peptide-independent manner [15] . FGFs have a key role in metazoan embryonic development and , in adults , are typically involved in regeneration processes ( angiogenesis , wound healing , liver regeneration , regeneration of nervous tissue ) [15] . In the liver , particularly FGF1 but also FGF2 are present as proteins in significant amounts [17] , are crucially involved in tissue regeneration upon damage [18 , 19] , and are also upregulated and released during fibrosis [20] . Secreted FGFs act through surface receptor TK of the FGF receptor family , of which four isoforms , Fgfr1-Fgfr4 , are expressed by humans [15] . The mammalian FGF receptors comprise an extracellular ligand-binding domain made up of three immunoglobulin ( Ig ) -like domains , a transmembrane domain , and a split intracellular kinase domain . FGF binding to the cognate FGF receptors typically results in receptor dimerization , transphosphorylation and subsequent activation of downstream signalling pathways such as the Ras-Raf-MAPK ( mitogen-activated protein kinase ) cascade or the PI3K/Akt pathway [15] . FGF signalling pathways have , in part , already been studied in flatworms . In the free-living planarian species Dugesia japonica , two members of the FGFR TK are expressed of which DjFGFR1 exhibits three immunoglobulin-like domains in the extracellular region whereas DjFGFR2 only contains two such domains [21] . Both receptors are expressed by X-ray sensitive planarian stem cells ( neoblasts ) and in cephalic ganglia and an important role of these FGFRs in planarian brain formation has been suggested [21 , 22] . Furthermore , similar FGF receptors were also detected in stem cells of the planarian Schmidtea mediterranea [23] . In the genome of the flatworm parasite species Schistosoma mansoni , two FGFR-encoding genes were identified of which fgfrA codes for a predicted protein with two extracellular immunoglobulin domains and a split TK domain whereas the fgfrB gene product only comprises one immunoglobulin domain in the extracellular region [24] . Expression of fgfrA and fgfrB in neoblast-like somatic stem cells has been shown and evidence was obtained for an important role of both receptors in schistosome stem cell maintenance [25–27] . Hahnel et al . [24] also demonstrated that both receptors are enzymatically active , are expressed in the gonads of schistosomes , and are upregulated following pairing , indicating a role in parasite fertility . Interestingly , these authors also showed that treatment of adult schistosomes with FGFR inhibitors leads to a reduction of somatic neoblast-like stem cells in both genders [24] . In the present work we provide a detailed analysis of three FGFRs in the cestode E . multilocularis and show that the expression patterns of these receptors differ from those in planaria and schistosomes . We also demonstrate that all three Echinococcus FGFRs are activated in response to human FGFs and that host FGF stimulates parasite development in vitro . Finally , we also show that inhibition of FGF signalling in Echinococcus larvae drastically reduces parasite development and survival .
FGF stimulation and inhibitor experiments were performed with the natural E . multilocularis isolate H95 [14] . Whole mount in situ hybridization was carried out using isolate GH09 which , in contrast to H95 , is still capable of producing brood capsules and protoscoleces in vitro [14] . All isolates were continuously passaged in mongolian jirds ( Meriones unguiculatus ) as previously described [9] . The generation of metacestode vesicles and axenic cultivation of mature vesicles was performed essentially as previously described [7 , 9] with media changes usually every three days . Primary cell cultures were isolated from mature vesicles of isolate H95 and propagated in vitro essentially as previously described [8–10] with media changes every three days unless indicated otherwise . For FGF stimulation assays , 10 nM or 100 nM of recombinant human acidic FGF ( FGF1 ) or basic FGF ( FGF2 ) ( both from ImmunoTools GmbH , Friesoythe , Germany ) were freshly added to parasite cultures during medium changes . In the case of primary cells , cultivation was usually performed in cMEM medium which is host hepatocyte-conditioned DMEM ( prepared as described in [10] ) . For inhibitor studies , specific concentrations of BIBF 1120 ( Selleck Chemicals LLC , Houston , TX , USA ) were added to parasite cultures as indicated and as negative control DMSO ( 0 . 1% ) was used . The formation of mature metacestode vesicles from primary cells and measurement of metacestode vesicles size was performed essentially as previously described [11] . RNA isolation from in vitro cultivated axenic metacestode vesicles ( isolate H95 ) , protoscoleces ( isolate GH09 ) , and primary cells ( H95 , GH09 ) was performed using a Trizol ( 5Prime , Hamburg , Germany ) -based method as previously described [11] . For reverse transcription , 2 μg total RNA was used and cDNA synthesis was performed using oligonucleotide CD3-RT ( 5’-ATC TCT TGA AAG GAT CCT GCA GGT26 V-3’ ) . PCR products were cloned using the PCR cloning Kit ( QIAGEN , Hilden , Germany ) or the TOPO XL cloning Kit ( invitrogen ) , and sequenced employing an ABI prism 377 DNA sequencer ( Perkin-Elmer ) . The full-length emfr1 cDNA was cloned using as starting material the partial sequence of a cDNA of the closely related cestode E . granulosus , which encoded a FGFR-like TK domain but which lacked the coding regions for transmembrane and extracellular parts [28] . Using primers directed against the E . granulosus sequence ( 5’-CTA CGC GTG CGT TTT CTG ATG-3’for first PCR; 5’-CCC TCT GAT CCA ACC TAC GAG-3’for nested PCR ) , the 3’ end of the corresponding E . multilocularis cDNA was subsequently PCR amplified from a metacestode ( isolate H95 ) cDNA preparation using primers CD3 and CD3nest as previously described [29] . 5’-RACE was performed using the SMART RACE cDNA amplification kit ( Clontech ) according to the manufacturer’s instructions using primers 5’-ACC GTA TTT GGG TTG TGG TCG-3’ ( first PCR ) and 5’-GAA CAG GCA GAT CGG CAG-3’ ( touchdown PCR ) as previously described [30] . The presence of an in frame TAA stop codon 110 bp upstream of the emfr1 ATG start codon indicated that the correct 5’ end had been identified . In a final step , the entire emfr1 cDNA was PCR amplified from metacestode cDNA using primers 5’-GAC ACA TCT CCT TGG CCG-3’ and 5’-GCG AGT TGA TAC TTT ATG AGA G-3’ and cloned using the TOPO XL PCR cloning kit ( Invitrogen ) . The sequence is available in the GenBankTM , EMBL , and DDJB databases under the accession number LT599044 . For emfr2 cloning we first identified by BLAST analyses on the published E . multilocularis genome sequence [14] a reading frame encoding a FGFR-like TKD annotated as EmuJ_000196200 . Transcriptome analyses [14] and 5’_RACE experiments , however , indicated that there is actually read-through transcription between gene models EmuJ_000196300 and EmuJ_000196200 . We thus designed primers 5’-ATG TGT CTC CGA GCT CTC TG-3’ , binding to the 5’ end regions of gene model EmuJ_000196300 , and primer 5’-TTA CTC GCT CGA TCG TGG GG-3’ , binding to the reading frame 3’ end of gene model EmuJ_000196200 , to PCR amplify the entire reading frame from metacestode cDNA . The resulting PCR fragment was subsequently cloned using the TOPO XL cloning kit ( Invitrogen ) and fully sequenced . The sequence is available in the GenBankTM , EMBL , and DDJB databases under the accession number LT599045 . For emfr3 cloning and sequencing we used primers directed against the CDS 5’ end ( 5’-ATG GCA CCT AAG GTT GTG TCA GGA-3’ ) and 3’ end ( 5’-GCA GAT GAG TAA GAA ACC CTC-3’ ) of gene model EmuJ_000893600 [14] for direct PCR amplification of the reading frame from metacestode cDNA . The resulting PCR fragment was subsequently cloned using the TOPO XL cloning kit ( Invitrogen ) and sequenced . The sequence is available in the GenBankTM , EMBL , and DDJB databases under the accession number LT599046 . Proliferation of E . multilocularis metacestode vesicles and primary cells was assessed by a bromodesoxyuridine ( BrdU ) -based method . Axenically cultivated metacestode vesicles ( 2–4 mm in diameter ) were manually picked and incubated in 12-well plates ( Greiner BioOne , Kremsmünster , Germany; 8 vesicles per well ) in DMEM medium without serum for 2 days . Freshly isolated primary cells were plated on 12-well plates and grown for 2 days under axenic conditions in conditioned DMEM ( cMEM ) medium with serum [8] . BrdU ( SigmaAldrich , taufkirchen , Germany ) as well as recombinant human FGF1 and FGF2 were added at 1mM ( BrdU ) and 100 nM or 10 nM ( FGF1 , FGF2 ) final concentrations as indicated . Cultures were incubated for 48 h at 37°C under 5% CO2 for metacestode vesicles or under nitrogen atmosphere [7 , 8] in the case of primary cells . Samples were analysed in duplicates in three independent experiments . As controls , metacestode vesicles or primary cells were incubated in either DMEM without serum or conditioned DMEM , without addition of FGFs . Primary cells and metacestode vesicles were then isolated for genomic DNA analysis . In detail , vesicles and primary cells were first washed with 1xPBS , pelleted , and subsequently transferred to lysis buffer ( 100 mM NaCl , 10 mM Tris-HCl , pH 8 . 0; 50 mM EDTA , pH 8 . 0 , 0 , 5% SDS ) supplemented with 20 μg/ml RNAse A and 0 , 1 mg/ml proteinase K . Samples were then incubated at 50°C for 4 h under constant shaking for complete lysis . DNA was isolated by two rounds of phenol/chlorophorm extraction ( 1 vol of phenol/chlorophorm/isoamylalcohol 25:24:1 ) . DNA was then precipitated with 2 vol of 96% ethanol and 0 , 1 vol of LiCl ( pH 4 , 5 ) after overnight incubation at -20°C and centrifugation at 20 . 000 rcf for 30 min at 4°C and washed with 70% ethanol . The pellet was then air dried for 15 min an resuspended in 1 x TE buffer ( 10 mM Tris , 1 mM EDTA , pH 8 , 0 ) . The DNA was then prepared for coating onto a 96-well plate ( 96 well optical bottom plates , Nunc , Langenselbold , Germany ) . To this end , 5 μg of DNA were combined with 1 vol of Reacti-Bind DNA Coating solution ( Pierce Biotechnology , Rockford , IL , USA ) and mixed for 10 min . The DNA mix was then added to the microplates in duplicates and incubated overnight at room temperature with gentle agitation . The TE/Reacti-Bind DNA coating solution mix served as a negative control . Unbound DNA was removed by washing three times with 1xPBS . After blocking with 5% nonfat dry milk in 1xPBS for 1 h at room temperature and extensive washing with 1xPBS , 100 μl of anti-BrdU-POD ( Cell Proliferation ELISA , BrdU; Roche Applied Science , Mannheim , Germany ) was added and incubated for 90 min at room temperature . After incubation , microplates were washed three times with 1xPBS buffer before substrate solution ( Cell Proliferation ELISA , BrdU; Roche Applied Science , Mannheim , Germany ) was added and the wells were incubated for 60 min . Stop-solution ( 25 μl of 1 M H2SO4 ) was added and absorbance of the samples was measured using an ELISA reader at 450 nm . Coding sequences of FGF receptors from E . multilocularis were amplified by RT-PCR and cloned into the vectors pDrive ( Qiagen ) or pJET1 . 2 ( Thermo Fisher ) . In the case of emfr2 , the full length coding sequence was amplified using primers 5’-ATG TGT CTC CGA GCT CTC TG-3’ ( forward ) and 5’-TTA CTC GCT CGA TCG TGG GG-3’ ( reverse ) , whereas partial coding sequences were amplified for emfr1 ( using forward primer 5’-GCA GTG GGC GTC TTC TTT CAC-3’ and reverse primer 5’-GTA AAT GTG GGC CGA CAC TCA G-3’ ) and for emfr3 ( using forward primer 5’-TTG CCC AGT CAT CCG CTA CAA G-3’ and reverse primer 5’-GCA AGC GGT CAT GAG GCT GTA G-3’ ) . The recombinant plasmids were used for in vitro transcription of digoxigenin-labelled RNA probes as previously described [6] . These probes were used for fluorescent WMISH of in vitro cultured E . multilocularis larvae as described in [6] . Control WMISH experiments using the corresponding sense probes were always negative . For expression in the Xenopus system , the emfr1 , emfr2 , and emfr3 coding sequences without predicted signal peptide information were cloned into the pSecTag2/Hygro expression system ( ThermoFisher Scientific , Germany ) leading to an in frame fusion of the Igk leader sequence ( provided by the vector system ) and the E . multilocularis FGF receptor sequences under control of the T7 promoter . Capped messenger RNAs ( cRNA ) encoding EmFR1 , EmFR2 and EmFR3 were then synthesized in vitro using the T7 mMessage mMachine Kit ( Ambion , USA ) . Microinjection of EmFGFR cRNAs ( 60 ng in 60 μl ) was performed in stage VI Xenopus laevis oocytes according to the procedure previously described [31] . Following 48h of receptor expression , human FGF1 or FGF2 ( R & D systems , UK ) were added to the extracellular medium at the final concentration of 10 nM . cRNA of Pleurodeles FGFR1 identified as homologous to human receptor [32] was a gift of Shi D . L . ( CNRS UMR 722 , Paris VI ) and was used as a positive control . In some experiments , BIBF1120 ( stock solution 10mM in DMSO , Selleck Chemicals LLC ) was added ( 0 . 1 to 20 μM final concentration ) 1 h before the addition of 10 nM FGF1 or FGF2 on EmFR1 , EmFR2 , EmFR3 and Pleurodeles FGFR1 expressing oocytes . Following 15 h of FGF1 or FGF2 stimulation , oocytes were analyzed for their state of progression in the cell cycle . The detection of a white spot at the animal pole of the oocyte attested to G2/M transition and GVBD . Non-injected oocytes treated with progesterone ( 10 μM ) were used as positive controls of GVBD . For each assay , sets of 20–30 oocytes removed from 3 different animals were used . Dead kinase ( TK- ) receptors were obtained by site-directed mutagenesis of the EmFR1 , EmFR2 and EmFR3 constructs . The active DFG sites present in EmFR1 ( D442FG ) , EmFR2 ( D647FG ) and EmFR3 ( D701FG ) were replaced by an inactivating motif ( DNA ) as described in [31] . For western blot analysis , oocytes were lysed in buffer A ( 50 mM Hepes pH 7 . 4 , 500 mM NaCl , 0 . 05% SDS , 5 mM MgCl2 , 1 mg ml−1 bovine serum albumin , 10 μg ml−1 leupeptin , 10 μg ml−1 aprotinin , 10 μg ml−1 soybean trypsin inhibitor , 10 μg ml−1 benzamidine , 1 mM PMSF , 1 mM sodium vanadate ) and centrifuged at 4°C for 15 min at 10 , 000 g . Membrane pellets were resuspended and incubated for 15 min at 4°C in buffer A supplemented with 1% Triton X-100 and then centrifuged under the same conditions . Supernatants were analyzed by SDS-PAGE . Proteins were transferred to a Hybond ECL membrane ( Amersham Biosciences , France ) . Membranes were incubated with anti-myc ( 1/50 000 , Invitrogen France ) or anti-PTyr ( 1/8000 , BD Biosciences , France ) antibodies and secondary anti-mouse antibodies ( 1/50 000 , Biorad , France ) . Signals were detected by the ECL advance Western blotting detection kit ( Amersham Biosciences , France ) Axenically cultivated metacestode vesicles of about 0 . 5 cm in diameter were incubated in DMEM medium with or without 10% FCS for 4 days . Vesicles cultivated without FCS were subsequently incubated with 10 nM FGF1 ( aFGF ) or 10 nM FGF2 ( bFGF ) for 30 sec , 60 sec or 60 min . Immediately after stimulation , vesicles were harvested , cut by a scalpel to remove cyst fluid and then subjected to protein isolation as described previously [33] . Isolated protein lysates were then separated on a 12% acrylamide gel and analysed by Western blotting using a polyclonal anti-Erk1/Erk2 antibody ( ThermoFisher Scientific; #61–7400 ) , recognizing Erk-like MAP kinases in phosphorylated and non-phosphorylated form , as well as a polyclonal antibody against phospho-Erk1/Erk2 ( ThermoFisher Scientific; #44-680G ) , specifically directed against the double-phosphorylated ( activated ) form of Erk1/Erk2 ( Thr-185 , Tyr-187 ) . We had previously shown that these antibodies also recognize the Erk-like MAP kinase EmMPK1 of E . multilocularis in phosphorylated and non-phosphorylated form [33] . As secondary antibody , a peroxidase-conjugated anti-mouse IgG antibody ( Dianova , Hamburg , Germany ) was used . In inhibitor experiments , axenically cultivated metacestode vesicles were incubated with either 5 μM or 10 μM BIBF 1120 for 30 min and then processed essentially as described above . qRT-PCR experiments have been performed as previously described by del Puerto et al . [34] with several modifications . Briefly , RNA samples were prepared with TRI-reagent and the DirectzolTM RNA Mini Prep Kit ( Zymo Research , USA ) . Prior to RNA-isolation , in vitro cultured metacestode vesicles were deprived of stem cells by treatment with hydroxyurea ( HU ) and the Polo-like kinase inhibitor BI 2536 essentially as previously described by Koziol et al . [6] and Schubert et al . [35] , respectively . Reverse transcription was performed with Omniscript RT Kit ( Qiagen , Germany ) from 500 ng of RNA using oligo dT23 as primers according to the manufacturer’s instructions . qRT-PCRs were then performed using primers Forward F_emfr1_qPCR ( 5’-CCG TAT GAA GGG AAA TGG TCG TGT T-3’ ) and Reverse R_emfr1_qPCR ( 5’-TGG TGA ATC GCC AAG GCT GAA A-3’ ) for emfr1 , Forward F_emfr2_qPCR ( 5’-GGG AAT TTC CAA GGT CAT CAG GGA C-3’ ) and Reverse R_emfr2_qPCR ( 5’-ATC GTG GGG GCA CAA CAT AAT TGC-3’ ) for emfr2 , as well as Forward F_emfr3_qPCR ( 5’-GTC TAC CTT GAG GAA ATT GCT GTG GTC-3’ ) and Reverse R_emfr3_qPCR ( 5’-CGT GAG GAA TGA CGC AGG C-3’ ) for emfr3 . As a control gene for normalization , the constitutively expressed gene elp was used with primers as described previously [34] . qRT-PCR conditions and primer amplification efficiency control experiments were performed essentially as previously described [34] . The relative expression of emfr genes in each sample was estimated using the delt-delta-Ct method [36] , normalising with elp expression levels using the StepOne software ( Thermo Fisher ) . All experiments were carried out as technical triplicates . Amino acid comparisons were performed using BLAST on the nr-aa and swissprot database collections available under ( https://www . genome . jp/ ) . Genomic analyses and BLAST searches against the E . multilocularis genome [14] were done using resources at ( https://parasite . wormbase . org/index . html ) . CLUSTAL W alignments were generated using MegAlign software ( DNASTAR Version 12 . 0 . 0 ) applying the BLOSUM62 matrix . Domain predictions were carried out using the simple modular architecture research tool ( SMART ) available under ( http://smart . embl-heidelberg . de/ ) as well as PROSITE scans available under ( https://prosite . expasy . org/scanprosite/ ) . Two-tailed , unpaired student’s T-tests were performed for statistical analyses ( GraphPad Prism , version 4 ) . Error bars represent standard error of the mean . Differences were considered significant for p-values below 0 . 05 ( indicated by * ) . All experiments were carried out in accordance with European and German regulations on the protection of animals ( Tierschutzgesetz ) . Ethical approval of the study was obtained from the local ethics committee of the government of Lower Franconia ( permit no . 55 . 2 DMS 2532-2-354 ) .
By cDNA library screening and mining of the available E . multilocularis genome sequence we identified a total of three Echinococcus genes encoding members of the FGFR family of RTK . A partial cDNA for a gene encoding a TK with homology to FGFRs was previously cloned for E . granulosus [28] and by RT-PCR amplification of metacestode cDNA as well as 5’-RACE , the entire cDNA of the E . multilocularis ortholog , designated emfr1 ( E . multilocularis fibroblast growth factor receptor 1 ) , was subsequently cloned . As shown in Fig 1 , the encoded protein , EmFR1 , contained an N-terminal export directing signal peptide , followed by one single Ig-like domain , a transmembrane region , and an intracellular TK domain ( Figs 1 and S1 ) . In the recently released E . multilocularis genome information [14] , this gene was correctly predicted on the basis of genome and transcriptome data ( E . multilocularis gene designation: EmuJ_000833200 ) . In the upstream genomic regions of emfr1 , no information encoding potential Ig-like domains was identified which , together with the presence of a signal peptide sequence upstream of the single Ig-like domain , indicated that EmFR1 indeed contained only one single Ig-like domain . Amino acid sequence alignments indicated that the kinase domain of EmFR1 contains all residues critical for enzymatic activity at the corresponding positions ( S2 Fig ) and SWISS-PROT database searches revealed highest similarity between the EmFR1 kinase domain and that of human FGFR4 ( 42% identical aa; 59% similar aa ) . A second gene encoding a TK with significant homology to known FGFRs was identified on the available E . multilocularis genome sequence [14] under the annotation EmuJ_000196200 . The amino acid sequence of the predicted protein only contained an intracellular TK domain , a transmembrane region , and one extracellular Ig-like domain , but no putative signal peptide . We therefore carried out 5’-RACE analyses on a cDNA preparation deriving from protoscolex RNA and identified the remaining 5’ portion of the cDNA , which contained one additional Ig-like domain and a predicted signal peptide . In the genome annotation , these remaining parts were wrongly annotated as a separate gene under the designation EmuJ_000196300 . Hence , the second FGFR encoding gene of E . multilocularis , emfr2 , encoding the protein EmFR2 , actually comprises the gene models EmuJ_000196300 and EmuJ_000196200 of the genome sequence . EmFR2 thus comprises a signal peptide , two extracellular Ig-like domains , a transmembrane region , and an intracellular TKD ( Figs 1 and S1 ) . The TKD contained all residues critical for TK activity ( S2 Fig ) and , in SWISS-PROT BLASTP analyses , showed highest similarity to two FGF receptor kinases of the flatworm Dugesia japonica ( 45% identical , 65% similar residues ) , to the S . mansoni receptor FGFRB ( 55% , 68% ) and to human FGFR3 ( 48% , 62% ) . The third FGF receptor encoding gene of E . multilocularis , emfr3 , was identified under the designation EmuJ_000893600 and was originally listed as an ortholog of the tyrosine protein kinase Fes:Fps [14] . However , unlike the Fes:Fps kinase which contains FCH and SH2 domains , the EmuJ_000893600 gene product , EmFR3 , comprised an N-terminal signal peptide , two extracellular Ig-like domains , a transmembrane region , and an intracellular TKD ( Figs 1 and S1 ) , in which 22 of 30 highly conserved residues of TK are present at the corresponding position ( S2 Fig ) . Furthermore in SWISS-PROT BLASTP analyses the EmFR3 TKD displayed highest similarity to several vertebrate FGF receptors and to human FGFR2 ( 32% , 47% ) . We thus concluded that EmuJ_000893600 actually encoded a third Echinococcus FGF receptor TK . Apart from emfr1 , emfr2 , and emfr3 , no further genes were identified in the E . multilocularis genome which displayed clear homology to known FGFR TKDs and which contained characteristic IG domains in the extracellular portions . In vertebrates , structural homology has been described between the TKDs of the receptor families of FGF receptors , the vascular endothelial growth factor ( VEGF ) receptors , and the platelet-derived growth factor ( PDGF ) receptors , which also contain varying number of Ig domains in the extracellular parts [37] . Furthermore , VEGF receptor-like molecules have also been described in invertebrates such as Hydra [37] . We therefore carried out additional BLASTP searches on the E . multilocularis genome using human VEGF- and PDGF-receptors as queries , but only obtained significant hits with the above mentioned TKDs of EmFR1 , EmFR2 , and EmFR3 . These data indicated that members of the VEGF- and PDGF-receptor families are absent in Echinococcus , as has already been described for the closely related schistosomes [24] . Genes encoding canonical FGF ligands have so far neither been identified in genome projects of free-living flatworms [38] , nor in those of trematodes [39] or cestodes [14] . In vertebrates [15] as well as several invertebrate phyla [40–42] , however , canonical FGF ligands are clearly expressed . To investigate the situation more closely , we carried out BLASTP and TBLASTN analyses against the predicted proteins and E . multilocularis contig information , respectively , using FGF ligand sequences of insect , nematode , and cnidarian origin as queries . The product of only one E . multilocularis gene , EmuJ_000840500 ( annotated as ‘conserved hypothetical protein’ ) showed certain similarity to these FGF ligands and according to SMART protein domain analyses could contain a FGF-ligand domain between amino acids 166 and 258 , although this prediction was clearly below the prediction threshold and of low probability ( E-value: 817 ) . No export directing signal peptide was predicted for the EmuJ_000840500 protein , as would be typical for FGF ligands . Furthermore , although EmuJ_000840500 had clear orthologs in the cestodes Taenia solium ( TsM_000953800 ) and Hymenolepis microstoma ( HmN_000558500 ) , none of these gene models had any prediction of an FGF-ligand domain in SMART analyses ( nor predicted signal peptides ) . We thus considered it highly unlikely that EmuJ_000840500 encodes a so far not identified flatworm FGF-ligand . Taken together , our analyses indicated that E . multilocularis contains genomic information for three members of the FGFR family of RTK with either one ( EmFR1 ) or two ( EmFR2 , EmFR3 ) extracellular Ig-like domains , of which one , EmFR3 , showed higher divergence within the TKD as it contained only 22 of otherwise 30 highly conserved amino acid residues of TK . On the other hand , no members of the VEGF- and PDGF-receptor families are encoded by the E . multilocularis genome , nor does it contain genes coding for canonical FGF-ligands . To investigate gene expression patterns of the Echinococcus FGF receptors in parasite larvae , we first inspected Illumina transcriptome data for parasite primary cells after 2 and 11 days of culture ( PC2d , PC11d , respectively ) , metacestode vesicles without or with brood capsules ( MC- , MC+ , respectively ) , as well as protoscoleces before or after activation by low pH/pepsin treatment ( PS- , PS+ , respectively ) , that had be produced during the E . multilocularis genome project [14] . According to these data , emfr1 was moderately expressed in primary cells , metacestode vesicles , and protoscoleces ( S3 Fig ) . Likewise , emfr2 was expressed in all stages , but very lowly in primary cells , somewhat more in metacestode vesicles , and highest in protoscoleces . emfr3 , on the other hand , was low to moderately expressed in primary cells , low in metacestode vesicles , and highest in protoscoleces ( S3 Fig ) . Since primary cell preparations are characterized by a much higher content of germinative ( stem ) cells than metacestode vesicles [6] , these expression patterns could indicate that emfr3 is stem cell-specifically expressed . We therefore carried out quantitative RT-PCR experiments on cDNA preparations from in vitro-cultivated metacestode vesicles ( MV ) versus metacestode vesicles after treatment with hudroxyurea ( MV-HU ) or the Polo-like kinase inhibitor BI 2536 ( MV-BI ) , in which the stem cell population had been selectively depleted [6 , 40] . While emfr1 expression levels were equal in MV when compared to MV-HU and MV-BI , both emfr2 and emfr3 transcripts were significantly reduced in MV-HU and MV-BI ( S4 Fig ) . This indicated that emfr1 does not have a typical stem cell-specific expression pattern whereas emfr2 and emfr3 could be expressed in the parasite’s stem cells or their immediate progeny ( since HU- and BI 2536-treatment has to be carried out for at least one week [6 , 34] ) . To clarify the situation we carried out whole-mount in situ hybridization experiments on metacestode vesicles according to recently established protocols [6 , 43 , 44] . In these experiments , proliferating parasite stem cells were labeled by incorporation of the nucleotide analog EdU , which was combined with detection of gene transcripts by using fluorescently labeled probes . According to these experiments , emfr1 was expressed at low levels throughout the germinal layer and the signal was somewhat higher during the development of protoscoleces ( Fig 2 ) , maybe due to higher cell density in the protoscolex . The intensity of the signal was heterogenous , but no clear pattern could be discerned and it was too low to clearly determine the percentage of positive cells . emfr2 , on the other hand , was specifically expressed in a population of small sized cells , which comprised 1 . 7% to 6 . 3% of all cells in the germinal layer ( Fig 3A ) . None of these emfr2+ cells were also EdU+ , indicating that they were post-mitotic [6] . During initial protoscolex development , emfr2+ cells accumulated in the peripheral-most layer of cells , as well as in the anterior-most region ( which will form the rostellum ) , and in three longitudinal bands of cells in the interior of the protoscolex buds ( Fig 3B ) . Again , practically none of the emfr2 labelled cells was EdU+ ( less than 1% of emfr2+ cells were EdU+; Fig 3C ) . Later during development , some emfr2+ cells were found in the protoscolex body , but most accumulated in the developing rostellum and the suckers ( Fig 3D ) . Importantly , emfr2 expression was restricted to the sucker cup , where cells are differentiating into em-tpm-hmw+ muscle cells [6] , and not the sucker base , where EdU+ germinative cells accumulate [6] ( Fig 3D ) . Taken together , these data indicate that emfr2 is expressed in post-mitotic cells , of which many are likely to be differentiating or differentiated muscle cells . Finally , emfr3 was expressed in very few cells in the germinal layer ( less than 1% of all cells , although the number is difficult to estimate since they were absent in most random microscopy fields ) ( Fig 4A ) . emfr3+ cells accumulated in small numbers during brood capsule and protoscolex development ( Fig 4B ) . The emfr3+ cells had a large nucleus and nucleolus and , thus , had the typical morphology of germinative cells ( S5 Fig ) . At the final stages of protoscolex development , few emfr3+ cells were present in the body region and some signals were also present in the rostellar region ( S5 Fig ) . In the developing protoscolex and in the germinal layer , emfr3+ EdU+ double positive cells were found ( Fig 4 ) . These data indicated that emfr3 is expressed in a very small number of proliferating cells with the typical morphology of germinative cells . In summary , the three E . multilocularis FGF receptor genes showed very different expression patterns in metacestode and protoscolex larval stages . While emfr1 was lowly expressed in cells that are dispersed throughout the germinal layer , emfr2 displayed an expression pattern indicative of differentiating/differentiated muscle cells . emfr3 , on the other hand , appeared to be expressed in a minor sub-population of germinative cells . Since E . multilocularis larvae do not express canonical FGF ligands ( see above ) , but usually develop in an environment in which FGF1 and FGF2 are abundant [17] , we next tested whether host-derived FGF ligands can stimulate parasite development in vitro . To this end , we employed two different cultivation systems which we had previously established [7–10] . In the axenic metacestode vesicle cultivation system , mature metacestode vesicles of the parasite are cultivated in the absence of host cells under reducing culture conditions ( e . g . low oxygen ) [7 , 9] . In the primary cell cultivation system [8 , 10] , axenically cultivated metacestode vesicles are digested to set up cell cultures which are highly enriched in parasite germinative ( stem ) cells ( ~ 80% [6] ) , but which also contain certain amounts of differentiated cells such as muscle or nerve cells . In the primary cell cultivation system , mature metacestode vesicles are typically formed within 2–3 weeks , which is critically dependent on proliferation and differentiation of the germinative cell population [6] in a manner highly reminiscent of the oncosphere-metacestode transition [3] . As shown in Fig 5A , the exogenous addition of 10 nM FGF1 to mature metacestode vesicles already resulted in an elevated incorporation of BrdU , indicating enhanced proliferation of parasite stem cells , which was even more pronounced in the presence of 100 nM FGF1 . In the case of FGF2 , addition of 100 nM resulted in enhanced BrdU incorporation in a statistically significant manner ( Fig 5A ) . Likewise , metacestode vesicles cultured for 4 weeks in the presence of 10 nM FGF1 or FGF2 displayed a considerably larger volume ( about two-fold ) when compared to non-FGF-stimulated vesicles ( Fig 5B ) . In the primary cell cultivation system , 100 nM concentrations of host ligands had to be added to observe statistically significant effects . Again , the incorporation of BrdU by primary cell cultures was stimulated in the presence of host-derived FGF ligands ( Fig 5C ) , as was the formation of mature metacestode vesicles from primary cell cultures ( Fig 5D ) . Taken together , these results indicated that host-derived FGF ligands , and in particular FGF1 , can stimulate cell proliferation and development of E . multilocularis primary cell cultures and mature metacestode vesicles . Having shown that host-derived FGF ligands can stimulate parasite proliferation and development in vitro , we were interested whether these effects might be mediated by one or all three of the parasite’s FGF receptors which are expressed by the metacestode larval stage . To this end , we first made use of the Xenopus oocyte expression system in which the activity of heterologously expressed protein kinases can be measured by germinal vesicle breakdown ( GVBD ) . This system has previously been used to measure the activities of the TKD of schistosome FGF receptors [24] , as well as the host-EGF ( epidermal growth factor ) dependent activation of a schistosome member of the EGF receptor family [31] . We , thus , expressed Pleurodeles FGFR1 ( as a positive control ) , which is highly similar to human FGFR1 [32] , and all three parasite FGF receptors in Xenopus oocytes , which were then stimulated by the addition of 10 nM FGF1 or FGF2 . As negative controls , we also expressed kinase-dead versions of all Echinococcus FGF receptors in Xenopus oocytes . As can be deduced from Table 1 , control ( non-stimulated ) oocytes were negative for GVBD , whereas progesterone-stimulated oocytes displayed 100% vesicle breakdown . Expression of FGFR1 did not yield GVBD but , after stimulation with 10 nM FGF1 , 100% of oocytes underwent GVBD , indicating stimulation of the Pleurodeles receptor by FGF1 ( as expected ) . Upon expression of any of the parasite receptors in Xenopus oocytes , no GVBD was observed when no ligand was added . The addition of 10 nM FGF1 to these receptors , however , resulted in 100% GVBD for EmFR1 and EmFR2 , as well as to 80% GVBD in the case of EmFR3 . In the case of human FGF2 ( 10 nM ) , 90% GVBD was observed for EmFR1 and EmFR3 , and 85% for EmFR2 . No GVBD was observed upon addition of 10 nM FGF1 or FGF2 when the kinase-dead versions of the parasite receptors were expressed ( S6 Fig ) . These data clearly indicated that all three parasite receptors were responsive to host derived FGF ligands ( albeit to somewhat different extent ) and that the kinase activity of the parasite receptors was essential to transmit the signal . We also measured the phosphorylation state of the parasite FGF receptors upon addition of exogenous FGF1 and FGF2 ( 10 nM each ) to Xenopus oocytes and obtained significantly induced levels of tyrosine phosphorylation in all three cases ( S6 Fig ) . Taken together , these data indicated that all three parasite FGF receptors were activated by binding of host-derived FGF1 and FGF2 , which was followed by auto-phosphorylation of the intracellular kinase domain and downstream transmission of the signal to the Xenopus oocyte signaling systems which induce GVBD . The small molecule compound BIBF 1120 ( also known as Nintedanib or Vargatef ) is a well-studied and highly selective , ATP-competitive inhibitor of mammalian members of the FGF- , VEGF- , and PDGF-receptor families with very limited affinity to other RTK [45 , 46] . As a possible agent to selectively inhibit FGF RTK activities in the parasite , we measured the effects of BIBF 1120 on EmFR1 , EmFR2 , and EmFR3 upon expression in the Xenopus oocyte system . As can be deduced from Table 1 , a concentration of 1 μM of exogenously added BIBF 1120 already diminished the activity ( after stimulation with 10 nM FGF1 ) of FGFR1 to 40% , and led to a complete block of kinase activity upon addition of 10 μM BIBF 1120 . In the case of EmFR1 and EmFR2 , 1 μM BIBF1120 also led to a marked decrease of receptor kinase activity , although to a somewhat lower extent than in the case of FGFR1 . In the presence of 10 μM BIBF 1120 , on the other hand , the activities of EmFR1 and EmFR2 were completely blocked . In the case of EmFR3 , exogenous addition of 1 μM BIBF 1120 had only slight effects on GVBD , whereas 10 μM BIBF 1120 reduced the activity to less than 50% and 20 μM BIBF 1120 completely blocked TK-dependent GVBD ( Table 1 ) . Upon addition of 20 μM BIBF 1120 , TK activity of all receptors tested was completely inhibited . Taken together , these data indicated that all three Echinococcus FGF RTK were affected by BIBF 1120 , although in all three cases higher concentrations of the inhibitor were necessary to completely block TK activity when compared to FGFR1 . BIBF 1120 treatment had the lowest effects on the activity of EmFR3 . We next tested the effects of different concentrations of BIBF 1120 on parasite development and survival in the primary cell and metacestode vesicle culture systems . As shown in Fig 6A , the addition of 1–10 μM BIBF 1120 had clear concentration-dependent effects on mature metacestode vesicle survival which , after cultivation for 18 days , led to about 20% surviving vesicles in the presence of 1 μM BIBF 1120 , 10% surviving vesicles in the presence of 5 μM BIBF 1120 , and no survival when 10 μM BIBF 1120 was applied . To test whether the metacestode vesicles were indeed no longer capable of parasite tissue regeneration , we set up primary cell cultures from microscopically intact vesicles which had been treated with 5 μM BIBF 1120 for 5 days ( 90% intact vesicles ) and let the cultures recover in medium without inhibitor . In these cultures , however , we never observed the formation of mature vesicles , indicating that either the parasite’s stem cell population , and/or other cell types necessary for parasite development in the primary cell culture system , were severely damaged after BIBF 1120 treatment . We then also tested the effects of BIBF 1120 on fresh primary cell cultures from previously untreated metacestode vesicles . As shown in Fig 6B , a concentration of 1 μM BIBF 1120 had no effect on the formation of metacestode vesicles from these cultures , whereas vesicle formation was completely blocked in the presence of 5 μM or 10 μM BIBF 1120 . Altogether , these results clearly indicated detrimental effects of BIBF 1120 on parasite development already at concentrations as low as 5 μM . Since the parasite does not express known alternative targets for BIBF 1120 , such as VEGF- or PDGFR-receptors , we deduced that these effects are due to the inhibition of one or more of the parasite’s FGF RTK . One of the major downstream targets of FGF signaling in other organisms is the Erk-like MAPK cascade , a complete module of which we had previously identified in E . multilocularis [33 , 47 , 48] . In particular , we had previously shown that the phosphorylation status of the parasite’s Erk-like MAP kinase , EmMPK1 , can be measured by using antibodies against the phosphorylated form of the human Erk-kinase [33] . To investigate whether exogenously added host FGF can affect the E . multilocularis Erk-like MAPK cascade module , we first incubated mature metacestode vesicles for 4 days in serum-free medium ( which has no effect on vesicle integrity [29] ) and then stimulated these vesicles for 30 sec , 60 sec , and 60 min with 10 nM FGF1 and FGF2 . As shown in Fig 7A , FGF1 treatment had a clear effect of EmMPK1 phosphorylation already after 30 sec of exposure . In the case of FGF2 , the effect was still measurable , but clearly less pronounced than in the case of FGF1 ( Fig 7A ) . We then also measured the effects of BIBF 1120 treatment on EmMPK1 phosphorylation . To this end , metacestode vesicles were incubated in hepatocyte-conditioned medium and were then subjected to BIBF 1120 treatment ( 5 μM , 10 μM ) for 30 min . As shown in Fig 7B , this led to diminished phosphorylation of EmMPK1 when 10 μM BIBF 1120 was used . Taken together , these data indicated that , like in mammals and other invertebrates , the E . multilocularis Erk-like MAPK cascade can be activated through FGF signaling , initiated by exogenously added , host-derived FGF ligands .
In the present work we provide clear evidence that human FGF ligands are capable of activating evolutionarily conserved TK of the FGF receptor family that are expressed by the larval stage of E . multilocularis and that the parasite’s Erk-like MAPK cascade is stimulated upon exogenous addition of human FGFs to metacestode vesicles . We also showed that human FGF1 and FGF2 are stimulating the development of metacestode vesicles from parasite primary cell cultures and that they accelerate metacestode vesicle proliferation and growth in vitro . Since FGF1 and FGF2 are expressed in considerable amounts within the host liver , the primary target organ for the establishment of the E . multilocularis metacestode , and since FGF ligands are also constantly produced during liver regeneration and fibrosis , which are consequences of parasite growth within the intermediate host , we consider the observed in vitro effects on parasite FGF signalling and metacestode development also of high relevance in vivo . Liver-specific activities of host FGF could thus support the development of metacestode vesicles from stem cells that are delivered to the liver by the oncosphere larva , and could constantly stimulate asexual proliferation of the metacestode during an infection . We finally showed that at least one compound that inhibits the activities of mammalian FGF receptors , BIBF 1120 , also inhibits the parasite orthologs , leads to metacestode inactivation , and prevents parasite development of stem cell-containing primary cell cultures . This opens new ways for the development of anti-Echinococcus drugs using the parasite FGF receptors as target molecules . | To ensure proper communication between their different cell populations , animals rely on secreted hormones and cytokines that act on receptors of target cells . Most of the respective cytokines , such as FGFs , evolved over 500 million years ago and are present in similar form in all animals , including parasitic worms . The authors of this study show that the metacestode larva of the tapeworm E . multilocularis , which grows like a malignant tumor within the host liver , expresses molecules with homology to FGF receptors from mammals . The authors show that human FGF , which is abundantly present in the liver , stimulates metacestode development and that all parasite FGF receptors are activated by human FGF , despite 500 million years of evolutionary distance between both systems . This indicates that cells of the Echinococcus metacestode can directly communicate with cells of the mammalian host using evolutionarily conserved signaling molecules . This mode of host-pathogen interaction is unique for helminths and does not occur between mammals and single-celled pathogens such as protozoans or bacteria . The authors finally demonstrate that BIBF 1120 , a drug used to treat human cancer , targets the Echinococcus FGF receptors and leads to parasite death . This opens new ways for the development of anti-parasitic drugs . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"vesicles",
"electrochemical",
"cells",
"helminths",
"parasite",
"evolution",
"vertebrates",
"parasitic",
"diseases",
"animals",
"echinococcus",
"parasitology",
"xenopus",
"endocrine",
"physiology",
"animal",
"models",
"model",
"organisms",
"amphibians",
"experimental",
"organism",
"systems",
"growth",
"factors",
"cellular",
"structures",
"and",
"organelles",
"mapk",
"signaling",
"cascades",
"fibroblast",
"growth",
"factor",
"research",
"and",
"analysis",
"methods",
"animal",
"studies",
"endocrinology",
"flatworms",
"chemistry",
"xenopus",
"oocytes",
"primary",
"cells",
"signal",
"transduction",
"eukaryota",
"cell",
"biology",
"electrochemistry",
"physiology",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"cell",
"signaling",
"frogs",
"organisms",
"signaling",
"cascades"
] | 2019 | The role of fibroblast growth factor signalling in Echinococcus multilocularis development and host-parasite interaction |
Crimean-Congo haemorrhagic fever virus ( CCHFV ) is a member of the Orthonairovirus genus of the Nairoviridae family and is associated with haemorrhagic fever in humans . Although T lymphocyte responses are known to play a role in protection from and clearance of viral infections , specific T cell epitopes have yet to be identified for CCHFV following infection . A panel of overlapping peptides covering the CCHFV nucleoprotein and the structural glycoproteins , GN and GC , were screened by ELISpot assay to detect interferon gamma ( IFN-γ ) production in vitro by peripheral blood mononuclear cells from eleven survivors with previous laboratory confirmed CCHFV infection . Reactive peptides were located predominantly on the nucleoprotein , with only one survivor reacting to two peptides from the glycoprotein GC . No single epitope was immunodominant , however all but one survivor showed reactivity to at least one T cell epitope . The responses were present at high frequency and detectable several years after the acute infection despite the absence of continued antigenic stimulation . T cell depletion studies confirmed that IFN-γ production as detected using the ELISpot assay was mediated chiefly by CD8+ T cells . This is the first description of CD8+ T cell epitopic regions for CCHFV and provides confirmation of long-lived T cell responses in survivors of CCHFV infection .
Crimean-Congo haemorrhagic fever virus ( CCHFV ) is a member of the Nairoviridae family and has a tripartite , single-stranded , negative sense RNA genome [1–3] . The three segments are referred to as the large ( L ) , medium ( M ) and small ( S ) segments [4] . The L segment encodes the viral RNA dependant RNA polymerase which is responsible for mRNA synthesis and replication of the RNA genome [5] . The M segment encodes a number of non-structural proteins and two structural glycoproteins , GN and GC [6 , 7] . The structural glycoproteins are responsible for attachment to host cell surface receptors and therefore determine the host range and cell tropism and are also the targets for neutralizing antibodies . The viral nucleoprotein , encoded by the S segment , binds the RNA segments for the formation of ribonucleoprotein complexes and shows endonuclease activity , although the role of this activity in CCHFV infection is not yet clear [8 , 9] . CCHFV is the only member of the CCHFV serogroup of medical importance . The other members of the group , Hazara virus and Khasan virus , have not been associated with disease in humans [10–12] . CCHFV infection in humans is associated with haemorrhagic fever and is fatal in up to 30% of cases [13] . The principal vectors of the virus are ticks belonging to the genus Hyalomma [14] . The distribution of disease follows that of the principal vector of the virus [13 , 15] . Clinical disease is well described in Africa , Asia , Eastern Europe and the Middle East and has recently emerged in Turkey , Greece , India and Spain [16–19] . With nearly 10 000 human cases reported to the Ministry of Health in Turkey between 2002 and December 2015 , as well as expanding areas of endemicity , the development of effective preventative and therapeutic measures have now become a priority [20] . To this end , the correlates of protection against CCHFV need to be determined . Antibody responses develop within 7–9 days of infection and include a transient IgM response , and an IgG response which likely persists for life [21 , 22] . In cases where an antibody response is not detectable by day nine , a fatal outcome is almost invariably seen [21] . However studies looking at viral load and antibody titre deduced that the detection of IgM had no influence on clearance of the virus or outcome , and viral load decreased independently of IgG [23] . Hence , antibody production does not always correlate with viral clearance , implying that innate and T cell immunity likely also play an important role in viral clearance [24] . In addition , neutralizing antibodies do not always confer protection in vivo , while non-neutralizing antibodies may confer protection through mechanisms such as antibody-dependent cell-mediated cytotoxicity [25] . The nucleoprotein induces antibody production and this region is widely used as the antigenic target in enzyme-linked immune-sorbent assays ( ELISA ) due to the robust nature of the antibody response to this protein following natural infection with CCHFV [26–34] . A study in mice suggested that neutralizing antibodies were predominantly targeted to the glycoprotein GC [25] . Ideally , an effective vaccine will therefore likely require the induction of both B and T cell responses . As the immune correlates of protection for CCHF are unknown , approaches to vaccine design have targeted the CCHF nucleoprotein or glycoproteins . Currently , the only vaccine that is available for human use is an inactivated vaccine used only in Bulgaria which was associated with a four-fold reduction in the number of CCHF cases following its introduction . The Bulgarian vaccine has been shown to induce both neutralising antibody responses and T cell responses to peptides across the length of the nucleoprotein , although immunogenicity required the administration of multiple doses [35 , 36] . Data on immune responses to the M segment proteins and challenge studies are not available for this vaccine . A modified vaccinia virus Ankara vectored vaccine expressing the nucleoprotein induced cellular and humoral immune responses yet failed to protect mice from lethal disease [37] . Glycoprotein-based vaccines have been investigated with varying degrees of protection in mouse models [38 , 39] . The most promising approach thus far is a modified vaccinia virus Ankara recombinant vaccine expressing the viral glycoproteins which induced cellular and humoral responses and provided protection from lethal disease in mice [39 , 40] . There are currently no data available evaluating T cell responses following natural infection with CCHFV in human cases . To further understand the role of T lymphocytes in the immune response to CCHFV infection , we aimed to determine whether memory T cell responses could be identified in survivors of CCHFV infection and to identify the viral proteins on which the T cell epitopes are located . An overlapping peptide library was used to screen for T cell responses by IFN-γ ELISpot assay using peripheral blood mononuclear cells ( PBMC ) derived from survivors of previous CCHFV infection .
Written informed consent was obtained from all participants and approval for the study was obtained from the Ethics Committee of the Faculty of Health Sciences , University of the Free State ( ECUFS NR 152/06 ) . Eleven adult patients with a history of laboratory confirmed CCHFV infection were included in the study . Laboratory confirmation of CCHFV infection was performed at the National Institute of Communicable Diseases , Johannesburg , South Africa at the time of the acute illness by means of viral nucleic acid detection by reverse transcription polymerase chain reaction ( RT-PCR ) , virus isolation , or detection of CCHFV specific antibodies . Sodium heparin blood samples were transported to the laboratory for processing within 4 to 6 hours after collection . Two additional participants were included in the study as negative controls and were selected as they had no history of CCHFV infection , exposure to CCHFV or risk factors for such exposure . An overlapping peptide library containing 156 peptides consisting of 19-mers with a 9-mer overlap and spanning the nucleoprotein ( 482 amino acids ) and the mature glycoproteins , GN ( 292 amino acids ) and GC ( 648 amino acids ) , were synthesized ( Mimotopes , Victoria , Australia ) based on the deduced amino acid sequences of CCHFV isolate SPU103/87 ( GenBank accession numbers DQ211647 and DQ211634 ) . Fresh PBMC were isolated from whole blood using Ficoll-Hypaque density gradient centrifugation . IFN-γ ELISpot assays were performed using 96 well plates ( MultiScreen-IP , Millipore ) pre-coated with anti-IFN-γ antibody ( clone 1-D1K; Mabtech ) at 2μg/ml at 4°C overnight . After washing with sterile phosphate buffered saline ( PBS ) , the plate was blocked with R10 ( RPMI 1640 medium plus 1% penicillin-streptomycin , 1% L-glutamine and 10% foetal calf serum ) for 2 hours at room temperature . Screening of the peptide library was performed by means of a matrix of 36 peptide pools containing 7 to 9 peptides each so that each peptide was included in two separate pools . Peptides were added at a final concentration of 5μg/ml for each peptide . PBMC were then seeded at 1x105 to 2x105 cells per well in R10 . For positive controls , PBMC were stimulated with phytohaemaglutinin ( Sigma-Aldrich ) and 0 . 1μg/ml monoclonal antibody to human CD3 ( clone CD3-2 , Mabtech ) in separate wells . Unstimulated PBMCs in R10 and wells with media and no PBMCs were included as negative controls . Plates were incubated for 16 hours at 37°C and 5% CO2 . The plates were then washed 6 times with PBS and 2μg/ml of biotinylated anti-IFN-γ antibody ( clone 7-B6-1 , Mabtech ) was added and incubated for 3 hours at room temperature . The plates were again washed with PBS and streptavidin-alkaline phosphatase conjugated antibody ( Mabtech ) was added and the plate incubated for 1 hour at room temperature in the dark . After a final wash , the alkaline phosphatase conjugate substrate kit ( Bio-Rad ) was used according to manufacturer’s instructions to detect IFN-γ producing cells . The number of spots were counted manually and results were expressed as the number of spot forming cells per million cells ( SFC/106 cells ) . A response was considered positive if it exceeded 50 SFC/106 cells after subtraction of the background count from the negative controls , while negative controls were consistently below this cut-off . The peptides showing a positive response using the pool screening method were then tested individually using the same method and with a final peptide concentration of 5 μg/ml per well . T cell subset depletion assays were performed to determine whether the positive responses obtained were predominantly due to CD4+ or CD8+ T cells . Based on the outcome of initial ELISpot screening , a patient that reacted strongly to two epitopes was selected for subset depletion assays in which PBMC isolated from the patient were depleted of CD8+ T cells using Dynabeads CD8 ( Invitrogen ) according to the manufacturer’s instructions . The ELISpot assay was then performed using depleted and undepeleted cells stimulated with each peptide for which the survivor had previously shown a positive reaction . To determine whether the identified T cell epitopes were conserved amongst geographically distinct CCHFV isolates , 40 complete S and M segment sequences were retrieved from the GenBank database . The sequences were aligned using Clustal X version 2 . 0 [41] and further edited using BioEdit version 7 . 2 . 3 ( available at http://www . mbio . ncsu . edu/bioedit/bioedit . html ) .
Eleven survivors were included in the study , all of whom were Caucasian males residing in the Free State and North West provinces of South Africa at the time of CCHFV infection . The interval between CCHFV infection and sample collection for this study ranged from 10 months to 13 years . The details of the survivors are summarised in Table 1 . To identify epitopic regions targeted by T cell responses , PBMCs were collected from survivors and stimulated using 36 pools containing 7–9 overlapping peptides representing the nucleoprotein and glycoproteins , GN and GC . IFN-γ responses were detected against 16 peptides using the ELISpot assays , with all but one survivor responding to at least one peptide ( Fig 1 ) . Due to reactivity against adjacent peptides , it is likely that six of the epitopes reside within the overlapping regions , resulting in a total of ten epitopic regions identified . The details of the peptides showing positive responses are summarised in Table 2 . The majority of the potential epitopes were located on the nucleoprotein , with only one patient responding to two peptides located on the glycoprotein GC . No region of the nucleoprotein appeared immunodominant , with the epitopes distributed throughout the length of the protein . The two glycoprotein epitopes were restricted to GC , starting at amino acid positions 244 and 316 respectively of the 648 amino acid long coding region of this protein . No response was detected when PBMC from two volunteers with no history of CCHF infection were tested against the 16 reactive peptides , confirming that the responses detected in the study participants likely result from previous CCHFV exposure and were not due to non-specific reactivity . No clear conclusion could be reached on the effect of the time interval between CCHFV infection and sampling with regard to the magnitude and range of T cell responses . Although the two survivors with intervals of 12 and 13 years since infection respectively showed only low level responses ( 55–70 SFC/106 ) to one or two peptides , the survivor with the most recent history of infection only 10 months prior to sampling also showed a low level response ( 65 SFC/106 ) to a single peptide . This variation is therefore likely a consequence of individual patient responses rather than duration after infection . PBMCs from three individuals had strong responses ( 65 - >500 SFC/106 ) against peptide N262-280 and PBMCs from five individuals had significant responses ( 55 - >500 SFC/106 ) against N298-316 , hence these two peptides were selected for CD8+ depletion studies . Participant 11 was selected due to strong responses to both of these peptides , with 290 and >500 SFC/106 detected respectively during the screening ELISpot testing . Following CD8+ T cell depletion , the number of SFC/106 dropped to 5 and 25 respectively confirming a predominantly CD8+ T cell response ( Fig 2 ) . Alignment of the 40 predicted amino acid sequences of the epitopic regions obtained from GenBank were relatively well conserved among geographically distinct CCHFV isolates despite representing genetically diverse groups . The global sequences showed a maximum of three amino acid differences across each 19-mer compared to SPU103/87 , with the exception of isolate AP92 which showed higher variance . Amino acid sequence conservation was higher among the southern African isolates with a maximum of one amino acid difference across each 19mer . In addition to the geographical distribution , the CCHFV isolates were also grouped according to phylogenetic relatedness as described previously [42 , 43] and the number of amino acid differences were tabulated for each peptide ( Table 3 ) . This confirmed the higher variance of isolate AP92 ( group VI ) , while the other groups showed a maximum of 3 amino acid differences . The amino acid alignments for each peptide are included in the supporting information files ( S1 Data ) .
The present study is the first description of T cell epitopes in survivors of CCHFV infection . A library of overlapping peptides covering the nucleoprotein , GN and GC proteins of CCHFV was screened by ELISpot assay to determine in vitro induction of IFN-γ production using PBMC from survivors . The rationale for selection of these viral proteins was because they are the most likely to contribute towards a protective immune response based on knowledge from related viruses and CCHF vaccine studies [37 , 44–47] . Ten probable epitopic regions residing predominantly on the nucleoprotein with only two epitopes on the glycoprotein GC were identified . The predominance of nucleoprotein epitopes is similar to findings on Hantaan virus T cell epitopes [44] . The reason for this is unclear but may result from the abundance of nucleoprotein production during CCHFV replication in vivo . A nucleoprotein based vaccine for Rift Valley fever virus ( RVF ) , another bunyavirus , has been shown to protect against challenge with a lethal dose of virus in a mouse model despite the absence of neutralizing antibodies [45] . This , and other RVF nucleoprotein vaccines , have been shown to induce strong memory T cell responses and an early type I interferon response which likely contribute to protection [46 , 47] . The CCHFV nucleoprotein represents an enticing option for vaccine design due to the greater sequence conservation of this protein and strong antibody responses , however a recombinant nucleoprotein vaccine failed to provide protection despite inducing both cellular and humoral responses [37] . No single epitope was found to be immunodominant , rather a variety of epitopes were identified in different patients . The high level of amino acid sequence conservation amongst southern African CCHFV isolates implies that this did not result from sequence variation which may result in an inability of T cells to recognise the peptides , but likely represents the HLA diversity within the patient population studied and the differing HLA restrictions of the peptides . A range of epitopes will most probably need to be included in any prospective vaccine candidates in order to induce protective T cell and antibody responses in all vaccine recipients . The current findings suggest that vaccines based on the structural glycoproteins , GN and GC , may not induce sufficient T cell responses to provide protection in human populations . This is supported by the failure of a subunit vaccine comprised of GN and GC to provide protection despite inducing high levels of neutralizing antibodies [38] . Although the M segment is known to be less conserved than the S segment , with amino acid variation of up to 30 . 1% and 8 . 4% respectively amongst global isolates , differences in amino acid identity for the GN and GC proteins specifically are less marked at 8 . 5% and 4 . 8% respectively [42] . It is therefore unlikely that the limited responses to the glycoproteins resulted from the inability of memory cells to recognise epitopes on the synthetic peptides . The strength of the T cell responses identified differed between peptides and between patients responding to the same peptide . The magnitude of T cell responses induced by longer peptides such as the 19mers used in this study , may be underestimated in comparison to the more effective responses induced by optimal nonamers . This increased efficiency of T cell response induction may result from direct binding of antigen presenting cells to the nonamers [44] . This phenomenon would not have influenced the outcomes of this study which aimed to identify long lived T cell responses and the location of these epitopes on viral proteins rather than to accurately quantify the magnitude of these responses . Memory T cells , especially CD8+ cytotoxic cells , are able to rapidly proliferate and differentiate into effector cells which are an important part of long-lived protective immune responses following both vaccination and natural infection [48] . Depletion studies from one patient confirmed that the T cell responses observed following CCHFV infection in that patient resulted from CD8+ cells rather than CD4+ helper cells . Examination of additional patients and phenotype of reactive memory cells requires further confirmation . CCHFV is transmitted to humans by tick bites , squashing of ticks with bare hands and contact with blood or tissues of infected humans and animals . Persons at high risk of exposure to CCHFV therefore include those likely to be exposed to ticks and those performing procedures resulting in exposure to blood or tissues of infected animals such as castration or slaughtering including farmers , farm workers , abattoir workers and veterinarians [49 , 50] . From the first recognised case of CCHFV infection in South Africa in 1981 until 2013 , a total of 194 cases were recorded in this country . Of these cases , 91% occurred in males and more than 50% originated in the Free State and Northern Cape Provinces [51] . The exclusively male cohort analyzed in this study is therefore representative of the gender distribution of cases in South Africa . T cell responses to CCHFV peptides were detected in this study population up to 13 years after acute CCHFV infection . The longevity of the responses points towards memory T cells playing an important role and support findings of the long-term presence of memory CD8+ T cells following infection with another bunyavirus , Puumala virus . This virus causes an acute infection with no known latent or chronic phase in humans , however Puumala virus specific memory T cell responses were present up to 15 years after infection despite the absence of continued antigenic stimulation or re-exposure [52] . These data imply that effective long term protection from infection may be achieved through vaccination . The MVA vectored glycoprotein vaccine which provided 100% protection in mouse challenge studies , induced both cellular and humoral responses . This vaccine expressed both the structural and non-structural glycoproteins , namely the variable mucin-like domain , GP38 and NSM [39] . The T cell responses to this vaccine mapped largely to the N-terminus of GC and the non-structural glycoproteins . Similarly , only two epitopes were identified on GC following natural infection , with no responses to GN . Further studies are underway to investigate T cell responses to the non-structural domains in patients following CCHFV infection . The novel epitopic regions which were identified within the nucleoprotein of CCHFV represent the first such T cell epitopes to be described following natural infection and may play an important role in vaccine design and evaluation of vaccine immunogenicity . | Crimean-Congo haemorrhagic fever virus ( CCHFV ) is a tick-borne virus that causes haemorrhagic fever with fatalities in humans . Clinical disease has been described in Africa , Asia , Eastern Europe , the Middle East , Turkey and more recently Greece , India and Spain . In the absence of specific anti-virals and vaccines , supportive therapy is the method of treatment . Specific T cell epitopes have yet to be identified for CCHFV following infection and although antibody responses are detectable in survivors , antibody production does not always correlate with viral clearance , implying that innate and T cell immunity likely also play an important role . To further understand the role of T lymphocytes in the immune response to CCHFV infection , we aimed to determine whether memory T cell responses could be identified in survivors of CCHFV infection and to identify the viral proteins on which the T cell epitopes are located . T cell responses to CCHFV peptides were detected in this study population up to 13 years after acute CCHFV infection and provides confirmation of long-lived CD8+ T cell responses . The longevity of the responses points towards memory T cells playing an important role and implies that effective long term protection from infection may be achieved through vaccination . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"blood",
"cells",
"sequencing",
"techniques",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"enzyme-linked",
"immunoassays",
"immune",
"physiology",
"immunology",
"vaccines",
"nucleoproteins",
"protein",
"sequencing",
"cytotoxic",
"t",
"cells",
"molecular",
"biology",
"techniques",
"infectious",
"disease",
"control",
"glycoproteins",
"immunologic",
"techniques",
"antibodies",
"research",
"and",
"analysis",
"methods",
"immune",
"system",
"proteins",
"infectious",
"diseases",
"white",
"blood",
"cells",
"animal",
"cells",
"proteins",
"t",
"cells",
"immunoassays",
"molecular",
"biology",
"biochemistry",
"cell",
"biology",
"physiology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"glycobiology"
] | 2017 | Long-lived CD8+ T cell responses following Crimean-Congo haemorrhagic fever virus infection |
Pathogenic Leptospira transmits from animals to humans , causing the zoonotic life-threatening infection called leptospirosis . This infection is reported worldwide with higher risk in tropical regions . Symptoms of leptospirosis range from mild illness to severe illness such as liver damage , kidney failure , respiratory distress , meningitis , and fatal hemorrhagic disease . Invasive species of Leptospira rapidly disseminate to multiple tissues where this bacterium damages host endothelial cells , increasing vascular permeability . Despite the burden in humans and animals , the pathogenic mechanisms of Leptospira infection remain to be elucidated . The pathogenic leptospires adhere to endothelial cells and permeabilize endothelial barriers in vivo and in vitro . In this study , human endothelial cells were infected with the pathogenic L . interrogans serovar Copenhageni or the saprophyte L . biflexa serovar Patoc to investigate morphological changes and other distinctive phenotypes of host cell proteins by fluorescence microscopy . Among those analyzed , 17 proteins from five biological classes demonstrated distinctive phenotypes in morphology and/or signal intensity upon infection with Leptospira . The affected biological groups include: 1 ) extracellular matrix , 2 ) intercellular adhesion molecules and cell surface receptors , 3 ) intracellular proteins , 4 ) cell-cell junction proteins , and 5 ) a cytoskeletal protein . Infection with the pathogenic strain most profoundly disturbed the biological structures of adherens junctions ( VE-cadherin and catenins ) and actin filaments . Our data illuminate morphological disruptions and reduced signals of cell-cell junction proteins and filamentous actin in L . interrogans-infected endothelial cells . In addition , Leptospira infection , regardless of pathogenic status , influenced other host proteins belonging to multiple biological classes . Our data suggest that this zoonotic agent may damage endothelial cells via multiple cascades or pathways including endothelial barrier damage and inflammation , potentially leading to vascular hyperpermeability and severe illness in vivo . This work provides new insights into the pathophysiological mechanisms of Leptospira infection .
The causative agents of leptospirosis , Leptospira species , are Gram-negative spirochetes of the class Spirochaetales , along with Borrelia and Treponema [1] . The genus Leptospira has at least 22 species with 300 serovars and are classified as pathogenic , saprophytic , and intermediate types [2–4] . Pathogenic Leptospira transmission to humans and susceptible animals causes the zoonotic infection leptospirosis . This life-threatening infection is reported in temperate and especially tropical regions worldwide [5 , 6] . The reservoirs of these bacteria are rodents and other domestic and wild animals , which release bacteria-containing urine into water , mud , and soil . Humans exposed to these contaminated sources can be infected through damaged skin or through mucous membranes , including the conjunctiva [1 , 2 , 6] . The global burden of leptospirosis is estimated to be more than 1 million cases and nearly 60 , 000 deaths annually [5] . Symptoms are often non-specific but may include high fever , severe headache , chills , myalgia , rash , vomiting , jaundice , red eyes , abdominal pain , and diarrhea [1 , 6–8] . Invasive species of Leptospira rapidly disseminate to multiple tissues where they damage host endothelial cells and increase vascular permeability , causing more severe illness such as acute renal injury , aseptic meningitis , liver failure , and respiratory distress from acute lung injury [8–12] . Due to the variety of symptoms , patients can be misdiagnosed as having other common viral or bacterial infectious diseases [5 , 6 , 8] . Leptospirosis can lead to multiple organ failure or fatal hemorrhagic diseases [1 , 8–10] . Leptospires are known to adhere to fibroblasts , renal epithelial cells , macrophages , and endothelial cells in vitro [13–17] . Multiple Leptospira adhesins have been reported to bind cells via VE-cadherin or the extracellular matrix ( ECM ) molecules fibronectin , collagen , laminin , elastin , and plasminogen [18–22] . Binding to glycosaminoglycans ( GAGs ) may promote attachment to cells and to ECM [18 , 19 , 22] . The adhesion of pathogenic leptospires is likely an important early stage of the infectious process . Pathological characteristics of leptospirosis are vasculitis and endothelial cell damage , leading to inflammatory infiltrates , localized ischemia and hemorrhage in organs [1 , 2 , 11 , 12] . Despite the burden in humans and animals , the pathogenic mechanisms of Leptospira infection at the cellular and molecular levels are poorly understood . It has been demonstrated that the pathogenic L . interrogans or its proteins adhere to endothelial cells and permeabilize endothelial cell monolayers in vitro [23 , 24] . At the molecular level , genetic tools for Leptospira work less efficiently than for many other bacteria [25] , making studies of Leptospira biology challenging . Even when Leptospira mutants are constructed , there are few available efficient methods to elucidate the pathogenic mechanisms [26] . Most often , the mutants have been examined for attenuated phenotypes in the mortality of animals or altered histological perturbations of organs from infected animals , such as hamsters [26–30] , guinea pigs [30 , 31] , transgenic mice [32 , 33] , and zebrafish [34 , 35] . Because wild-type mice and rats are carrier animals for leptospires , these animals are used as negative controls to study pathogenicity , e . g . to examine leptospire colonization without disease [26 , 36] . As in vitro assays to screen Leptospira strains , cell attachment and transmigration through polarized epithelial cells have been used [16 , 37] . The epithelial translocation of leptospires does not alter the transepithelial electrical resistance [37] , so the resistance measurement itself is not informative . Human umbilical vein endothelial cells ( HUVEC ) are also used to test recombinant leptospiral proteins for changes in host protein expression or cell-junction permeability [24 , 38 , 39] . To study Leptospira pathogenicity more intensely , other in vitro screening systems remain to be explored . In this study , morphological changes and other distinctive phenotypes in Leptospira-infected human endothelial cells were investigated . Antibodies and reagents recognizing human proteins were screened by fluorescence microscopy . Most proteins analyzed demonstrated little change in Leptospira-infected endothelial cells . Yet , 17 host proteins from five biological classes demonstrated distinctive phenotypes in the morphology and/or signal intensity upon infection: 1 ) extracellular matrix , 2 ) intercellular adhesion molecules and cell surface receptors , 3 ) intracellular proteins , 4 ) cell-cell junction proteins , and 5 ) a cytoskeletal protein . The most prominent phenotype of pathogenic L . interrogans sv . Copenhageni-infected cells was the loss of the adherens junction proteins VE-cadherin and p120- , alpha- , and beta-catenins from the original site at intercellular junctions . Copenhageni infection also influenced the actin cytoskeleton as well as a tight junction protein , ZO-1 . Infection with both pathogenic and non-pathogenic Leptospira strains altered other host proteins belong to multiple biological classes , although the pathogenic strain caused more intense changes . This work provides the insights in biological and pathological effects of Leptospira infection .
L . interrogans serovar Copenhageni strain Fiocruz L1-130 ( pathogen ) , L . interrogans sv . Canicola strain Moulton ( pathogen ) , and L . biflexa sv . Patoc strain Patoc 1 ( non-pathogen ) were purchased from ATCC . Bacteria were grown at 30°C in Ellinghausen-McCullough-Johnson-Harris ( EMJH ) medium supplemented with 1% rabbit serum [40 , 41] . The viability , motility , and general morphology of strains were periodically checked using darkfield microscopy . When bacterial cultures reached 1 to 2 × 108 cells/ml , bacteria were used for infection or subcultured in fresh medium . The bacterial cell number was determined using a Petroff-Houser chamber under darkfield microscopy prior to infection . Bacterial cultures of 8 passages or less were used for all experiments . All procedures involving Leptospira were performed in a biosafety cabinet . In this study , the wild-type B . burgdorferi B31-A3 strain was used as a control bacterium . B . burgdorferi was grown in Barbour-Stoenner-Kelly ( BSKII ) medium [42] at 33°C to a density of 1 × 108 cells/ml . Prior to each infection , the presence of genomic plasmids was confirmed in each culture by PCR [43 , 44] . The dermal endothelial cell line human microvascular endothelial cells ( HMEC-1 ) , was originally a gift from Dr . E . Ades and Dr . T . J . Lawley ( Emory School of Medicine and the Centers for Disease Control and Prevention ) . This cell line is currently available from ATCC . HMEC-1 were cultured in MCDB 131 ( Gibco ) supplemented with 10 mM L-glutamine , 10 ng/ml epidermal growth factor ( Corning ) , 1 μg/ml hydrocortisone ( Sigma ) , 15% Hyclone Fetal bovine serum ( FBS , Thermo ) , and 25 mM HEPES ( Gibco ) . Two types of primary cells , human dermal lymphatic endothelial cells ( HDLEC ) and human dermal microvascular endothelial cells ( HDMEC ) , were purchased from ScienCell . These primary cells were cultured in endothelial cell medium ( ECM ) with the endothelial cell growth supplement ( ECGS ) and 5% FBS ( all from ScienCell ) , according to the vendor’s protocol . All human cells were grown at 36 . 5°C under 5% CO2 . For infection , passage of endothelial cells was limited to 20 or less for HMEC-1 and 13 or less for HDLEC and HDMEC . In preliminary experiments , we confirmed the morphological similarity of HDMEC to the cell line HMEC-1 . For screening , HMEC-1 was selected for use as a stable microvascular endothelial cell type , and for some experiments was compared to HDLEC , which possesses morphologically well-organized intercellular junctions . Human endothelial cells were grown on sterile coverslips placed in wells of 12-well plates . Seeding numbers used are 4 . 2 x 105/well for HMEC-1 and 2 . 2 x 105/well for HDLEC for 2-day growth . Cells were checked under a brightfield microscope for confluence and maturation of intercellular junctions prior to infection . Cells were washed with phosphate-buffered saline ( PBS ) once , placed in cell culture medium ( supplemented MCDB 131 or ECM ) , and infected with L . interrogans sv . Copenhageni or L . biflexa sv . Patoc in EMJH ( similar volumes of EMJH were added to the uninfected control cells ) at a multiplicity of infection ( MOI ) of 20 for 24 h at 36 . 5°C in 5% CO2 . Infection with B . burgdorferi B31-A3 was performed under the same conditions as for Leptospira . Infected endothelial cells were fixed with 2% para-formaldehyde for 15 min and then rinsed with PBS three times prior to immunofluorescence procedures . All procedures were performed in a biosafety cabinet . The infected and then fixed cells were rinsed with PBS and then either directly used for immunofluorescence procedures without permeabilization or treated with 0 . 1% Triton X-100 in PBS for 15 min for permeabilization , depending on the cell localization of a host protein or the specificity of an antibody ( shown in Table 1 ) . The samples were blocked with 3% bovine serum albumin ( BSA ) in PBS for 1 h and then incubated in a primary antibody diluted in 3% BSA/PBS for 1 h . For this study , antibodies and reagents were titrated and optimized for immunofluorescence microscopy analyses . The primary antibodies for which data are shown and the dilution factor used are listed in Table 1 . Unbound primary antibody was washed away with 3% BSA/PBS three times prior to incubation with either an anti-mouse-IgG or anti-rabbit-IgG antibody conjugated with Alexa Fluor 488 ( Molecular Probes ) for detection . After 1 h incubation , the unbound secondary antibody was rinsed away with 3% BSA/PBS twice and then with PBS twice . Filamentous actin was labeled with Alexa Fluor 488-conjugated phalloidin ( Table 1 ) for 20 min and washed with 3% BSA/PBS twice and then with PBS twice . Coverslips were mounted on glass slides using ProLong Diamond containing DAPI ( Molecular Probes ) . The mountant was cured in the dark for 12 h or longer before sealing of the coverslips with nail polish . Fluorescence microscopy images were acquired by a Nikon Eclipse Ti-U inverted microscope equipped with a CoolSNAP ES2 CCD camera ( Photometrics ) and a multifluorescent Sedat Quad ET filter set ( multichroic splitter , Chroma ) using the 20× Plan Apo objective lens ( N . A . 0 . 75 , Nikon ) . NIS-Elements software ( Nikon ) was used for image acquisition , processing , and analysis . Scale bars represent 50 μm . In each microscopy experiment , at least 3 to 5 images were acquired , and the experiment was independently repeated at least 3 times for each host protein tested . For the quantification of signal intensity , raw images of each host protein in endothelial cells were processed in NIS-Elements software ( Nikon ) as follows: adjusting the color values of indexed-color pixels , converting to RGB format , selecting the whole field of an image as a region of interest ( ROI ) , and obtaining the number of the mean intensity indicated in the ROI statistics in the software . Graphs of the quantified signal intensity were created in Microsoft Excel 2016 . Error bars indicate standard deviations ( SD ) from the means . Statistical analysis was performed using two-tailed unpaired t-test in GraphPad Prism version 7 . 00 . The p-values are indicated inside or below the graph .
Infection with pathogenic Leptospira strains and serovars causes vascular leakage in the tissues and organs of the host organism by increasing the permeability of endothelial layers [8–12] . The mechanisms of this disruption of the endothelial barrier at pathological and molecular levels are unclear . In this study , we used two types of cultured human endothelial cells to investigate which host proteins are affected during infection with pathogenic L . interrogans strains , compared to the non-pathogenic , saprophytic strain , L . biflexa sv . Patoc . For primary screening , the endothelial line HMEC-1 was used due to the faster growth rate and stability of this cell line . The phenotypes of interest detected in HMEC-1 were confirmed and further characterized by infection of primary endothelial cells , human dermal lymphatic endothelial cells ( HDLEC ) . Advantages to the use of HDLEC are: 1 ) cell size is large , 2 ) cell structure is generally flat without overlapped cell edges , and 3 ) the well-defined structure of cell-cell junctions . Endothelial cells were infected with leptospires at a multiplicity of infection ( MOI ) of 20 for 24 h throughout the screening process of candidate proteins ( Methods ) . Leptospires remained motile throughout the 24 h co-incubation . During incubation , we did not observe a hallmark of apoptosis , nuclear condensation or fragmentation in infected endothelial cells ( S5 Fig , DAPI ) . To investigate the effect of Leptospira infection on endothelial cells , we screened antibodies and reagents to detect any changes in biological structures of human proteins by immunofluorescence microscopy . After repeated screening , we found that the signal intensity or overall morphology of most of the host proteins tested were not significantly influenced by infection with either pathogenic or non-pathogenic leptospires . The lack of change could be a result of the irrelevance of the host protein to Leptospira infection or could be due to technical issues , such as the epitope position ( s ) in the tested protein or the antibody specificity for immunofluorescence microscopy . Among the analyzed reagents , we identified that the signal intensity and/or cellular morphology of 17 host proteins were affected by Leptospira infection ( Table 2 ) . These 17 human proteins , belonging to 5 biological groups , are the focus of this study . We first analyzed several extracellular matrix ( ECM ) proteins that are known to be the targets of many of the Leptospira adhesins [18–21] . In this study , L . interrogans sv . Copenhageni infection was found to influence three ECM proteins . One is collagen type IV , which is one of the most abundant ECM proteins and is located exclusively in the basement membrane [45] . Collagen type IV provides a scaffold for cell structural stability and also plays a role in interaction of cells with underlying basement membranes , critical for cell adhesion [45] . Compared to uninfected endothelial cells , Copenhageni infection increased the signal intensity by 1 . 5- to 2-fold with concomitant morphological changes leading to formation of puncta ( Fig 1A , Table 2 ) . The signal increase was minor in L . biflexa sv . Patoc-infected cells , ~1 . 3-fold ( Fig 1A ) . The signal increase and punctate morphology were also observed with another type of collagen , type VI ( S1A Fig ) . Decorin is another ECM protein affected by infection . Decorin is a small leucine-rich proteoglycan that associates with fibrillar collagen type I [46 , 47] . After Leptospira infection , the signal intensity of decorin slightly increased by 1 . 2- to 1 . 4-fold , in combination with increases in puncta in both Patoc- and Copenhageni-infected endothelial cells ( Fig 1B , Table 2 ) . These data indicate that the changes of decorin signal and morphology are caused by Leptospira infection in general , and are not specific to the pathogenic strain . Laminin is a glycoprotein , a major component of basal lamina located in the basement membrane [48] . In micrographs of uninfected endothelial cells , laminin displayed an intricate net-shaped structure ( Fig 1C , uninfected ) . When cells were infected with Copenhageni , the laminin network appeared to form bundles of small , rolled up , rope-like nets ( Fig 1C ) . There was a minor reduction of the signal intensity in Copenhageni-infected HDLEC but not in HMEC-1 ( Fig 1C ) . In Patoc-infected cells , there were only subtle structural rearrangements or changes in the signal intensity of laminin ( Fig 1C ) . These data suggest that the rearrangement of the laminin structure is a pathogenic Copenhageni-specific phenotype . Fibronectin is one of most abundant ECM proteins in tissues , along with collagen and laminin . There was no detectable change in fibronectin morphology upon infection with either Copenhageni or Patoc , although the signal intensity of fibronectin was slightly increased ( S1B Fig ) . In addition to ECM proteins , other host cell surface proteins may be involved in Leptospira infection or pathogenicity . To test this hypothesis , we examined intercellular adhesion molecules ( ICAMs ) and other cell surface receptors . ICAMs belong to the immunoglobulin superfamily , and participate in inflammatory responses [49] . Compared to uninfected endothelial cells , the signal intensities of ICAM-1 and ICAM-2 were elevated by infection with either L . interrogans sv . Copenhageni or L . biflexa sv . Patoc ( Fig 2A and 2B , S2A Fig ) . The signal increase was more apparent with Copenhageni infection , especially for ICAM-1 in both cell types: a 4 . 5-fold increase in HMEC-1 and a 12-fold in HDLEC ( Fig 2A , Table 2 ) . For the ICAM-2 signal , Copenhageni infection caused an increase of 1 . 6- to 2 . 2-fold in endothelial cells ( Fig 2B , S2A Fig ) . Patoc infection demonstrated an intermediate increase in both ICAM-1 and ICAM-2 ( Fig 2A and 2B , S2A Fig ) . CD36 is a fatty acid/scavenger receptor , and is involved in microvascular endothelial cell migration and metastasis [50 , 51] . Leptospira infection increased the signal intensity of CD36 by 1 . 3- to 1 . 8-fold , and was slightly higher in Copenhageni-infected than Patoc-infected HMEC-1 ( Fig 2C ) . Infection also elevated the CD36 signal in HDLEC , although the difference between Copenhageni- and Patoc was miniscule , and mainly caused by the increase in punctate morphology ( Fig 2C , S2B Fig ) . These data suggest that this CD36 phenotype is induced by Leptospira infection with both pathogenic and non-pathogenic strains . Vascular endothelial growth factor-receptor 2 ( VEGF-R2 ) is another Leptospira-influenced cell surface protein . This VEGF-specific receptor is involved in the proliferation of vascular endothelial cells and the regulation of the endothelial barrier function [52] . Leptospira infection with both Copenhageni and Patoc elevated the VEGF-R2 signal , more so in HMEC-1 ( 1 . 5- to 2-fold ) than the slight increase ( 1 . 1- to 1 . 4-fold ) in HDLEC ( Fig 2D , S2C Fig ) . Again , these signal elevations in VEGF-R2 were induced by both pathogenic and non-pathogenic Leptospira species . Although the screening of cell surface proteins was originally our focus , we also examined several intracellular proteins . We found three intracellular host proteins that were affected by Leptospira infection ( Table 2 ) . One protein was vascular endothelial growth factor ( VEGF ) , which plays roles in the control of vascular endothelial cell proliferation and vascular permeability [53] . Infection with either L . interrogans sv . Copenhageni or L . biflexa sv . Patoc slightly elevated the fluorescence signal of VEGF by 1 . 1- to 1 . 9-fold with puncta formation ( Fig 3A ) . The signal increase caused by the pathogenic Copenhageni was higher than by the nonpathogenic Patoc in HMEC-1 but there was no difference between the changes caused by the two Leptospira strains in HDLEC ( Fig 3A ) . A second Leptospira-affected protein , the small GTPase RhoA , is an important molecule that regulates the assembly of the actin cytoskeleton and the remodeling of cell junction proteins . RhoA activity accompanied by actin remodeling can lead to a loss of endothelial barrier integrity [52 , 54–56] . In both endothelial cell types we tested , the fluorescence intensity of small GTPase RhoA was slightly elevated by Copenhageni ( 1 . 3- to 1 . 4-fold ) and Patoc ( 1 . 1- to 1 . 2-fold ) infection ( Fig 3B ) . The signal increase was slightly higher when cells were infected with Copenhageni than with Patoc ( Fig 3B ) . An increase in punctate morphology of the RhoA signal was also observed in infected cells , but not specific to the Leptospira species ( Fig 3B ) . Another intracellular protein affected by Leptospira infection is integrin-linked kinase ( ILK ) . ILK associates with integrins as a regulator of integrin-mediated signaling , correlating with multiple cellular functions such as cell proliferation , migration , adhesion , and vascular integrity [57–59] . The intensity of ILK signal was slightly higher ( 1 . 1- to 1 . 5-fold ) with puncta formation when HMEC-1 were infected with leptospires regardless of pathogenic status ( Fig 3C ) . Overall , Leptospira-mediated changes were found in intracellular host proteins that have roles in the regulation of cell proliferation , endothelial barrier integrity , and actin remodeling , but differences between the pathogenic and non-pathogenic strains were not always apparent . In endothelial cells , intercellular connections are formed through multiple adhesive structures , regulating the passage of blood constituents and circulating cells to the underlying tissues [60 , 61] . Pathological conditions of endothelial paracellular permeability lead to severe or fatal organ dysfunction [60 , 61] , similar to the symptoms in severe leptospirosis patients . To determine the effect of Leptospira infection on cell-cell junctions , we examined the transmembrane proteins and cytosolic adaptor proteins of three major intercellular junction types: 1 ) adherens junction , 2 ) tight junction , and 3 ) gap junction . L . interrogans sv . Copenhageni infection apparently disrupted multiple proteins of the adherens junction , the tight junction protein ZO-1 , and the gap junction protein connexin ( Figs 4 and 5 ) . The adherens junction and tight junction proteins directly or indirectly interact with actin filaments to stabilize the cellular structure and cell junctions [72 , 73] . To examine if Leptospira infection influences actin filaments ( microfilaments ) in endothelial cells , filamentous actin was labeled with Alexa Fluor-conjugated phalloidin . In permeabilized , uninfected HMEC-1 , actin filaments were visualized as net-shaped structures , spreading ubiquitously in the cell ( Fig 6 , uninfected ) . When this cell type was infected with Copenhageni , the signal intensity of actin filaments decreased to ~50% of uninfected cells ( Fig 6 , Table 2 ) . L . biflexa sv . Patoc infection demonstrated no reduction in signal intensity ( Fig 6 ) . In another type of endothelial cells , HDLEC , the morphology of actin filaments appeared well-organized , with long straight filaments rather than the net-shaped morphology ( Fig 6A , uninfected ) . In contrast to the reduction of the actin signal in Copenhageni-infected HMEC-1 , the signal decrease in HDLEC was minor ( Fig 6 ) . Instead , Copenhageni infection induced a morphological rearrangement of actin filaments: intense localization of filamentous actin at the cell periphery and reduction of stress fibers inside the cell ( Fig 6 , Table 2 ) . Patoc infection induced slightly more stress fibers but did not affect the overall morphology and signal intensity of filamentous actin in HDLEC ( Fig 6 ) . Throughout our screening of host proteins , only this protein demonstrated host cell-type specific changes in morphology and phenotype . We also examined another cytoskeletal structure , the microtubule , by immune-labeling the most critical protein , alpha-tubulin ( Table 1 ) . There was no detectable change in morphology or signal-intensity in HMEC-1 or HDLEC infected with either Copenhageni or Patoc ( S8 Fig ) . These data suggest that the pathogenic Leptospira specifically modifies the actin filaments of the cytoskeleton .
To investigate the pathogenic effects of Leptospira infection of human endothelial cells , we utilized immunofluorescence microscopy to screen for changes in host protein abundance and distribution . Seventeen proteins indicated minor to major changes in Leptospira-infected endothelial cells . These 17 proteins are classified into five biological groups: 1 ) extracellular matrix , 2 ) intercellular adhesion molecules and cell surface receptors , 3 ) intracellular proteins , 4 ) cell-cell junction proteins , and 5 ) a cytoskeletal protein . The most prominent phenotype of infection with pathogenic L . interrogans sv . Copenhageni was the dramatically reduced multiple adherens junction proteins and one of the tight junction proteins ( ZO-1 ) , a gap junction protein ( connexin ) as well as filamentous actin ( summarized in Table 2 ) . Infection with Leptospira , regardless of the strain’s pathogenicity or ability to harm host cells , increased the signal intensity of some of the ECM proteins , ICAMs , cell surface receptors , and intercellular proteins ( Table 2 ) . In general , the signal increase was more intense when endothelial cells were infected with pathogenic L . interrogans sv . Copenhageni than the saprophyte strain , L . biflexa sv . Patoc ( Table 2 ) . Among ECM proteins , collagen type IV , decorin , and laminin were influenced by Leptospira infection ( Fig 1 ) . It has been reported that outer membrane protein ( s ) of pathogenic Leptospira can increase the production of an ECM protein , collagen type IV [76] . In this study , the signal elevations of collagen and decorin were caused by both pathogenic and non-pathogenic leptospires ( Fig 1A and 1B ) . Kassegne et al . reported that L . interrogans possesses a collagenase , which is involved in the invasion and transmission of the pathogenic species [77] . Morphological changes we observed , formation of puncta as a result of infection ( Fig 1A ) , may be a result of protein degradation by collagenase activity , which could expose epitopes for antibody binding , increasing the fluorescence signal . For laminin , the structural rearrangement was specifically caused by the pathogenic Copenhageni ( Fig 1C ) . The net-shape structure of laminin appeared rolled up , forming thick bundles with little change in total signal intensity , suggesting that this phenotype could be a secondary effect of the Copenhageni-mediated disassembly of adherens junctions . Leptospira infection influenced several cell-surface proteins/receptors , ICAM-1 , ICAM-2 , CD36 , and VEGF-receptor 2 ( Fig 2 , S2 Fig , Table 2 ) . We observed a significant signal increase of ICAMs in both HMEC-1 and HDLEC when infected with leptospires and the increase was more prominent with Copenhageni in HMEC-1 ( Fig 2 ) . Because ICAMs cluster at intercellular junctions distinct from the adherens junction and tight junction [61] , the elevation of ICAMs is likely an independent phenomenon from the disruption of adherens junctions by pathogenic Leptospira . In pulmonary leptospirosis patients , an increase in ICAM-1 expression was detected in the alveolar septa and pulmonary vessels [7] . However , it was also reported that there was no significant change in ICAM-1 cell surface expression in HUVEC after 24 h and 48 h infection as detected by horseradish peroxidase reaction [78] . This difference may be due to the sources of the cells or specific experimental conditions , as other work has shown that leptospiral lipopolysaccharide ( LPS ) and outer membrane proteins increase ICAM-1 expression in HUVECs [38 , 39] . During early stages of leptospirosis , leptospiral LPS and outer membrane lipoproteins induce inflammation primarily via Toll-like receptor 2 ( TLR2 ) and in some experiments , via TLR4 activation [33 , 34 , 79–81] . Interestingly , pathogenic Leptospira infection induces pro-inflammatory reactions in human ( susceptible to leptospirosis ) but activates anti-inflammatory pathways in mice ( not susceptible to clinical leptospirosis ) [18 , 36] . Thus , the phenotype of ICAM increase in human endothelial cells may be a result of inflammatory reaction induced by TLR-mediated signaling pathways and other pro-inflammatory responses . The fatty acid/scavenger receptor CD36 is involved in the regulation of microvascular endothelial cell migration and is implicated as having a role in inflammation [50 , 51 , 82] . CD36 is also known to interact with collagens [83] . Since we observed Leptospira-mediated changes in collagen type IV and type VI ( Fig 1A and S1A Fig ) , the signal increase phenotype of CD36 that we observed could be induced by multiple factors , such as changes in collagen or inflammatory signaling . Another receptor , VEGF-R2 , specifically reacts to VEGF in controlling the growth of vascular endothelial cells . Another function of VEGF-R2 is that this receptor interacts with VE-cadherin , which physically limits the internalization of VE-cadherin from adherens junctions [52] . Overproduction of VEGF-R2 might inhibit the internalization of VE-cadherin , but we observed a reduction in VE-cadherin signal without apparent internalization ( Fig 4A and S3C Fig ) . The elevation of the VEGF-R2 signal may be a response to leptospire-mediated VEGF increase rather than a direct response to Leptospira contact with endothelial cells . Leptospira infection increased the fluorescence signal and puncta formation of three intracellular proteins , VEGF , RhoA , and ILK ( Fig 3 , Table 2 ) . In addition to its function in vascular endothelial growth , VEGF promotes vascular permeability via the phosphorylation and endocytosis of VE-cadherin; this modulation is reversible [61] . There was no detectable internalization of VE-cadherin ( S3C Fig ) and the signal increase of VEGF was not specific to the pathogenic strain ( Fig 3A ) , implying that any involvement of VEGF in pathogenicity is less likely . ILK is a regulator of integrin-mediated signaling to regulate cell migration , adhesions , and vascular integrity [57–59] . The Leptospira-mediated changes were statistically significant but modest and also detected in cells infected both Copenhageni and Patoc strains ( Fig 3C ) . Endothelial permeability is controlled by the opening and closing of cell-cell junctions via the rearrangement of junction proteins and cytoskeletal proteins [61] . To regulate the adherens junction organization and endothelial permeability , some small GTPases are involved [53 , 84] . For example , the small GTPase RhoA controls the endothelial barrier integrity via remodeling of the actin cytoskeleton and of junction proteins [52 , 54–56] . In this study , Leptospira infection altered the actin cytoskeleton along with a moderate elevation of the RhoA signal ( Figs 6 and 3B ) . The increase of RhoA signal was not specific to Copenhageni infection , though the signal intensity was higher than with Patoc inoculation ( Fig 3B ) . These data suggest that the involvement of these intracellular proteins in the Leptospira pathogenicity , especially in disrupting the endothelial integrity , is relatively minor . In this study , the most prominent L . interrogans pathogenic phenotype was the disruption of adherens junctions ( Fig 4 , S4 Fig ) . The adherens junction proteins , VE-cadherin , p120 catenin , alpha-catenin , and beta-catenin , showed drastically reduced immunofluorescence signals specifically at the cell-cell junctions ( Fig 4 , S4 Fig; also see Fig 7 for structural features ) . Because Copenhageni-infection did not disturb occludin and claudin ( tight junction markers , S7 Fig ) , the VE-cadherin/catenin-complex at the adherens junction is likely to be the main target of pathogenic Leptospira species in endothelial cells ( Fig 7 ) . VE-cadherin is essential for adherens junction formation and barrier maintenance in endothelial cells , playing a critical role in vascular morphogenesis , especially remodeling and maturation [71 , 85] . The administration of the anti-VE-cadherin antibody BV13 redistributes VE-cadherin molecules at adherens junctions in cultured endothelial cells and increases vascular permeability in heart and lungs of mice [60] . VE-cadherin-associated catenins are also important for the formation of the dynamic endothelial barriers . For example , VE-cadherin/beta-catenin complexes are involved in the regulation of endothelial cell survival , and in VE-cadherin mutant cells , beta-catenin was not localized at intercellular junctions [71] . Inactivation of the beta-catenin gene disrupted the cell junctional organization by reduction in alpha-catenin expression and cell-adhesion strength , leading to hemorrhagic vessels [72] . Also , p120 catenin is involved in maintaining VE-cadherin expression [68] . Thus , VE-cadherin and catenins are interconnected physically and functionally to regulate the expression and interaction of these proteins , maintaining the stability and flexibility of endothelial junctional barriers required for normal biological function . Our data indicate that pathogenic Leptospira strains target these adherens junction proteins important for the endothelial barrier integrity and disassemble the junction-protein complex ( Fig 7 ) . In contrast to the drastic disturbance of the adherens junction ( Fig 4 , S4 Fig ) , there were no detectable changes in the tight junctional transmembrane proteins claudin 5 and occludin in endothelial cells infected with L . interrogans sv . Copenhageni ( S7 Fig ) . These tight junction markers are less critical to the vascular barrier integrity since the deletion of the genes encoding claudin 5 or occludin does not influence on the vascular morphology and barrier function in mice [86 , 87] . In our experiments , only ZO-1 , a cytosolic peripheral protein , was apparently mislocalized from tight junctions ( Fig 5A ) . Reduction of ZO-1 was previously observed in Leptospira-infected HUVEC [24] . ZO-1 is a tight junction protein when intercellular junctions are mature but localizes at adherens junctions at an early stage of cell-cell contact [62 , 73 , 88] . In the immature cell junctions , ZO-1 can directly associate with alpha-catenin and the actin cytoskeleton [67 , 89] and indirectly associates with beta-catenin [90] . In the mature tight junction , ZO-1 indirectly influences the endothelial integrity via association with claudin , occuldin , and actin filaments [60 , 75] . The published information and our data imply that Leptospira primarily disrupts the adherens junctions , resulting in mislocalization of one of the actin-binding proteins , ZO-1 ( Fig 7 ) . A gap junction protein , connexin 43 , was also mislocalized and showed reduced signal at cell junctions in endothelial cells when infected with pathogenic Copenhageni ( Fig 5B ) . The non-pathogenic Patoc strain caused translocation of connexin 43 from the cell periphery to intracellular locations without losing signal intensity ( Fig 5B ) . Connexin 43 associates with a variety of proteins located at adherens junctions and tight junctions , the cytoskeleton , and actin-binding proteins , including p120 catenin , beta-catenin , and ZO-1 [74 , 75] . Moreover , the gap junction is not directly involved in endothelial permeability [62] . These data imply that the gap junction is not a primary target of pathogenic Leptospira and that the phenotype we observed is likely induced as a secondary effect following the disruption of the adherens junction . Infection with the pathogenic Copenhageni demonstrated cell-type-specific phenotypes in the actin cytoskeleton: 1 ) reduction of the actin signal in HMEC-1 and 2 ) bundling and rearrangement of filamentous actin structure in HDLEC ( Fig 6 ) . Actin filaments physically interact with multiple cell-junction proteins , which regulates the dynamic rearrangement of the actin-filament structure [62] . For instance , alpha-catenin interacts with either cadherin/beta-catenin complex or actin filaments , regulating actin assembly and organization [67 , 69] . The inactivation of the beta-catenin gene influences the morphology of actin filaments in endothelial cells [72] , and ZO-1 regulates the cortical cytoskeleton at cell junctions [73] . The phenotypes of Copenhageni infection , a decrease in filamentous actin in HMEC-1 and translocation of the bundled-actin filaments to the cell periphery of HDLEC ( Fig 6 ) , may be induced by Copenhageni-mediated disruption of the VE-cadherin-catenin complex . One of the functions of filamentous actin is stabilizing or reorganizing the intercellular junctions through interacting with cell-junctional proteins [62] . The cadherin-catenin complex is known to dynamically influence the actin cytoskeleton and vice versa: filamentous actin is necessary for the regulation of endothelial opening/closing in addition to the stabilization of cell-junctions [61 , 62] . We considered the possibility that filamentous actin is the primary target of pathogenic Leptospira infection , but the inhibition of typical actin distribution or mobilization at cell-cell junctions by cytochalasin D or jasplakinolide do not influence the dynamics of cadherin and alpha-catenin [67] , suggesting that filamentous actin is unlikely to be the primary target of pathogenic Leptospira . In physiologic conditions in vivo , dynamic and transient remodeling of intercellular junctions is well-controlled and critical to cellular maintenance , especially in endothelial cells [62 , 91] . However , drastic and irreversible changes in endothelial junctions contribute to pathological endothelial permeability and leakage as well as vascular network disruption [62] . Miyahara et al . identified intact cell attachment with some disturbance of intercellular junctions in hepatocytes of pre-icteric hamsters , with cell detachment plus disrupted junctional association in icteric hamsters [23] . In leptospirosis patients , proinflammatory response and vascular damage are pathologic features of leptospirosis-associated pulmonary hemorrhage syndrome or acute lung injury [11 , 12 , 25] . Thus , in the later stage of severe leptospirosis , Leptospira infection and detrimental inflammatory responses , independent of TLR activation [33] , overwhelm the cellular maintenance system , leading to devastating damage to cell-cell junctions and vascular systems of the host . We used cultured human endothelial cells to investigate how Leptospira may lead to endothelial permeability and , by inference , vascular damage seen in human patients and susceptible animals . Our study demonstrated that the primary targets of L . interrogans are intercellular junctions , primarily adherens junctions . Other host proteins affected by L . interrogans infection may be indirectly impacted by the damage to a modification of the primary targets . The changes in host proteins that were impacted by non-pathogenic Patoc , though not as robust as those impacted by Copenhageni , may be a consequence of pro-inflammatory responses induced by Leptospira LPS , cell-surface proteins , or secreted proteins . Our systematic analyses of host proteins in Leptospira infected-human endothelial cells demonstrated pathogen-specific phenotypes in the adherens junction , filamentous actin and actin-associated proteins . Several phenotypes were observed with infection with either the pathogen or the non-pathogen in multiple biological groups . These data suggest that this zoonotic agent may damage endothelial cells via multiple cascades or pathways , potentially leading to the increased vascular permeability followed by severe illness in vivo . In addition , morphological and quantitative analyses of infected human cells by immunofluorescence microscopy constitute a reliable method to investigate the pathogenicity and biological functions of Leptospira strains and specific proteins in vitro . Further work based on these results will contribute to our understanding of pathophysiological mechanisms of Leptospira infection . | Pathogenic Leptospira causes the life-threatening infection called leptospirosis worldwide . Symptoms of leptospirosis range from mild illness to severe illness such as organ damage , meningitis , and fatal hemorrhagic disease . Invasive species of Leptospira spread to multiple tissues , and damage the linings of blood vessels . Despite the burden in humans and animals , how Leptospira bacteria cause damage remains to be determined . In this study , human endothelial cells were infected with L . interrogans ( pathogen ) or L . biflexa ( non-pathogen ) to investigate changes to host cell proteins . Among those analyzed , 17 proteins from five biological classes demonstrated distinctive changes upon infection with Leptospira . Infection with the pathogenic strain most profoundly disturbed the adherens junction group of proteins that hold neighboring cells together . In addition to the changes in cell-cell junctions , Leptospira infection , regardless of pathogenic status , influenced other host proteins belonging to multiple biological classes . Our data suggest that L . interrogans may damage endothelial cells via multiple cascades or pathways . The damage may include endothelial barrier disruption and inflammation , potentially leading to leaky blood vessels in patients . This work contributes to our understanding of how of Leptospira causes widespread , disseminated infection and disease in humans and animals . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"cell",
"physiology",
"cell",
"motility",
"medicine",
"and",
"health",
"sciences",
"leptospira",
"actin",
"filaments",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"endothelial",
"cells",
"microbiology",
"junctional",
"complexes",
"epithelial",
"cells",
"tight",
"junctions",
"bacteria",
"bacterial",
"pathogens",
"animal",
"cells",
"proteins",
"medical",
"microbiology",
"microbial",
"pathogens",
"biological",
"tissue",
"biochemistry",
"cell",
"biology",
"anatomy",
"epithelium",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"leptospira",
"interrogans",
"organisms",
"extracellular",
"matrix",
"proteins"
] | 2017 | Leptospira interrogans causes quantitative and morphological disturbances in adherens junctions and other biological groups of proteins in human endothelial cells |
Integration and down-regulation of cell growth and differentiation signals rely on plasma membrane receptor endocytosis and sorting towards either recycling vesicles or degradative lysosomes via multivesicular bodies ( MVB ) . In this process , the endosomal sorting complex-III required for transport ( ESCRT-III ) controls membrane deformation and scission triggering intraluminal vesicle ( ILV ) formation at early endosomes . Here , we show that the ESCRT-III member CHMP1B can be ubiquitinated within a flexible loop known to undergo conformational changes during polymerization . We demonstrate further that CHMP1B is deubiquitinated by the ubiquitin specific protease USP8 ( syn . UBPY ) and found fully devoid of ubiquitin in a ~500 kDa large complex that also contains its ESCRT-III partner IST1 . Moreover , EGF stimulation induces the rapid and transient accumulation of ubiquitinated forms of CHMP1B on cell membranes . Accordingly , CHMP1B ubiquitination is necessary for CHMP1B function in both EGF receptor trafficking in human cells and wing development in Drosophila . Based on these observations , we propose that CHMP1B is dynamically regulated by ubiquitination in response to EGF and that USP8 triggers CHMP1B deubiquitination possibly favoring its subsequent assembly into a membrane-associated ESCRT-III polymer .
Endocytosis of activated plasma membrane receptors is induced by ubiquitin linkage and plays a crucial role in cell signaling modulation through their subsequent sorting to either recycling vesicles or to lysosomes via multivesicular bodies ( MVBs ) . For example , endocytosis of the Epidermal Growth Factor Receptor ( EGFR ) represents the major mechanism of long-term attenuation of EGF signaling [1 , 2] . In this process , the conserved ESCRT ( Endosomal Sorting Complex Required for Transport ) machinery drives endosomal membrane deformation and scission leading to the formation of intraluminal vesicles ( ILVs ) within MVBs [3–6] . The ESCRT machinery consists of five complexes called ESCRT-0 , I , II , III and VPS4 that are all required for MVB biogenesis: ESCRT-0 , I and II cluster the internalized ubiquitinated cargoes and may initiate membrane bending while ESCRT-III and VPS4 are responsible for membrane fission [6–8] . The human ESCRT-III family is composed of 11 proteins termed CHarged Multivesicular Proteins ( CHMP1A , B , CHMP2A , B , CHMP3 , CHMP4A , B , C , CHMP5 , CHMP6 and IST1 ) which are recruited to membranes [9 , 10] . They are found in a closed auto-inhibited conformation in the cytosol and activation is thought to displace a C-terminal region from the core helical hairpin [11–15] . Activated ESCRT-III proteins polymerize as homo- or heteromers adopting spiral structures [13 , 16–21] . CHMP1B polymerization requires extensive conformational changes from the closed conformation to the open polymer conformation . The latter is stabilized by domain swapping of the C-terminal α-helices 4 and 5 and the extension of α-helix 2 by α-helix 3 [22] . Similar conformational changes have been reported for yeast Snf7 ( orthologous to CHMP4 ) indicating common principles for activation and polymerization of the CHMP family of proteins [23 , 24] . In the particular case of IST1 however , in vitro polymerization in a closed conformation has been observed in association with open CHMP1B , resulting in a polymer composed of an external layer of closed IST1 and an internal layer of open CHMP1B [22] . Regulation by phosphorylation has been reported for CHMP1A and CHMP4C [25 , 26] yet the regulatory mechanisms of ESCRT-III activation and polymerization on membranes in vivo remain poorly understood . The regulatory C-terminal part of ESCRT-III proteins contains one or two MIT interacting motifs ( MIM ) that recruit partners possessing a Microtubule Interacting and Trafficking ( MIT ) domain including the Ubiquitin Specific Protease 8 ( USP8/UBPY ) or the Associated Molecule with SH3 domain of STAM ( AMSH ) [27–32] . The interaction between the MIT domain of USP8/UBPY and the MIM domain of CHMP1B is required for USP8 association with endosomal membranes and EGFR sorting [30] . In fact , these two Deubiquitinating ( DUBs ) enzymes are strongly associated with endosomes where they regulate the stability and ubiquitination status of ESCRT-0 members STAM and/or Hrs [30 , 33–39] as well as of internalized receptors [27 , 28 , 30 , 38–41] . Particularly , both AMSH and USP8/UBPY deubiquitinate the Epidermal Growth Factor Receptor ( EGFR ) , a member of the receptor tyrosine kinase family ( RTK ) , by acting at the level of the plasma membrane and/or of the endosome where deubiquitination of EGFR precedes its incorporation into MVBs [27 , 33 , 34 , 37 , 41–49] . In addition to EGFR , USP8 deubiquitinates numerous plasma membrane receptors , making this enzyme a promising target in cancer therapy to overcome chemoresistance associated with RTK stabilization [50 , 51] . Furthermore , gain of function mutations of USP8 have been found in microadenoma of patients with Cushing’s disease , a rare disease where the secretion of large amounts of adenocorticotrophin hormone ( ACTH ) by pituitary corticotroph adenomas results in excess of glucocorticoids and hypercortisolism putatively due to defective EGFR sorting [52–55] . In the present study , we describe that USP8/UBPY also targets the ESCRT-III machinery . Indeed , we show that CHMP1B is regulated by linkage of ubiquitin and its subsequent removal by USP8/UBPY . The physiological importance of CHMP1B ubiquitination is highlighted by the observation that it is inducible within minutes of EGF stimulation and results in transient accumulation of ubiquitinated forms of CHMP1B on membranes . Moreover , mutation of four lysine residues in , or close to , a flexible loop of CHMP1B makes the protein non-functional in EGFR trafficking in human cells or during Drosophila wing morphogenesis . Thus , our findings establish a new ubiquitin-dependent mechanism controlled by USP8/UBPY ( hereafter designed as USP8 ) that may act as a check-point for the spatial and temporal control of CHMP1B activity such as polymerization at endosomal membranes .
CHMP1B was previously shown to interact with USP8 in both co-immunoprecipitation and yeast two-hybrid experiments [30 , 56] . In order to further map the domains of CHMP1B implicated in this interaction , GFP-tagged constructs of full length or helical fragments of CHMP1B ( Fig 1A ) [57] were expressed in HEK293T cells and tested for their ability to co-immunoprecipitate with Flag-USP8 . Full-length CHMP1B and α-helices 4 , 5 and 6 interacted with Flag-USP8 in this assay ( Fig 1B ) . Interaction of CHMP1B was maintained with a catalytically-inactive version of the enzyme ( USP8C748A ) ( Fig 1C ) . These results map the interacting region of USP8 to CHMP1B residues 105 to 199 , similar to the CHMP1B-Spastin interaction [58] . Next , we designed a Venus complementation assay in which USP8 was fused to the N-ter of Venus ( VN-USP8 ) and full length or truncated CHMP1B constructs to its C-ter ( VC-CHMP1B ) ( Fig 1D ) . The expression of corresponding constructs was verified by immunoblot ( S1 Fig ) . In this assay , a cytoplasmic signal was clearly observed upon co-expression of VN-USP8 and VC-CHMP1B indicating a direct interaction between the two partners in vivo ( Fig 1E and 1E’ ) . Consistent with the above results , truncation of the helices situated in the C-ter part of CHMP1B led to a decrease ( Δα6 and Δα5 , 6 constructs ) or a total loss ( Δα4 , 5 , 6 ) of fluorescence ( Fig 1E and 1E’ ) . Cells transfected with VN-USP8 and VC-CHMP1B were stained with a set of endosomal markers revealing a strong overlap of the Venus signal with Lamp1 , a marker of late endosomes/lysosomes , but not with the early endosomal marker EEA1 ( Fig 1F and 1F’ ) . Taken together , these observations confirm that the two proteins USP8 and CHMP1B are part of a same protein complex in living cells , interacting at the level of the late endosome , which is consistent with the known function of CHMP1B in the multivesicular body biogenesis and sorting of receptors . We next investigated whether CHMP1B is ubiquitinated in cells . To this end , the GFP-CHMP1B protein was expressed in HEK293T cells together with HA-ubiquitin ( HA-Ub ) [59] and immunoprecipitated from cell lysates with an antibody against GFP . To specifically detect ubiquitin moieties covalently linked to CHMP1B and not to putative partners , highly stringent conditions were used for immunoprecipitation . Western blot of GFP immunoprecipitated product using anti-HA antibodies revealed the presence of a major Ub-CHMP1B product migrating at 70kDa on SDS-PAGE , putatively corresponding to a mono- or di-ubiquitinated form of the recombinant protein , as well as higher molecular weight ( mw ) species that may correspond to multi-mono- or poly-ubiquitinated forms of CHMP1B ( Fig 2A , “WT” ) . Accordingly , previous whole-proteome analyses revealed the existence of ubiquitin linkage at multiple sites of the CHMP1B protein [60–62] . We then generated a CHMP1B mutant construct in which four lysine residues exposed to the solvent were replaced by arginine residues ( CHMP1B-4K>R ) ( Fig 1A , S2A–S2H Fig ) . This mutant displayed a strong reduction of ubiquitin linkage compared to wild-type CHMP1B ( Fig 2A and 2A’; S2I Fig ) . We also observed a slight reduction of ubiquitin-linked forms of CHMP1B carrying a single K>R substitution in the case of either the K87>R or the K90>R but not in the case of the K42>R or the K59>R single substitutions ( S2I Fig ) . Residual ubiquitination observed with the CHMP1B-4K>R construct may result from other ubiquitination sites within the protein [60–62] . These results indicate that GFP-CHMP1B is ubiquitinated in human cells and that ubiquitin linkage occurs mostly at lysine residues K87 or K90 . Remarkably , these two lysine residues are located in the flexible linker between α2 and α3 which becomes helical in the polymer structure ( S2A–S2H Fig ) [22] . We then tested whether ubiquitinated CHMP1B is a target of the ubiquitin hydrolase activity of USP8 . To this end , HEK293T cells were co-transfected with HA-Ub and GFP-CHMP1B together with Usp8 silencing ( shUsp8 ) or wild-type or mutated USP8 expressing constructs [63] . Silencing of Usp8 was only partial and resulted in a slight increase of the ubiquitinated pool of GFP-CHMP1B ( Fig 2B and 2B’ ) . In contrast , the expression of the wild-type or the constitutively active form USP8S680A , but not of the catalytic mutant USP8C748A , caused a strong reduction of the Ub-CHMP1B pool ( Fig 2C and 2C’ ) . We repeated the experiment using antibodies specifically directed against K48-linked ubiquitin polymers showing that GFP-CHMP1B is unlikely to be modified by this type of chains or at very low level ( S2J Fig ) . Reinforcing this observation , inhibiting the proteasomal activity did neither increase the amount of ubiquitinated GFP-CHMP1B nor stabilize the protein ( S2K Fig ) . In these experiments again , Ub-CHMP1B forms were strongly reduced by expressing USP8 or USP8S680A , but not the catalytic mutant USP8C748A ( S2J and S2K Fig ) . Our results thus strongly suggest that USP8 deubiquitinates CHMP1B . The analysis of the whole cell lysate from HEK293T or HeLa cells with an anti-CHMP1B antibody revealed three major bands migrating at 25–28 kDa , 55 kDa and 200 kDa ( Fig 3A and S3A Fig ) . These bands were strongly diminished upon silencing of CHMP1B using two independent shRNAs ( S3A Fig ) . These results indicate that endogenous CHMP1B is present as distinct species corresponding to monomers and SDS-PAGE-resistant putative dimers and polymers . Longer heat denaturation of the samples resulted in a loss of the 200 kDa band and an increase of the 55 kDa band , supporting the hypothesis that the high mw species corresponds to a SDS-PAGE resistant polymeric form of CHMP1B ( S3B Fig ) . In contrast to the endogenous protein , the recombinant GFP-tagged CHMP1B was only detected as monomeric form possibly due to the presence of the N-terminal GFP tag that may perturb CHMP1B polymerization ( see above , Fig 2A ) . Then , HEK293T cell lysates were separated on a sucrose gradient as a first step to enrich samples in CHMP1B polymers . Western blot analysis confirmed the presence of polymeric SDS-PAGE resistant CHMP1B in the 20 and 30% sucrose fractions while the putative dimers were present in the upper fractions ( 0% and 10% ) and monomers were not detected by this method ( Fig 3B ) . We then analyzed the 30% sucrose fraction by size exclusion chromatography followed by western blot analysis . This revealed the presence of CHMP1B in fractions 6 to 9 ( elution volume 9 . 5 to 11 ml ) ( Fig 3C ) . Because two marker proteins of 669 kDa ( thyroglobubin ) and 158 kDa ( γ-globulin ) eluted at 9 . 8 and 13 . 2 ml , respectively , we conclude that CHMP1B is part of a ~500 kDa complex . Probing the same fractions with the anti-IST1 antibody showed further that endogenous IST1 is also present in this complex ( Fig 3C ) . Taken together , our results indicate that endogenous CHMP1B exists as SDS-PAGE resistant polymers that are part of a larger complex that also contains IST1 . We next analyzed the ubiquitination profile of endogenous CHMP1B . Ubiquitinated proteins from HEK293T cells were immunoprecipitated using the anti-ubiquitin FK2 antibody and further analyzed by western blot using the anti-CHMP1B antibody . As shown above , the same three forms of CHMP1B could be detected in the whole cell lysate ( Fig 3D , WCL ) . In the pool of ubiquitinated proteins , CHMP1B immunoblotting revealed the presence of a major ubiquitinated form of CHMP1B migrating at ~70 kDa that may correspond to an ubiquitinated dimer of CHMP1B ( Fig 3D , IP FK2 ) while the minor band detected at 35 kDa might be non-specific ( see S3C Fig ) . In contrast , the polymeric form of CHMP1B was not found in the ubiquitinated fraction ( Fig 3D , IP FK2 ) . Thus , polymers of CHMP1B that are part of the ESCRT-III complex with IST1 are most likely devoid of ubiquitin . In parallel , endogenous CHMP1B was immunoprecipitated from the HEK293T cell extract and detected by immunoblotting with anti-CHMP1B . Likewise , the monomers , the putative dimers and the polymers could be detected in the whole cell lysate ( Fig 3D , WCL ) . Interestingly , only the polymers and the ubiquitinated , but not the non-ubiquitinated , forms of CHMP1B , were found in the CHMP1B fraction immunoprecipitated from native cell extracts ( Fig 3D , IP CHMP1B ) . Since we used a polyclonal human anti-CHMP1B antibody raised against a peptide covering residues 35 to 84 ( Fig 3E ) , the epitope ( s ) recognized by this antibody might be completely or partially hidden by the auto-inhibitory helix 6 in the monomeric closed conformation [22] thereby preventing its recognition and immunoprecipitation by the antibody from native cell lysates . We thus suggest that these epitope ( s ) are exposed for antibody recognition in polymeric as well as in ubiquitinated forms of CHMP1B while they are masked in non-ubiquitinated ones . Taken together , our results show the existence of endogenous ubiquitinated forms of CHMP1B , which may correspond to ubiquitinated dimers , and that CHMP1B polymeric forms are not ubiquitinated . Finally , we observed that the ubiquitination level of endogenous CHMP1B was higher in partially Usp8-silenced cells compared to control cells , strengthening the hypothesis that USP8 deubiquitinates CHMP1B ( Fig 3F ) . Given the role of the ESCRT machinery in the downregulation of a vast array of receptors in response to their ligands , including EGFR and the pro-inflammatory IL1R ( Interleukine 1 Receptor ) [64 , 65] , we tested the effect of EGF and IL1β on CHMP1B ubiquitination when added at doses known to induce the internalization of their respective receptors [64 , 65] . In cells co-expressing GFP-CHMP1B and HA-Ub , a transient increase of the amount of ubiquitinated GFP-CHMP1B was observed at 5 min of stimulation with EGF ( Fig 4A ) . This timing coincides with the onset of the interaction between EGFR and the ESCRT proteins [45] . In the case of IL1β treated cells , a transient accumulation of ubiquitinated GFP-CHMP1B was observed at 10 min of stimulation ( S4A Fig ) . These experiments show that ubiquitination of CHMP1B is dynamically regulated in response to EGF or cytokine stimulation . We then tested the effect of EGF stimulation on the ubiquitination profile of endogenous CHMP1B . HEK293T cells were treated with 100 ng/ml of EGF and ubiquitinated proteins were immunoprecipitated using the FK2 antibody at different time points . A transient , although modest , increase of the amount of endogenous Ub-CHMP1B putative dimers was observed at 5 min post-EGF stimulation in the FK2 immunoprecipitate from whole cell lysates ( Fig 4B , blot ( 1 ) ) . Fractioning of cytosolic ( Fig 4B , blot ( 2 ) ) versus membrane ( Fig 4B , blot ( 3 ) ) fractions prior to immunoprecipitation revealed that the pool of ubiquitinated CHMP1B in response to EGF was strongly enriched in the membrane fraction from 5 to 15 min of EGF stimulation ( Fig 4B , blot ( 3 ) ) . Analysis of the corresponding cell fractions prior to immunoprecipitation revealed no significant change in the CHMP1B profile in the whole cell lysates or the cytosolic fraction . In the membrane fraction however , the 55kDa band was not detected as opposed to a new lower band whose molecular nature remains to be determined ( Fig 4B ) . Nuclear versus cytoplasmic fractions were analyzed in the same way and showed that the accumulation of ubiquitinated CHMP1B following EGF stimulation did not correspond to nuclear CHMP1B ( S4B Fig ) . From these results , we conclude that endogenous CHMP1B is rapidly ubiquitinated upon EGF stimulation resulting in the accumulation of ubiquitinated CHMP1B at cellular membranes . In order to investigate the physiological relevance of CHMP1B ubiquitination , we analyzed the EGFR degradation kinetic following EGF stimulation in stably CHMP1B-silenced ( shCHMP1B ) HeLa cells that were transiently transfected with shRNA-resistant CHMP1B-WT or CHMP1B-4K>R constructs ( Fig 5A; S5A and S5B Fig ) . CHMP1B-silenced cells exhibited delayed EGFR degradation compared to the control cells ( Fig 5A and 5A’ ) . Expressing CHMP1B-WT , but not CHMP1B-4K>R , restored the kinetic of EGFR degradation in CHMP1B-silenced cells to the control cell situation ( Fig 5A and 5A’ ) . We then analyzed the plasma membrane expression of EGFR in CHMP1B-silenced cells expressing or not the shRNA-resistant versions of either wild-type or 4K>R mutant forms of CHMP1B ( Fig 5B and 5B’ ) . The amount of EGFR at the plasma membrane before and after EGF stimulation was followed by immunostaining on fixed , but non-permeabilized cells to better preserve the membrane structure and mostly visualize EGFR located at the plasma membrane . We observed a higher level of plasma membrane EGFR expression in CHMP1B-silenced cells compared to control cells both before and at each time point of EGF stimulation ( Fig 5B and 5B’ ) . Treatment of control HeLa cells with 100 ng/ml EGF induced the expected progressive removal of EGFR from the plasma membrane in the ensuing 30 min and an almost complete loss of staining at 60 min post-stimulation ( Fig 5B and 5B’ ) . In contrast , EGFR staining at the plasma membrane was reduced but still present at 60 min in CHMP1B-silenced cells ( Fig 5B and 5B’ ) . Interestingly , the expression in CHMP1B-silenced cells , of CHMP1B-WT , but not of CHMP1B-4K>R , was able to induce EGFR disappearance from the membrane after EGF stimulation , clearly supporting an important role for CHMP1B ubiquitination in EGFR trafficking ( Fig 5B and 5B’ ) . Co-staining of EGFR and RAB4 , a marker for early recycling endosomes [66] , revealed that both the EGFR amount and the RAB4 staining were higher , with an enhanced correlation of the two signals , in CHMP1B-silenced versus control cells ( Fig 5C and 5C’ ) . We further investigated if the internalization process per se was defective in CHMP1B-silenced cells . To this end , we performed a trafficking assay using an EGFR antibody that recognizes the extracellular domain of EGFR [67] which was directly added to the culture medium on living cells . Then , cells were washed , fixed and secondary antibody was added . This procedure allowed to detect plasma membrane EGFR at time 0 and intracellular EGFR associated with the primary antibody internalized from the cell surface at different time points following EGF stimulation ( Fig 5D ) . The quantification of internalized EGFR over time showed a proper or even enhanced level of internalized EGFR in CHMP1B-silenced cells versus control cells ( Fig 5D and 5D’ ) . Co-staining with endosomal markers to reveal the early ( EEA1 ) , late ( Lamp1 ) and early recycling ( RAB4 ) endosomes indicates that EGFR traffics through these different endosomal compartments although with a different kinetic in shCHMP1B-silenced cells compared to control cells ( S6 Fig ) . Taken together , our results show that EGFR is internalized but most likely less efficiently degraded in CHMP1B-silenced cells compared to control cells following EGF stimulation . Moreover , CHMP1B ubiquitination is required for proper EGFR degradation . To assess the generality of a CHMP1B regulation by the ubiquitin system , we extended our experiments to the Drosophila melanogaster model organism . The Drosophila melanogaster genome encodes one orthologue of HsUSP8 , the protein DmelUSP8 ( syn . DmelUBPY ) that shares 45% sequence identity with its human counterpart and a unique CHMP1 protein , DmelCHMP1 , displaying 90% identity with human CHMP1B [68] . Three out of the four lysine residues targeted in this work are conserved and the MIM domain is almost identical except for one conservative substitution ( Fig 6A ) . Ubiquitous silencing of Chmp1 or Usp8 in Drosophila is lethal before the pupal stage , so we used transgenic fly lines expressing silencing hairpins under the control of the UAS promoter [69] together with the specific wing Gal4 driver MS1096-Gal4 inducing the expression of the UAS constructs in the dorsal wing layer . Consistent with previous studies showing the role of these two genes in wing development [63 , 68 , 70] , silencing either Chmp1 or Usp8 resulted in a similar curved and growth defective wing phenotype ( Fig 6B ) . During wing development , a complex set of signaling events establishes a row of active Notch pathway cells in the wing larval imaginal disc , at the level of the future wing margin [71 , 72] . Using an EGFP reporter gene expressed under Notch Responsive Elements ( NRE ) and the driver engrailed-Gal4 ( en-Gal4 ) to induce the silencing construct in the posterior part of the wing imaginal disc , we observed that silencing endogenous Chmp1 resulted in an enlarged domain of EGFP expression both in the domain of engrailed ( en ) expression and slightly beyond ( Fig 6C ) . Defects outside the domain of shCHMP1 expression is in accordance with a non-cell autonomous role of ESCRT proteins in the regulation of developmental signals [70 , 73–80] . This resulted in observable wing margin defects at the adult stage ( Fig 6C ) . Transgenic flies conditionally expressing human transgenes encoding either wild-type HsCHMP1B or HsCHMP1B-4K>R were then generated to test their ability to rescue the wing phenotype induced by DmelChmp1 silencing . Remarkably , restriction of the Notch signaling domain in larval imaginal discs was restored by expressing wild-type HsCHMP1B resulting in normal adult wings . In contrast , expressing HsCHMP1B-4K>R mutant only partially restored the restriction of Notch activity in the wing margin , resulting in a high proportion of adults presenting wing margin defects ( 85% ) and a few adults with almost normal wings ( 15% ) ( Fig 6C ) . Incomplete rescue by the ubiquitination defective HsCHMP1B-4K>R mutant indicates that modification of CHMP1B by ubiquitin linkage contributes to the protein function in Drosophila thus providing evidence for the evolutionary conservation of a ubiquitination-based regulation of HsCHMP1B and DmelCHMP1 proteins .
Presently , little is known about the regulation and activation of human ESCRT-III proteins in vivo . Here we show that the ESCRT-III member CHMP1B is regulated by ubiquitin linkage which is tightly controlled by the ubiquitin specific protease USP8 . We propose that CHMP1B ubiquitination serves as a checkpoint for spatial and temporal control of its polymerization at endosomes . We identified the lysine residues K87 and/or K90 situated in the linker region connecting α-helices 2 and 3 of CHMP1B as major site ( s ) of ubiquitination . Compared to the human protein , K87 is conserved but not K90 in Drosophila melanogaster and inversely in Arabidopsis thaliana ( S7 Fig ) , pointing out the importance of the presence of at least one lysine residue in this region . This linker region is crucial for the transition from the closed inactive state to the open active polymer conformation of ESCRT-III CHMP1B [22 , 23] . Notably , in the conversion from the closed to the open conformation , the flexible linker region extends helix 2 of the hairpin ( S2A–S2H Fig ) . We hypothesize that ubiquitination of K87 or K90 residues could possibly induce or stabilize an open conformation of CHMP1B monomers or dimers . In agreement with this hypothesis , we have shown that non-ubiquitinated monomers or dimers of CHMP1B are not recognized by the CHMP1B antibody that targets an epitope presumably masked by the auto-inhibitory helix while ubiquitinated forms of CHMP1B expose the corresponding epitope . Remarkably , ubiquitinated forms of CHMP1B were not detected as polymers indicating that ubiquitin moieties on CHMP1B may prevent its assembly into ESCRT-III filaments/complexes . We further purified the CHMP1B containing complexes and showed that they are part of a bigger complex containing IST1 , an ESCRT-III member previously shown to co-polymerize with CHMP1B in vitro [22] . The finding that CHMP1B is free of ubiquitin in these complexes is in accordance with the fact that CHMP1B:IST1 polymerization was observed in vitro with non-ubiquitinated recombinant proteins [22] . Furthermore , we have demonstrated that USP8 deubiquitinates CHMP1B . Thus , deubiquitination of CHMP1B by USP8 at the endosomal membrane may favor CHMP1B oligomerization and co-assembly with IST1 in vivo . It is well described that EGF induces the active sorting of the EGFR at the endosomal membranes where it can be either directed to the lysosomal degradation pathway or recycled back to the plasma membrane [81] . Here , we observed that ubiquitinated CHMP1B strongly accumulate on membranes upon EGF stimulation . Therefore , our observation provides strong evidence for a correlation of the transient accumulation of ubiquitinated CHMP1B at the membrane with the activation of EGFR trafficking . Thus , ubiquitination of one or two of the exposed lysine residues may also favor CHMP1B binding to membrane-associated complexes prior to its incorporation in ESCRT-III complexes . We show that CHMP1B-silenced cells present delayed degradation kinetic of EGFR following EGF stimulation . By staining specifically plasma membrane EGFR in non-permeabilized cells , we were also able to observe that CHMP1B-silenced cells present a delay in the disappearance of EGFR from the plasma membrane following EGF stimulation . However , the CHMP1B-silenced cells present normal or even enhanced internalization of EGFR following EGF stimulation . These results together with the data from the literature showing the role of CHMP1B in MVB biogenesis [82] are consistent with a defect in EGFR sorting , in which decreased lysosomal degradation could be coupled with an enhanced recycling rate of EGFR at the plasma membrane . This hypothesis was further supported by enhanced correlation of EGFR staining with RAB4 following EGF stimulation . While further investigation is needed to understand the exact function of CHMP1B in EGFR sorting , we observed that expression of the CHMP1B-4K>R mutant failed to rescue the degradation kinetic of EGFR while wild-type CHMP1B did rescue it . Although we cannot completely exclude that the K>R mutations impair CHMP1B function by themselves , this substitution is conservative and the side chains of the four lysine residues are solvent exposed in the closed conformation model and in the CHMP1B polymer structure . We thus propose that defective ubiquitination alters CHMP1B function in receptor sorting . Hence , the ubiquitination of CHMP1B would play a major role in the regulation of the sorting of the EGFR at the MVB . Several functions have been described for CHMP1 proteins . The yeast Did2 protein ( CHMP1 ) was suggested to act in concert with IST1 in the control of Vps4 , which disassembles ESCRT-III polymers at a late stage in MVB sorting [83–85] . Furthermore , CHMP1B in complex with IST1 has been implicated in the formation of recycling tubules from endosomes by stabilizing positively curved membrane tubules in overexpressing conditions [22] . These various functions together with the observed dynamic ubiquitination of CHMP1B upon EGF or IL1β treatment could indicate that CHMP1B ubiquitination is part of an ubiquitin-dependent sensing mechanism that might control the fate of receptors towards either recycling or degradation . The crucial role of both CHMP1B ubiquitination and interaction with USP8 was further confirmed in vivo by analyzing its role in Drosophila melanogaster wing development . Proper wing development depends on signaling molecules , such as Hedgehog ( HH ) , Wingless ( Wg ) or the EGFR ligand Vein ( Vn ) , which activate signal transduction pathways through endocytosis of their receptors [41 , 70 , 79 , 86 , 87] . In this process , CHMP1B-dependent secretion of morphogens is essential [70] . Our results indicate that defective ubiquitination of CHMP1B impairs its normal function in the regulation of the developmental signals controlling Notch activation at the wing margin . In summary , we propose that dynamic CHMP1B ubiquitination in response to plasma membrane receptor activation and internalization regulates its association to membrane bound complexes and that subsequent ubiquitin hydrolysis allows its incorporation into ESCRT-III polymers exerting their function in receptor sorting at the endosomes . Our results are in line with regulation of protein polymerization by ubiquitin linkage [88] . Furthermore , the finding that CHMP1B is a target of USP8 may shed new light in the future on understanding its contribution to membrane receptor trafficking , resistance to chemotherapy or EGFR stabilization in Cushing’s disease .
HEK293T and HeLa cells were purchased from ATCC ( LGC Standards , United Kingdom ) . HEK293T and HeLa cells were cultured in Dulbecco's modified Eagle's medium ( DMEM ) and RPMI 1640 respectively ( Life Technologies ) supplemented with 10% heat inactivated fetal bovine serum and 1% Penicillin/Streptomycin mix , and grown in 5% CO2 at 37°C in a humidifier incubator . Cell transfection , immunostaining and imaging ( confocal and high content analysis by automated microscopy ) are described in supporting information ( S1 Detailed procedures ) . Flies were raised and crossed at 18 or 25°C using standard procedures . Stocks used for gene silencing are: BL#28906 ( Chmp1-IR ) and VDRC#v107623 ( Usp8-IR ) and neutralizing UAS transgenes on second and third chromosomes are VDRC#3955 ( UAS-LacZ ) and VDRC#58760 ( UAS-BirA ) , respectively . The drivers used are MS1096 ( BL#8860 ) and enGal4 combined with the Notch signaling reporter gene ( BL#30729 ) . Human CHMP1B wild-type and mutated constructs were sub-cloned into pUAST-attb plasmid using EcoR1 and XhoI as restriction sites . Stable transgenic lines were generated by injection into the fly stock attP40 in order to integrate each rescuing construct at the same genomic location ensuring similar expression levels ( genotype y1 x67 c23; PattP40 ) . Therefore , variability in the amounts of recombinant protein are the results of differences in protein stability rather than to variability of transcription levels . Flies crosses and manipulation are described in supporting information ( S1 Detailed procedures ) . The construct expressing HA-Ubiquitin ( HA-Ub ) was obtained from Dr . Mathias Treier [59] . Mammalian expression constructs of human USP8 were cloned by PCR into pmyc-VN155 ( I152L ) vector at KpnI/SalI , and human CHMP1B wild-type and mutated sequences were cloned into pHA-VC155 vector at KpnI/SalI . All PCR primers were purchased from Sigma-Aldrich . Constructs expressing FLAG-USP8 , FLAG-USP8C748A and FLAG-USP8S680A were kindly provided by Dr . M . Komada [63] . Full length and truncated GFP-tagged constructs of CHMP1B were kindly provided by Dr . M . Maki [57] . GFP-CHMP1B was used as a template to generate substitution of lysine to arginine ( K>R ) residues at position K42 , K59 , K87 or/and K90 by site-directed mutagenesis using the QuikChange II Site-Direct Mutagenesis Kit ( Agilent Technologies ) . GFP-CHMP1B and GFP-CHMP1B truncated constructs were used as a template to generate HA- or VN- and VC- CHMP1B tagged constructs . Flag-USP8 construct was used as a template to amplify the fragment at position 3311: 5’-AATCTTCAGCAGCTTATATCC-3’ which was cloned into the pSilencer plasmid to generate the silencing construct shUsp8 . Silencing CHMP1B in HeLa cells was achieved by stable transfection with shRNA-CHMP1B TRCN0000159294 ( sh-94 ) ( targets 3’UTR region ) or shRNA-CHMP1B TRCN0000165547 ( sh-47 ) ( targets CDS region ) from Mission Sigma shRNA library . Cells stably transfected with non-target shRNA ( indicated as Ctrl or shNT ) were used as controls . RPMI culture medium ( time 0 ) or EGF diluted at 100 ng/ml in RPMI/0 . 5% BSA was added to serum-starved HeLa cells together with a primary antibody specifically directed against the extracellular part of the EGFR ( ATCC mAb Hybridoma 108 , 1/100 ) . EGFR bound to the antibody was let allowed to internalize for 10 , 30 or 60 min of EGF stimulation . Then cells were fixed , permeabilized and immunostained as described in supporting information ( S1 Detailed procedures ) . Cell lysis , fractioning , sucrose gradient , gel filtration , immunoblotting and immunoprecipitation methods are described in supporting information ( S1 Detailed procedures ) . | In multicellular organisms , the interpretation and transmission of cell growth and differentiation signals strongly rely on plasma membrane receptors . Once activated by their ligands , these receptors activate downstream signaling cascades and are rapidly internalized into intracellular vesicles that fuse inside the cell to form the endosomal compartment . From there , the receptors are sorted towards either recycling vesicles or degradative lysosomes via multivesicular bodies . Receptors sorting therefore plays a crucial role in the integration and regulation of intracellular signals during development and numerous physio-pathological processes . It requires extensive membrane remodeling and scission events at the level of the endosomal compartment by so-called ESCRT proteins , including CHMP1B . In this study , we provide evidence for dynamic regulation of CHMP1B function and subcellular localization by ubiquitin linkage . We also show the contribution of the ubiquitin specific protease USP8 in this regulation , which is a known actor of intracellular trafficking and Cushing’s disease . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"invertebrates",
"chemical",
"compounds",
"hela",
"cells",
"molecular",
"probe",
"techniques",
"biological",
"cultures",
"immunoblotting",
"organic",
"compounds",
"animals",
"animal",
"models",
"drosophila",
"melanogaster",
"model",
"organisms",
"immunoprecipitation",
"materials",
"science",
"experimental",
"organism",
"systems",
"basic",
"amino",
"acids",
"amino",
"acids",
"cell",
"cultures",
"molecular",
"biology",
"techniques",
"cellular",
"structures",
"and",
"organelles",
"macromolecules",
"drosophila",
"materials",
"by",
"structure",
"research",
"and",
"analysis",
"methods",
"polymers",
"polymer",
"chemistry",
"proteins",
"ubiquitination",
"chemistry",
"cell",
"lines",
"molecular",
"biology",
"cell",
"membranes",
"precipitation",
"techniques",
"insects",
"arthropoda",
"biochemistry",
"eukaryota",
"cell",
"biology",
"post-translational",
"modification",
"organic",
"chemistry",
"biology",
"and",
"life",
"sciences",
"lysine",
"physical",
"sciences",
"cultured",
"tumor",
"cells",
"organisms"
] | 2018 | CHMP1B is a target of USP8/UBPY regulated by ubiquitin during endocytosis |
Proper development of the immune system is an intricate process dependent on many factors , including an intact DNA damage response . The DNA double-strand break signaling kinase ATM and its cofactor NBS1 are required during T cell development and for the maintenance of genomic stability . The role of a second ATM cofactor , ATMIN ( also known as ASCIZ ) in T cells is much less clear , and whether ATMIN and NBS1 function in synergy in T cells is unknown . Here , we investigate the roles of ATMIN and NBS1 , either alone or in combination , using murine models . We show loss of NBS1 led to a developmental block at the double-positive stage of T cell development , as well as reduced TCRα recombination , that was unexpectedly neither exacerbated nor alleviated by concomitant loss of ATMIN . In contrast , loss of both ATMIN and NBS1 enhanced DNA damage that drove spontaneous peripheral T cell hyperactivation , proliferation as well as excessive production of proinflammatory cytokines and chemokines , leading to a highly inflammatory environment . Intriguingly , the disease causing T cells were largely proficient for both ATMIN and NBS1 . In vivo this resulted in severe intestinal inflammation , colitis and premature death . Our findings reveal a novel model for an intestinal bowel disease phenotype that occurs upon combined loss of the DNA repair cofactors ATMIN and NBS1 .
Defects in T cell development can result due to inefficient repair of DNA lesions that are generated in a programmed manner during the recombination of variable , diverse and joining ( VDJ ) gene segments , a process that is crucial for the generation of the T cell receptor ( TCR ) [1] . Therefore , proper repair of such breaks is vital for lymphocyte generation and survival . An important kinase that functions in the repair of such DNA lesions is Ataxia Telangiectasia Mutated ( ATM ) [2] . Patients ( known as AT patients ) and mice deficient for ATM show T and B cell developmental defects and lymphoma generation [3–11] . Although the development of thymic lymphoma has been linked to aberrant TCR recombination [11 , 12] , it has also been proposed that oxidative damage plays an important part in generating these tumors [13 , 14] . In line with this hypothesis , treatment of ATM-deficient mice with scavengers of reactive oxygen species ( ROS ) alleviates the lymphocyte developmental defects observed in these mice , as well as the development of thymic lymphomas [13] . ATM is regulated by its cofactor NBS1 , mutated in Nijmegen Breakage Syndrome , following the generation of DNA double-strand breaks [15 , 16] . NBS1 functions as part of the MRN complex , consisting of MRE11 , RAD50 and NBS1 , that is a major sensor of DNA double-strand breaks [2 , 17] . The MRN complex binds to broken DNA ends and induces ATM activation to repair the DNA lesions [17] . Recently , however , the MRN complex has also been linked to activating another kinase that belongs to the ATM superfamily known as ATR ( for ataxia telangiectasia and Rad3 related ) [18–22] . The role of ATR is to resolve replication stress by binding single-stranded DNA [23] . Thus , MRN participates in the activation of ATM and ATR . Within the immune system , loss of NBS1 leads to defects in T and B cell development characterized by lymphopenia [24–27] . Nijmegen Breakage Syndrome patients are also predisposed to malignancies , particularly those of the lymphoid system [28] . Furthermore , a ‘humanized’ NBS1 mouse model has been generated and as well as displaying immunodeficiency , this model also develops T cell lymphoma , in a p53 dependent manner [27] . ATM has also been shown to be regulated by a second cofactor , ATMIN ( for ATM Interactor ) [29] also known as ASCIZ ( ATM substrate Chk2-interacting Zn2+-finger protein ) [30] . It is known that ATMIN functions in resolving DNA damage . ATMIN has been reported to function as an ATM-cofactor following replicative stress and hypotonic stress [29 , 31] . It is also required to localize RAD51 following DNA methylation damage [30] . Furthermore , in the ageing mouse brain ATMIN-deficient mice accumulate oxidative DNA damage [32] and in B cells loss of ATMIN during later stages of development ( the pro B cell stage ) leads to genomic instability , chromosomal translocations and tumourigenesis [33] . Yet the functions of ATMIN are not limited to DNA repair: in B cells ATMIN also functions as a transcription factor where it is required to regulate the expression of DYNLL1 [34] . Loss of ATMIN during early B cell development leads to increased apoptosis due to reduced DYNLL1 expression hence inducing Bim-dependent apoptosis [34] . Mechanistically , it has been shown that NBS1 and ATMIN compete for ATM binding and hence regulate ATM function [35] . ATM activity following DNA double-strand breaks is increased in ATMIN mutant cells whereas ATMIN-dependent ATM signaling is increased in cells deficient for NBS1 [35] . Hence , the absence of one cofactor increases activity through the alternative pathway . Because of this mechanism of ATM regulation , ATMIN deficiency rescues NBS1-dependent cellular lethality [35] . Mutations in DNA repair genes , including ATM and NBS1 have been linked with immunodeficiencies in patients and furthermore immune deficiency is an important factor in causality of human inflammatory diseases such as inflammatory bowel disease ( IBD ) [28 , 36 , 37] . For example , patients with Omenn syndrome or common variable immunodeficiency ( CVID ) carry hypomorphic mutations in RAG1/2 , the enzymes that initiate recombination in B and T cells [38–40] . These patients suffer from immunodeficiency but also from chronic inflammation affecting multiple tissues including the gut [38 , 41] . This is , in some cases , due to abnormal T cell production that displays increased affinity for self-antigens . T cells from these patients are autoreactive and hence give rise to chronic inflammation . In addition , it was recently shown that unrepaired lesions in AT patients induce a type I interferon response , which leads to inflammatory manifestations [42] . However , the underlying genetic causes for such inflammatory diseases , including IBD , are largely not known . In order to address whether loss of ATMIN , alone or in combination with loss of NBS1 , leads to T cell-related defects and pathologies , we generated murine models with deletion of ATMIN and NBS1 either alone or in combination . We show that loss of NBS1 led to a developmental block at the double-positive ( DP ) stage of T cell development and reduced TCRα recombination . Unexpectedly , these developmental functions of NBS1 were neither exacerbated nor alleviated by concomitant loss of ATMIN . In contrast , compound mutant mice lacking both ATMIN and NBS1 in T cells displayed immune hyperactivation and an IBD phenotype , which is mediated by T cells and transplantable into control mice . ATMIN/NBS1-deficient mice carried higher levels of DNA damage and their T cells produced elevated levels of proinflammatory cytokines , coupled with increased proliferation . This generated a proinflammatory environment in the intestine , as well as the spleen , leading to premature death . However , the pathology-causing T cells were found to be largely proficient for both ATMIN and NBS1 .
To determine the contribution of ATMIN and NBS1 in T cell development and function , we generated mice lacking ATMIN ( ATMINΔL ) , NBS1 ( NBS1ΔL ) or both ATMIN and NBS1 ( ATMINΔLNBS1ΔL ) , in T lymphocytes by crossing ‘floxed’ mice [33 , 43] to CD2-cre expressing mice [44] ( denoted as ΔL for ‘lymphocyte’ ) ( Fig 1A ) . These mice were then intercrossed with mice that expressed YFP in CD2-cre expressing cells , from the ROSA26 locus [45] ( Fig 1A ) . The efficiency of deletion of ATMIN and/or NBS1 was determined by PCR , performed on DNA from thymus and spleen ( S1A Fig ) . We additionally confirmed the deletion of ATMIN and NBS1 ( as well as ATM ) at the protein level in the thymus and the spleen ( S1B and S1C Fig ) . Deletion achieved with CD2-cre was minimal in the spleen , since this tissue is not made up exclusively of T cells . CD2-cre mediated deletion of NBS1 led to a severe reduction in thymic cellularity by >50% and this surprisingly was neither exacerbated nor alleviated by co-deletion of ATMIN ( Fig 1B ) . Moreover , NBS1ΔL mice displayed decreased CD4 , CD8 double-positive ( DP ) T cells and CD4 single-positive ( SP ) as well as CD8 SP cells ( Fig 1B and 1C ) and these phenotypes were not alleviated by concomitant loss of ATMIN . The cre recombinase was expressed from the double-negative ( DN ) stage of T cell development ( S2A Fig ) and efficiently through DN1 to DN4 stages of T cell development ( S2B Fig ) . This in turn led to efficient deletion of NBS1 in DN1-4 stages of T cell development ( S2C Fig ) . We observed an increase in the total number and percentage of TCRγδ+ T cells upon loss of NBS1 ( Fig 1B and 1C and S3A Fig ) . Loss of ATMIN alone during T cell development did not lead to an apparent phenotype . The NBS1-dependent block in T cell development was also apparent on the periphery , as indicated by a decrease in the total number of splenocytes in mice deficient for NBS1 ( S3B Fig ) . Similarly , the relative percentage of CD4+ , CD8+ and TCRβ+ T cells was dramatically reduced in the spleens of NBS1ΔL and ATMINΔLNBS1ΔL mice ( S3B and S3C Fig ) . Since both ATM and NBS1 are required for the repair of DNA double-strand breaks , a process that occurs during VDJ recombination , we investigated whether recombination of the TCRβ locus was affected in NBS1ΔL and ATMINΔLNBS1ΔL mice . Southern blotting of two distinct TCRβ recombination regions did not reveal a difference in efficiency of TCRβ recombination in mice lacking ATM , ATMIN , NBS1 or both ATMIN and NBS1 ( S4A Fig ) . Also , there was no effect on the quantification of DN T cell subpopulations in any of the genotypes ( DN1-DN4; S4B and S4C Fig ) . Annexin V staining did not reveal an increase in the percentage of apoptotic cells in DN1-4 T cells ( S4D and S4E Fig ) and in vivo bromodeoxyuridine ( BrdU ) labelling did not suggest impairment in the proliferation of DN cells ( S4F and S4G Fig ) . In summary , we did not detect any defects in the DN T cells that would underlie the developmental block at the DP stage in NBS1ΔL mice . Next , we investigated the efficiency of TCRα recombination by quantitative PCR in DP T cells at 16 different recombination regions in the TCRα locus ( represented schematically in S5A Fig ) . This revealed a requirement for NBS1 in TCRα recombination ( for most of the recombination regions we investigated ) ( Fig 1D and S5B Fig ) . Indeed , this defect in recombination is similar to that which occurs due to lack of ATM [46] . Hence , these data reveal NBS1 , and not ATMIN , to be the ATM cofactor required for TCRα recombination . Since TCRα recombination is essential for T cell maturation we assessed the proportion of mature HSAlowTCRβ+ T cells in the thymus . We observed a decrease in the percentage of mature cells in the thymi of NBS1ΔL and ATMINΔLNBS1ΔL mice ( Fig 1E and 1F ) . Annexin V staining revealed elevated apoptotic and necrotic cells in ATMINΔLNBS1ΔL and NBS1ΔL mice ( Fig 1G and 1H ) . To test whether this could contribute to T cell activation , we measured CD44+ cells and found CD44+CD4+ T cells to be enhanced in the thymi of ATMINΔLNBS1ΔL ( S6A and S6B Fig ) . In summary , we reveal novel functions of NBS1 in T cell development since we observe that loss of NBS1 leads to a block in T cell development at the DP stage as well as defective TCRα recombination . Hence , NBS1ΔL mice show reduced mature T cells , and this phenotype is surprisingly exacerbated ( and not rescued ) upon co-deletion of ATMIN . We also observe increased DNA damage and apoptosis in NBS1ΔL mice , but more so in ATMINΔLNBS1ΔL mice . ATMINΔLNBS1ΔL mice but not ATMINΔL or NBS1ΔL mice display increased mortality ( Fig 2A ) . Whereas ATM-/- mice developed thymic lymphomas , the compound mutant mice developed splenomegaly marked by an accumulation of CD3+ cells ( Fig 2B and 2C ) that were of a CD4+ subtype ( Fig 2D ) . The CD8+ T cells were decreased ( Fig 2D ) . An infiltration of CD3+ T cells was observed in multiple organs , including the liver and lungs of moribund ATMINΔLNBS1ΔL mice ( Fig 2E ) . T cells from the spleens of ATMINΔLNBS1ΔL ( but also to a lesser extent in NBS1ΔL ) mice showed an activation phenotype where the proportion of CD62LlowCD44+ activated T cells was increased ( Fig 2F and 2G ) . Moreover , this activation phenotype , characterized by the increased proportion of potentially antigen-experienced CD62LlowCD44+ T cells , correlated with the weight loss of ATMINΔLNBS1ΔL mice ( Fig 2H ) , as such cells were increased in healthy ATMINΔLNBS1ΔL mice and continued to gradually increase with the progression of splenomegaly and systemic inflammation . We observed increased proliferation of ATMINΔLNBS1ΔL T cells by in vivo BrdU incorporation ( Fig 2I ) and furthermore we identified the CD62LlowCD44+ T cells to be of a CD4+ subset ( S7A Fig ) . Unexpectedly , however we observed a decrease in the number of splenic T cells that expressed YFP in ATMINΔLNBS1ΔL mice ( S7B Fig ) . Furthermore , the vast majority of activated CD62LlowCD44+ T cells from ATMINΔLNBS1ΔL mice did not express YFP ( S7C Fig ) and the majority of proliferating BrdU+ T cells from ATMINΔLNBS1ΔL mice were not expressing YFP ( S7D Fig ) . To determine whether YFP- T cells were deleted for ATMIN and/or NBS1 we performed a genotyping PCR on FACS sorted T cells ( naïve and activated ) from a control mouse , YFP- T cells from ATMINΔLNBS1ΔL mice and YFP+ T cells from an ATMINΔLNBS1ΔL mouse . Approximately 80% of YFP- ATMINΔLNBS1ΔL T cells were proficient for ATMIN and/or NBS1 ( S7E and S7F Fig ) . As expected YFP+ T cells displayed approximately 100% deletion of ATMIN and/or NBS1 ( S7E and S7F Fig ) . Taken together , these data indicate that T cells that are largely proficient for ATMIN and NBS1 are causative of the inflammatory phenotype . Since regulatory T cells ( Treg ) are known to maintain T cell homeostasis we assessed their abundance by staining cells isolated from the spleens of control , ATMINΔL , NBS1ΔL and ATMINΔLNBS1ΔL mice for CD4 and Foxp3 . We did not find a reduction in the proportion of Foxp3+ Treg cells . The CD4+Foxp3+ cells were found to be slightly elevated in both ATMINΔLNBS1ΔL and NBS1ΔL mice ( S8A and S8B Fig ) . Therefore we conclude that a decrease in Foxp3+ Treg cells is not the cause for the auto activation of T cells that results due to a concomitant loss of ATMIN and NBS1 , however , we cannot rule out that their function is impaired . In summary , the ATMINΔLNBS1ΔL mice ( and to a lesser degree the NBS1ΔL mice ) display an immune activation phenotype on the periphery , coupled with T cell proliferation . The extent of the autoactivation is marginal in NBS1ΔL mice but is exacerbated and deleterious upon simultaneous loss of ATMIN in T cells . Yet the proliferating and activated T cells were mostly ATMIN/NBS1 proficient . To test whether elevated DNA damage could be a contributing factor to T cell activation , we measured the presence of alkali-labile sites as well as DNA single- and double-strand breaks in splenic T cells using the alkali comet assay . We observed a significant contribution of ATMIN to suppress these types of lesions , as well as of NBS1 ( Fig 3A and 3B ) . Indeed , splenic T cells isolated from ATMINΔLNBS1ΔL mice displayed an elevated amount of DNA lesions . DNA lesions were observed in YFP+ T cells but not YFP- T cells . DNA damage was also confirmed by γH2AX staining , where increased foci were observed in the nuclei of spleens from ATMINΔLNBS1ΔL mice ( Fig 3C ) . We also observed elevated levels of phosphorylated p53 in the spleens of both NBS1ΔL and ATMINΔLNBS1ΔL mice ( Fig 3D ) . Since p53 is stabilized upon phosphorylation , the total levels of p53 were also increased ( Fig 3D ) . This data for NBS1 is in line with a previous report that show increased apoptosis in NBS1-deficient neuronal cells , which is dependent on p53 [43] . Having observed elevated DNA damage , activation , inflammation and proliferation in the spleens of ATMINΔLNBS1ΔL mice we next asked whether this could lead to neutrophil infiltration hence we stained for CD11b and Gr1 to identify neutrophils . We detected an enrichment of neutrophils ( but potentially also monocytic cells ) in the spleens of NBS1ΔL mice and ATMINΔLNBS1ΔL mice ( Fig 3E and 3F ) . The recruitment of neutrophils might be a consequence of the observed T cell activation [47] . To address the causes of the peripheral inflammation more closely , we performed expression analyses by RNA sequencing on total splenocytes isolated from control , ATMINΔL , NBS1ΔL and ATMINΔLNBS1ΔL mice . We chose to analyze total splenic RNA as we aimed to obtain a global representation of pathways and molecules affected in the spleens of moribund ATMINΔLNBS1ΔL mice . Hence , we isolated RNA from spleens of moribund ATMINΔLNBS1ΔL mice , along with ATMINΔL and NBS1ΔL mice . Compared to the ATMINΔL and NBS1ΔL mice , the moribund ATMINΔLNBS1ΔL mice show a profound inflammatory phenotype ( Fig 4A and S9A and S9B Fig ) . Among the most enriched gene ontology terms were ‘inflammatory response’ and ‘regulation of cytokine and chemokine production’ . We selected several genes that were among those most highly upregulated in ATMINΔLNBS1ΔL mice and validated their expression by quantitative RT-PCR analysis . We observed a dramatic increase in the expression of several proinflammatory markers such as Il17a , Ifnγ , Tnfα , IL-1β and Ifitm1 , and chemokines such as Ccl1 , Cxcl10 , Ccl22 and Xcl1 , in ATMINΔLNBS1ΔL mice displaying splenomegaly , compared to the other genotypes ( Fig 4B ) . Since we observed that only a proportion of ATMINΔLNBS1ΔL mice develop splenomegaly , we assessed the activation potential of T cells in the spleens of ATMINΔLNBS1ΔL mice displaying no signs of splenomegaly . To this end we MACS sorted T cells to 90% purity and quantified the expression of a panel of cytokines in the presence or absence of in vitro stimulation ( using anti-CD3 and CD28 antibodies ) . We detected increased levels of mainly Th1 and Th17 proinflammatory cytokines such as Il1β , Tnfα and Il17a specifically in splenic T cells from ATMINΔLNBS1ΔL mice ( Fig 4C ) . After in vitro stimulation there was a substantial increase in production of all displayed cytokines in ATMINΔLNBS1ΔL mice . We confirmed these results by flow cytometry , showing that splenic T cells from ATMINΔLNBS1ΔL mice produced high amounts of IL17A and IFNγ and similarly to previous findings , most of the cytokine producing cells were YFP- ( Fig 4D ) . These data indicate that in ATMINΔLNBS1ΔL mice the expanded ATMIN/NBS1-proficient T cells are highly prone to eliciting an inflammatory response when stimulated in vitro , although the double-deficient mice do not show signs of splenomegaly . Enlarged spleens from ATMINΔLNBS1ΔL mice contain activated , proliferating ATMIN/NBS1-proficient T cells as well as neutrophils . Since neutrophils are known to produce reactive oxygen species ( ROS ) , we investigated the expression levels of genes involved in the oxidative stress response from the RNA sequencing data obtained from ATMINΔLNBS1ΔL mice ( Fig 4E ) . We observed an increase in the expression of genes involved in the clearance of oxidative stress ( Fig 4E ) . To validate this finding , we analysed the expression of selected genes involved in the detoxification of oxidative stress ( Mt1 , Mt2 , Gpx4 and Pdia6 ) and confirmed their upregulation in the spleens of ATMINΔLNBS1ΔL mice ( Fig 4F ) . In support of these findings , we measured ROS production in splenocytes and showed that ROS production was increased in all genotypes but additively so in ATMINΔLNBS1ΔL mice ( Fig 4G ) . A proportion of ATMINΔLNBS1ΔL mice became moribund and displayed systemic inflammation , which also involved the intestine since we observed the development of spontaneous intestinal prolapses . We histologically investigated the large intestine that was found to be thickened and inflamed ( Fig 5A ) . Moreover , histological scoring of the spontaneously sick ATMINΔLNBS1ΔL mice revealed extensive inflammation of the intestine ( Fig 5B ) . Since we observed an infiltration of CD3+ T cells to the large intestine ( Fig 5C ) , we next aimed to address what type of T cells these represented and whether the T cells were deficient or proficient for ATMIN/NBS1 . Hence , we isolated intra-epithelial lymphocytes ( IEL ) from the small intestine of all genotypes , which revealed a specific enrichment of CD4+ T cells in ATMINΔLNBS1ΔL mice ( Fig 5D and 5E ) . The CD8+ T cell compartment was reduced accordingly . Similarly the TCRβ+ T cells were increased accompanied by a concomitant reduction in TCRγδ+ T cells; a population that is normally abundant in the intestine . In conclusion , we observed a severe inflammatory phenotype in ATMINΔLNBS1ΔL mice characterized by infiltration of CD4+ T cells and a reduction of CD8+ T cells in the intestine . In line with our observations from splenic T cells from ATMINΔLNBS1ΔL mice , we observed increased amounts of IL17A and IFNγ in the IELs isolated from ATMINΔLNBS1ΔL mice , which were mostly produced by YFP- T cells ( Fig 5F ) . Moreover , we observed elevated levels of γH2AX , indicative of DNA damage , in the intestine of ATMINΔLNBS1ΔL mice ( Fig 5G and S10 Fig ) . Hence , we conclude that in the intestine of ATMINΔLNBS1ΔL mice , T cells proficient for ATMIN and NBS1 produce enhanced proinflammatory cytokines and interleukins , which drive severe inflammation . To confirm that T cells are the driving cause of the severe inflammation in ATMINΔLNBS1ΔL mice , we isolated CD3+TCRβ+ splenic T cells from control and ATMINΔLNBS1ΔL mice and transferred these cells into immunodeficient RAG2-/- mice ( Fig 6A ) . RAG2-/- mice injected with T cells from ATMINΔLNBS1ΔL mice displayed high mortality due to severe inflammation , including that of the spleen and intestine ( Fig 6B ) . Phenotypic characterization of splenic cells from moribund RAG2-/- mice revealed a decreased proportion of TCRβ+ T cells along with an elevated percentage of CD4+ T cells ( Fig 6C ) hence recapitulating the phenotype of ATMINΔLNBS1ΔL mice . We also detected an increase in CD11b+Gr1+ neutrophils in the spleens of the RAG2-/- mice reconstituted with ATMINΔLNBS1ΔL T cells ( Fig 6C ) . Genotyping PCR was used to quantify the deletion status of ATMIN and NBS1 in T cells from host ATMINΔLNBS1ΔL mice transferred to recipient RAG2-/- mice 6 months post transfer . The deletion efficiency was found to be approximately 10–30% ( S11A Fig ) , indicating that similar to the phenotype observed in ATMINΔLNBS1ΔL mice , the activation of ATMIN/NBS1 proficient T cells leads to the intestine inflammation phenotype in this RAG2 transfer model . However , since the amount of deletion of ATMIN and NBS1 does not change over a 6-month engraftment period ( i . e . 20–40% of the deletion efficiency prior to transfer , S11B Fig ) , it might be possible that ATMIN/NBS1 double-deficient T cells are also involved in the maintenance of the disease ( in addition to its initiation ) under this experimental setting . Taken together , these results clearly demonstrated that the inflammation phenotype observed in ATMINΔLNBS1ΔL mice is due to defects in T cells ( and not in B cells ) . Having observed increased cytokine production by IELs from ATMINΔLNBS1ΔL mice , we next isolated IELs from control , ATMINΔL , NBS1ΔL and ATMINΔLNBS1ΔL mice and analyzed the expression of proinflammatory cytokines by quantitative RT-PCR . We assessed gene expression in both unstimulated and anti-CD3 and CD28 antibody stimulated IELs . We observed high expression levels of the inflammatory cytokines Il-1β , Tnfα and Il17a in both unstimulated and stimulated IEL from ATMINΔLNBS1ΔL mice ( Fig 7A–7C ) . Since only a proportion of ATMINΔLNBS1ΔL mice develop spontaneous colitis we assessed whether these mice would be more susceptible to chemically induced colitis with dextran sodium sulphate ( DSS ) . The DSS model of colitis has similarities to human IBD hence is an ideal model to mimic this disease [48] . We treated control , ATMINΔL , NBS1ΔL , ATMINΔLNBS1ΔL and ATM-/- mice for 8 days with 2% DSS , at which point the control mice did not show any signs of weight loss . In striking contrast to ATMINΔL and NBS1ΔL , only the ATMINΔLNBS1ΔL mice were sensitive to DSS-induced colitis as apparent by the weight loss of approximately 20% in these mice over an 8-day period ( Fig 7D ) . Our data also confirmed the reported mild sensitivity of ATM-/- to DSS induced colitis [49] . Following DSS treatment , mice were sacrificed and the expression of inflammatory cytokines was assessed in the large intestine , as this is the tissue mostly affected following exposure to DSS . We detected a substantial increase in the expression of the inflammatory cytokines Il17a , Ifnγ and Tnfα after DSS treatment , specifically in ATMINΔLNBS1ΔL mice ( Fig 7E–7G ) . Moreover , we observed a thickening of the large intestine , in the DSS treated ATMINΔLNBS1ΔL mice as assessed by histological analysis ( Fig 7H ) . Assessment of the colitis score showed a severe inflammation of the intestine of ATMINΔLNBS1ΔL mice treated with DSS ( Fig 7I ) . These data indicate that although a proportion of ATMINΔLNBS1ΔL mice develop spontaneous inflammation and colitis , these mice are prone to chemically induced colitis .
By utilizing murine models for the conditional deletion of ATMIN and/or NBS1 in T cells , achieved via the use of CD2-cre , we have identified a novel role for NBS1 in TCRα recombination . In this study we have confirmed findings showing that ATM is required for recombination of the TCRα locus [46] and our data indicate that this process is regulated by NBS1 . Hence , NBS1 appears to be the cofactor of ATM that drives TCRα recombination . Loss of ATMIN does not affect TCRα recombination and furthermore loss of ATMIN in NBS1-deficient mice does not rescue this NBS1 dependent defect . As well as uncovering a role for NBS1 in TCRα recombination , we show that loss of NBS1 leads to a block in T cell development at the DP stage of development . This novel finding differs from the developmental block reported by Saidi and colleagues [25] where the use of Lck-cre to mediate NBS1 deletion resulted in a T cell developmental block at the DN3 to DN4 stage [25] . The differences between the developmental block at DP and DN3-DN4 observed by Saidi and colleagues [25] could be due to the use of different cre lines , which delete with varying efficiency during T cell development , with CD2-cre appearing to delete target genes more efficiently . As with our findings for TCRα recombination , ATMIN does not play a role in T cell development and moreover loss of ATMIN cannot rescue NBS1-mediated functions with regard to T cell development . The inability of ATMIN-loss to rescue the NBS1-dependent reduction in thymic cellularity was unexpected as it contrasts to other cellular systems where loss of ATMIN rescues NBS1-dependent cellular lethality [35] . Unexpectedly , however , the combined loss of ATMIN and NBS1 results in spontaneous activation of peripheral T cells , including in the spleen and intestine that results in the development of intestinal prolapses in approximately 30% of ATMINΔLNBS1ΔL mice . Moreover , in an in vivo experimental system for intestinal colitis , using DSS to mimic IBD , we report enhanced colitis , significantly and specifically , in ATMINΔLNBS1ΔL mice . Although we observe a tendency towards spontaneous T cell activation in mice lacking only NBS1 , as indicated by increased antigen experienced cells in the spleen and increased proliferation of splenic T cells , this phenotype is not pronounced enough to give rise to pathology , that is spontaneous colitis development . Hence , in T cells , loss of ATMIN exacerbates a phenotype of spontaneous T cell activation observed upon loss of NBS1 . Interestingly , the T cells that are activated in the ATMINΔLNBS1ΔL mouse model are predominantly T cells that have ‘escaped’ cre-mediated deletion . One could speculate that the spontaneous inflammation driven by ATMIN/NBS1-proficient T cells is a secondary phenotype that occurs due to lymphopenia . In such a scenario the wild-type T cells in ATMINΔLNBS1ΔL mice proliferate to fill an ‘empty space’ . Yet we would argue against this since we do not observe spontaneous intestinal inflammation in NBS1ΔL mice that are as lymphopenic as ATMINΔLNBS1ΔL mice . Furthermore , spontaneous inflammation is not a general feature of lymphopenic mice and to our knowledge this is the first report of a DNA repair deficiency that leads to spontaneous colitis . ATM-deficient mice do not develop spontaneous systemic inflammation and yet the combinatorial loss of ATMIN and NBS1 does . There are two potential explanations for this; firstly , these cofactors have ATM-independent roles that contribute to the development of colitis . Secondly , when removing these two cofactors , ATM is still present but it cannot function . This would allow the kinase to function in a ‘dominant-negative manner’ , binding its substrates but being unable to phosphorylate them . In doing so , other kinases belonging to the ATM-superfamily ( such as DNA-PKcs ) would be unable to compensate for ATM activity . In support of this , the ATM-deficient mouse is viable whereas the kinase-dead ATM mouse is lethal [3 , 50 , 51] . We consolidate the data presented in this manuscript in the form of a model as displayed in Fig 8 . Co-deletion of ATMIN and NBS1 in T cells leads to excessive DNA damage , and in turn , increased apoptosis in T cells , leading to a reduction in T cells numbers . Surviving , mostly wild-type , T cells move to the periphery where they show increased proliferation and activation , as marked by the production of cytokines , including IL-17A . Subsequently , neutrophil infiltration leads to ROS production hence explaining the increased expression of ROS-detoxifying genes , including Mt1 , Mt2 , Gpx4 and Pdia6 that we observe in the spleens of ATMINΔLNBS1ΔL mice . The increased ROS could also exacerbate the DNA damage observed in ATMINΔLNBS1ΔL cells . We propose that the increase in neutrophils and proliferating T cells in moribund ATMINΔLNBS1ΔL mice are the causes of splenomegaly and intestinal inflammation that eventually leads to premature death of ATMINΔLNBS1ΔL mice . Hence our data support a model where ATMIN and NBS1 proficient T cells are the source of inflammation . In summary we have generated a novel mutant mouse strain that develops an IBD-like phenotype that occurs due to the combined loss of the ATM cofactors , ATMIN and NBS1 in T cells . The underlying genetic causes of many patients displaying immunodeficiency and/or IBD are to a large extent unknown . Here we shed light on factors leading to the development of such defects .
Mice were maintained and bred at the Institute of Molecular Biotechnology , Vienna . All experimental procedures were approved by the ethical committee of the Medical University of Vienna and by Federal Ministry of Science and Research and conform to Austrian law ( license number: BMWF-66 . 009/0069-II/3b/2012 ) . The generation of ATMINF/F and NBS1F/F mice has been described previously [33 , 43] . ATMINF/F and NBS1F/F mice were bred to achieve ATMINF/FNBS1F/F mice . For cre-mediated deletion ATMINF/F , NBS1F/F and ATMINF/FNBS1F/F mice were crossed with heterozygous CD2-cre mice [44] and were designated ATMINΔL , NBS1ΔL and ATMINΔLNBS1ΔL . ATM mice were bred as ATM+/- as described previously [3] . Control mice include ATMINF/F-CD2cre- , NBS1F/F-CD2cre- , ATMINF/FNBS1F/F-CD2cre- and CD2cre+ mice . Unless otherwise stated , all mice were used at 6–12 weeks of age . CD2-cre deletion efficiency and genotyping of mice were determined on DNA using PCR-based assay . Primers are listed in Table 1 . RAG2-/- mice were used for the T cell reconstitution experiments [52] . For detection of cell proliferation , mice were injected with 1 mg BrdU or supplemented with 0 . 8 mg/ml of BrdU in drinking water and analysed at the indicated time points . For colitis induction , mice were challenged with 2% ( mass/vol ) dextran sodium sulphate ( DSS; molecular weight 36–50 kDa; MP Biomedicals ) in autoclaved drinking water ad libitum for 8 consecutive days . Weight of the animals was monitored every day . At the end of the treatment mice were sacrificed and the colon tissue was analyzed for cytokine production by quantitative reverse transcription ( RT ) -PCR . RAG2-/- mice were injected intravenously with 5x105 of sorted splenic CD3+ TCRβ+ T cells in 200 μl of PBS . Moribund mice were sacrificed and samples were analyzed by flow cytometry . Tissue was fixed directly after harvesting in 4% paraformaldehyde solution and transferred into 70% ethanol after 24 hours . The samples were dehydrated using an increasing ethanol series and embedded in paraffin . Tissue sections were prepared using a microtome at a thickness of 5 μm . Samples were then rehydrated using xylene , ethanol solutions and water . Hematoxylin and eosin ( H&E ) staining and CD3 ( Dako ) staining was performed and finally slides were mounted in Entellan ( Merck ) and subjected to microscopy . Axio Imager A1 microscope ( Zeiss ) and Axio Cam MRc5 were used to acquire the images . Alternatively , the rehydrated samples were stained with an anti-γH2AX antibody ( Cell Signaling ) and with an In Situ Cell Death Detection Kit ( Roche ) , according to manufacturers’ instructions . Samples were counterstained with diamidino-2-phenylindole ( DAPI ) . Images were acquired on an Axio Imager M2 microscope ( Zeiss ) . Cells at a density of 5×104 were washed in pre-chilled PBS and then mixed in 100 μL 0 . 6% low melting agarose ( Sigma , type VII ) maintained at 37°C . The cell suspension was then immediately layered onto pre-chilled frosted glass slides pre-coated with 1 . 5% agarose and maintained in the dark at 4°C for all following steps . Slides were immersed in pre-chilled lysis buffer ( 2 . 5 M NaCl , 10 mM Tris–HCl , 100 mM EDTA pH 8 . 0 , 1% Triton X-100 , 1% DMSO , pH 10; DMSO and Triton X added shortly before use ) overnight . Slides were washed with pre-chilled distilled water ( 2×10 minutes ) , and next placed for 45 minutes in pre-chilled alkaline electrophoresis buffer ( 55 mM NaOH , 1 mM EDTA , 1% DMSO ) . Electrophoresis was conducted at 30 V for 25 minutes , followed by neutralisation in 400 mM Tris–HCl pH 7 . 0 for 1 h . Finally , DNA was stained with SYBR Gold ( 1:10 , 000 dilution in H2O; Life Technologies ) for 10 minutes . The comet tail moment was measured for at least 50 cells per sample in 3 replicates using the CASP image-analysis program [53] . To detect ROS , splenic T cells were plated on a poly-L lysine ( Sigma ) coated plate ( Corning ) and stained with 5 μM CellROX Deep Red Reagent ( Life Tech ) for 30 minutes at 37°C , washed with PBS , fixed in 4% paraformaldehyde 10 minutes and counterstained with 5 μg/ml Hoechst 33258 for 5 min . Quantification of immunofluorescence images with CellROX was performed based on the mean fluorescence intensity of cytoplasmic area defined by the distance from the nuclei using the CellProfiler cell image analysis software v2 . 0 [54] . T cells were isolated from the spleen using the Pan T Cell Isolation Kit II ( Miltenyi Biotec ) according to manufacturer’s instructions . For cytokine production experiments , cells were stimulated with 25 ng/ml PMA and 10 mg/ml ionomycin in the presence of 10mg/ml Brefeldin A ( all from Sigma ) overnight . For assessment of cytokine gene expression by quantitative RT-PCR , cells were incubated in a 48-well plate , at 37°C with 5% CO2 and 3% O2 in the presence or absence of 2 μg/ml anti-CD3 and anti-CD28 immobilized antibodies ( both from BD ) overnight . Intraepithelial lymphocytes ( IEL ) were isolated from the small intestine . In brief , the small intestine was removed and flushed with PBS . The tissue was cut into pieces and incubated with RPMI medium containing 5 mM EDTA three times for 15 minutes Supernatant was collected and centrifuged and cells were purified on a Percoll ( Sigma ) gradient . Subsequently cells were subjected to flow cytometry analysis . In some cases , cells were cultured in RPMI ( Invitrogen ) supplemented with penicillin and streptomycin ( Invitrogen ) , 10% FCS ( Invitrogen ) and mercaptoethanol . One x 105 cells were incubated in a 96-well plate , at 37°C with 5% CO2 and 3% O2 in the presence or absence of 2 μg/ml anti-CD3 and anti-CD28 immobilized antibodies ( both from BD ) overnight . Cells were then harvested and used for quantitative RT-PCR analysis to determine cytokine expression . For cytokine production experiments , cells were stimulated with 25 ng/ml PMA and 10 mg/ml ionomycin in the presence of 10mg/ml Brefeldin A ( all from Sigma ) overnight . Cells were washed with PBS containing 0 . 5% BSA and incubated for 30 minutes on ice with the following antibodies: anti-CD4 ( RM-4 . 5; eBioscience ) , anti-CD8 ( 53 . 6 . 7; BD ) , anti-CD44 ( IM . 7; BD ) , anti-CD25 ( PC61; BD ) , anti-CD62L ( MEL14; Biolegend ) , anti-TCRβ ( H57-597; eBioscience ) , anti-TCRγδ ( BD ) , anti-CD69 ( H1 . 2F3; eBioscience ) , anti-CD11b ( M1/70; BD ) , anti-Gr1 ( RB6-8C5; eBioscience ) and anti-HSA ( M1/69 , eBioscience ) . Cells were then washed in PBS with 0 . 5% BSA and data was collected using a Fortessa cytometer ( BD Bioscience ) and analyzed using FlowJo software ( Treestar , Ashland , OR ) . In the case of intracellular staining , cells were fixed and permeabilized using the Foxp3 buffer staining kit ( eBioscience ) according to the manufacturer’s instructions prior to staining for intracellular Foxp3 expression using an anti-Foxp3 antibody ( FJK-16s , BD ) , for 30 minutes . For Annexin V staining , cells were washed with PBS and stained with BD Pharmingen Annexin V Apoptosis Detection Kit I according to the manufacturer’s instructions . The detection of BrdU was performed using BD Pharmingen BrdU Flow Kit according to the manufacturer’s instructions . Splenocytes and IELs were harvested and RNA was isolated from cells using phenol-chlorophorm extraction . RNA was treated with 1 μl DNase ( Sigma ) and then reverse transcribed with the SuperScript III Reverse Transcriptase protocol ( Invitrogen ) to obtain cDNA . An amount of 50 ng of cDNA template was used for the qRT-PCR using SYBR Green qPCR Mastermix ( Qiagen ) . For determination of cytokine expression mEF1α was used as reference gene . Alternatively , the DP population of thymocytes ( CD4+CD8+ ) was isolated using fluorescence activated cell sorting ( FACS ) . PCR quantification of TCR recombination regions was performed as published previously using total DNA from the isolated DP ( CD4+CD8+ ) thymocytes [46] . The PCR was performed on a 7900HT Fast Real-Time PCR System ( Applied Biosystems ) . The DN4 population ( CD4-CD8-CD25-CD44- ) of thymocytes was isolated using fluorescence activated cell sorting . Cells were lysed and subjected to PCR amplification of selected recombination regions using the following primers combinations: Jβ2 and Vβ5 . 1; Jβ2 and Vβ8 . 2; Thy1 F and Thy1 R for which the sequences are found in Table 1 . The Thy1 non-recombining region was used as a positive control . PCR products were separated on a 1 . 2% agarose gel and blotted onto a Hybond N+membrane and subjected to Southern blot analysis using a TCRβ probe which corresponds to the Jβ2 . 6 fragment and was obtained by PCR amplification with the Dβ2 and Jβ2 primers followed by gel purification . The Thy1 probe was generated by isolating the PCR fragment resulting from amplification using the Thy1 primers and gel purification . Both probes were labeled using the RandomPrimed DNA Labeling Kit ( Roche Life Science ) and α-32P-dCTP ( Hartmann Analytic ) . The amount of total RNA was quantified using Qubit 2 . 0 Fluorometric Quantitation system ( Life Technologies ) and the RNA integrity number ( RIN ) was determined using Experion Automated Electrophoresis System ( Bio-Rad ) . RNA-seq libraries were prepared with TruSeq Stranded mRNA LT sample preparation kit ( Illumina ) using Sciclone and Zephyr liquid handling robotics ( PerkinElmer ) . Library amount was quantified using Qubit 2 . 0 Fluorometric Quantitation system ( Life Technologies ) and the size distribution was assessed using Experion Automated Electrophoresis System ( Bio-Rad ) . For sequencing libraries were pooled , diluted and sequenced on Illumina HiSeq 2000 using 50 bp single-read chemistry . Base calls provided by the Illumina Realtime Analysis software were converted into BAM format using Illumina2bam and demultiplexed using BamIndexDecoder ( https://github . com/wtsi-npg/illumina2bam ) . Transcriptome analysis was performed using the Tuxedo suite . TopHat2 ( v2 . 0 . 10 , http://genomebiology . com/2013/14/4/R36/abstract ) was supplied with reads passing vendor quality filtering ( PF reads ) and the Ensembl transcript set ( Mus musculus , e73 , September 2013 ) as reference . TopHat2 analyses were run independently for each replicate . Cufflinks ( v2 . 1 . 1 , http://www . nature . com/nbt/journal/v31/n1/full/nbt . 2450 . html ) was used to assemble transcripts from spliced read alignments , using the Ensembl e73 transcriptome as the reference as well as de novo assembly of transcript models . Differential expression was assessed with Cuffdiff v2 . 1 . 1 ( http://www . nature . com/nbt/journal/v28/n5/full/nbt . 1621 . html ) . Transcriptome sets of all replicates for each sample group were combined with Cuffmerge . Finally , cummeRbund ( http://www . bioconductor . org/packages/release/bioc/html/cummeRbund . html ) and biomaRt ( http://www . bioconductor . org/packages/release/bioc/html/biomaRt . html ) were used in combination with custom R scripts to perform quality assessment and further refine analysis results . Cells were extracted in RIPA lysis buffer ( NEB ) supplemented with protease inhibitors ( Sigma ) and phosphatase inhibitors ( Sigma , NEB ) . Western blots were performed using standard procedures . Protein samples were separated by SDS–PAGE ( 3–8% or 4–12% gradient gels; Invitrogen ) , and subsequently transferred onto nitrocellulose membranes . All primary antibodies were used at 1:1000 dilution and secondary antibodies at 1:5000 . The following antibodies were used: ATM ( Santa Cruz ) , ASCIZ ( Millipore ) ; p95 ( known as NBS1 ) ( NEB ) , β-actin ( Sigma ) , pS15-p53 ( Cell Signalling ) , pS824-KAP1 ( Bethyl Labs ) , total p53 ( Pab-421; CR-UK generated antibody ) and HRP-conjugated goat anti-mouse or rabbit IgG ( Sigma ) . The statistical significance of differences between the means of individual experimental groups compared to the control group was calculated using the Student’s t-test . Values with a p<0 . 05 were considered as statistically significant . | Defects in DNA repair pathways can lead to pathogenesis within the immune system , an example of which is inflammatory bowel disease ( IBD ) . Yet the underlying genetic causes of IBD are often unclear . The DNA repair kinase ATM is crucial for the proper development and function of the immune system . ATM is regulated in a stimulus dependent manner by its cofactors , ATMIN and NBS1 . These cofactors compete for ATM binding and in doing so regulate ATM kinase activity . Whereas both ATM and NBS1 function in T cell development and in the maintenance of genomic stability within such cells , the role of ATMIN ( and the contribution of ATMIN and NBS1 ) in T cell function is unknown . Here , we show that whereas NBS1 has distinct ATMIN-independent functions during VDJ recombination , loss of both cofactors resulted in exacerbated DNA damage , T cell hyperactivation , inflammation and an IBD phenotype . The pathology was driven by T cells largely proficient for both ATMIN and NBS1 . These data demonstrate additive effects revealed upon loss of both ATMIN and NBS1 , thus illustrating the importance of these two DNA repair cofactors in proper T cell development and function . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | DNA Repair Cofactors ATMIN and NBS1 Are Required to Suppress T Cell Activation |
Heme-iron recycling from senescent red blood cells ( erythrophagocytosis ) accounts for the majority of total body iron in humans . Studies in cultured cells have ascribed a role for HRG1/SLC48A1 in heme-iron transport but the in vivo function of this heme transporter is unclear . Here we present genetic evidence in a zebrafish model that Hrg1 is essential for macrophage-mediated heme-iron recycling during erythrophagocytosis in the kidney . Furthermore , we show that zebrafish Hrg1a and its paralog Hrg1b are functional heme transporters , and genetic ablation of both transporters in double knockout ( DKO ) animals shows lower iron accumulation concomitant with higher amounts of heme sequestered in kidney macrophages . RNA-seq analyses of DKO kidney revealed large-scale perturbation in genes related to heme , iron metabolism and immune functions . Taken together , our results establish the kidney as the major organ for erythrophagocytosis and identify Hrg1 as an important regulator of heme-iron recycling by macrophages in the adult zebrafish .
Heme is an iron-containing porphyrin that acts as an essential cofactor in numerous biological processes such as oxygen transport , miRNA processing , electron transfer , and circadian clock control [1] . In humans , total body iron store is a composite of dietary iron absorption , which accounts for around 10% of the daily iron requirement after compensating for daily iron losses , and heme-iron recycling through clearance of senescent red blood cells ( RBCs ) [2 , 3] . RBCs contain approximately 70% of the overall body iron in the form of heme , and consequently heme-iron recycling is a significant contributor of systemic iron homeostasis [4] . Thus , a better understanding of heme recycling and transport is critical to understanding the role of heme and iron in red cell synthesis and turnover . The lifespan of mature RBCs is limited in circulation , with approximately40 and 120 days for mouse and human , respectively [5 , 6] . Furthermore , circulating RBCs can be subjected to damage under stress conditions leading to hemolysis . When RBCs become senescent or the number of damaged RBCs increases in the circulation , macrophages from the reticuloendothelial system ( RES , spleen and liver ) contribute to RBC clearance and subsequently promote heme-iron recycling through erythrophagocytosis ( EP ) . Upon degradation of RBCs in the erythrophagosome , heme is imported into the cytoplasm for degradation by the heme-degrading enzyme heme oxygenase-1 ( HMOX1 ) [7] . Defects in erythrophagocytosis ( EP ) lead to aberrant iron homeostasis , culminating in iron deficient anemia ( IDA ) or iron overload [8 , 9] . The Heme Responsive Gene-1 ( HRG1 , SLC48A1 ) was identified previously as a heme importer in the intestine of Caenorhabditis elegans [10] . In ex vivo cultured mouse bone marrow derived macrophages ( BMDMs ) , HRG1 was recruited to the erythrophagosome membranes where it colocalizes with NRAMP1 , an iron transporter found on the phagolysosomal membranes , surrounding ingested senescent RBCs [11 , 12] . HRG1 mRNA is upregulated during erythrophagocytosis ( EP ) in mouse bone marrow derived macrophages ( BMDMs ) and HRG1 protein is strongly expressed in macrophages of the reticuloendothelial system ( RES ) and specifically localizes to the phagolysosomal membranes [11] . Depletion of HRG1 by siRNA in bone marrow derived macrophages ( BMDMs ) causes defective heme transport from the phagolysosomal compartments and a failure to upregulate HMOX1 mRNA , demonstrating that HRG1 must mobilize heme from the erythrophagosome into the cytosol . However , whether HRG1 functions similarly in vivo remains to be elucidated . The teleost fish , Danio rerio ( zebrafish ) , is a powerful genetic animal model for studying vertebrate development and ontogeny of hematopoiesis [13] . In zebrafish , the kidney marrow ( head kidney ) is the adult hematopoietic organ and is functionally analogous to the mammalian bone marrow . Compared to the kidney marrow , the zebrafish spleen is not well characterized , although it has been proposed to function as a reservoir for RBCs where erythrocytes are stored and destroyed [14] . Histological analysis reveals that macrophages , although rare in the zebrafish spleen , contain phagosomes with erythrocytes and other cellular debris , suggesting that erythrophagocytosis ( EP ) could occur in the zebrafish spleen [15 , 16] . However , direct experimental evidence identifying the tissues that are involved in heme-iron recycling in the zebrafish is lacking . In this study , we show that Hrg1 plays an essential role in the recycling of damaged RBCs and that the kidney macrophages are primarily responsible for heme-iron recycling in the zebrafish .
Owing to ancient whole-gene duplication in teleosts , the zebrafish genome can typically contain more than one orthologue per human protein-encoding gene [17 , 18] . Indeed , two hrg1 paralogs are present—hrg1a ( slc48a1b ) and hrg1b ( slc48a1a ) are located on chromosome 6 and chromosome 23 , respectively ( Fig 1A ) . Both paralogs are phylogenetically related to C . elegans ( CeHRG-1 and CeHRG-4 ) , mouse ( MmHRG1 ) and human ( HsHRG1 ) HRG1 ( Fig 1B and 1C ) . Sequence alignment reveal that the protein sequences of Hrg1a and Hrg1b share 73% identity and 86% similarity to each other ( Fig 1C ) . Protein topological predictions show that Hrg1a and Hrg1b contain four transmembrane domains with a cytoplasmic N- and C-terminus and conserved amino acids that have been implicated in heme transport ( Fig 1C , asterisk ) [19] . To determine the spatiotemporal expression of hrg1 , total mRNA was extracted from embryos at different stages , from single cell to 4 days post fertilization ( dpf ) . RT-PCR analysis revealed that the temporal expression patterns of hrg1a and hrg1b are similar and that both hrg1a and hrg1b mRNAs are present in the one-cell embryo ( Fig 1D , S1A Fig ) . As zygotic mRNA expression is turned-on around 3 hour post fertilization , mRNA detected before this stage is typically associated with maternal deposition during oogenesis [20] . Whole-mount in situ hybridization ( WISH ) confirmed the RT-PCR analysis and revealed that hrg1 is ubiquitously expressed throughout developing embryos , with high expression levels in the central nervous system ( Fig 1E , S1B and S1C Fig ) . qRT-PCR from dissected adult zebrafish tissues showed that hrg1a and hrg1b are expressed at different levels in various organs ( S1D and S1E Fig ) . Ectopic expression of fluorescent-tagged zebrafish Hrg1a and Hrg1b in HEK293 cells revealed that Hrg1a and Hrg1b colocalize with LAMP1 , a lysosomal marker , consistent with previous studies that showed that the worm and human HRG1 localize to endo-lysosome-related organelles [10] . Moreover , coexpression of fluorescent-tagged Hrg1a and Hrg1b showed both proteins colocalize to similar intracellular compartments ( Fig 1F ) . These results are consistent with the presence of potential heme-binding ligands and sorting motifs in the C-terminus of Hrg1a and Hrg1b ( Fig 1C ) [10 , 11 , 19] . Expression of zebrafish hrg1a and hrg1b rescued growth of hem1Δ yeast at heme concentrations as low as 0 . 25 μM , comparable to the established heme transporters CeHRG-4 and CeHRG-1 , demonstrating that Hrg1a and Hrg1b are heme transporters ( Fig 1G ) [10] . To further define the function of zebrafish Hrg1 in vivo , we generated hrg1a and hrg1b double knockout zebrafish with CRISPR/Cas9 genome editing . The hrg1aiq261 mutant allele contains a 61 nt deletion and a 7 nt insertion ( -61 , +7 ) in exon 2 , resulting in loss of its original ATG translation start site ( S2A Fig ) . Although an ATG site downstream in the hrg1a ORF could be used for alternative translation initiation , this would cause a truncation of 38 amino acids at the N-terminus . The hrg1biq361 mutant allele carries a -61 nt deletion in exon 3 , resulting in a protein predicted to contain 28 less amino acids at the C-terminus and addition of 6 extra amino acids ( S2A Fig ) . To determine whether the truncated forms of Hrg1a and Hrg1b had residual function , we performed heme-dependent growth assays with the predicted mutant ORFs using the hem1Δ yeast mutant . Hrg1aiq261 and epitope-tagged Hrg1biq361 mutant constructs were expressed , as determined by immunoblotting ( S2B Fig ) , but failed to rescue hem1Δ growth even in the presence of 5 μM exogenous heme ( Fig 2A ) . hrg1 double knockout ( DKO ) zebrafish were generated by crossing hrg1aiq261 and hrg1biq361 mutant alleles ( S2C Fig ) . Intercrossing of hrg1a+/iq261; hrg1b+/iq361 generated progeny with the expected Mendelian ratios ( S2D Fig ) ( chi-square test , p > 0 . 05 ) and survival rates for DKO fish were comparable to wildtype Tü and hrg1aiq261/iq261 or hrg1biq361/iq36 single mutants . Immunoblotting of total membrane proteins obtained from 3 dpf embryos revealed no detectable Hrg1 protein in DKO fish ( Fig 2B ) . To evaluate whether hrg1aiq261/iq261 , hrg1biq361/iq361 or DKO mutants have defects in red cell homeostasis , we first evaluated red cell synthesis by collecting embryos and processing them for o-dianisidine staining which interrogates RBC hemoglobin production . No defects in embryonic hemoglobinization were detected at 3 dpf embryos ( Fig 2C ) . Quantification of GFP-positive RBCs , in hrg1 mutants crossed to the globinLCR-GFP transgenic fish , showed comparable numbers of RBCs demonstrating that neither globin expression nor RBC numbers were altered in the absence of hrg1 ( Fig 2D , S2E Fig ) ( Two-way ANOVA , p>0 . 05 ) . Correspondingly , WISH showed unaltered expression of gata1 , a transcription factor required for erythropoietic lineage specification , and ae1 , a marker for developing RBCs , further indicating normal erythropoietic differentiation in hrg1aiq261/iq261 , hrg1biq361/iq361 , and DKO mutant animals ( Fig 2E ) ( n = 30 ) . RBC morphology , as determined by May-Grünwald-Giemsa staining ( Fig 2F and 2G ) , and hemoglobinization levels , as determined by o-dianisidine staining , were normal ( S2F and S2G Fig ) in RBCs from embryos and adults . Collectively , these results suggested that red cell differentiation and maturation in hrg1aiq261/iq261 , hrg1biq361/iq361 , and DKO mutants were unaffected . Next , we determined whether Hrg1 plays a role in RBC degradation in the zebrafish . To analyze turnover , we had to first establish an in vivo model for hemolysis in the adult zebrafish and then locate the tissue involved in recycling damaged RBCs . To induce acute hemolysis and EP , phenylhydrazine ( PHZ ) was administrated to adult zebrafish . Histological analysis of the kidney , spleen , and liver conducted one day post-PHZ revealed infiltration of large-sized cells with irregular morphology in the kidney and spleen , but not the liver ( Fig 3A , yellow arrows ) . qRT-PCR revealed that expression of the zebrafish hmox1a homolog was significantly upregulated in the kidney , spleen , and liver after PHZ treatment ( Fig 3B ) . Perl’s Prussian blue showed iron-positive pigmentation only in the kidney macrophages but not in the spleen or liver ( Fig 3C , yellow arrows ) , indicating that only the kidneys were responding to hemolysis for heme-iron recycling . Since macrophage-specific antibodies are not available for the zebrafish , transgenic fish expressing gata1:gfp and mpeg1:gfp , which label the erythroid and macrophage cells respectively , were treated with PHZ to determine whether the large-sized cells were macrophages . Immunohistochemistry ( IHC ) with anti-GFP antibody detected GFP-positive cells in the kidney of PHZ-treated mpeg1:gfp zebrafish , indicating that the newly-populated cells are macrophages ( Fig 3D ) . Furthermore , PHZ-treatment of gata1:gfp fish enhanced GFP staining within macrophages while GFP staining was restricted only to erythroid cells in untreated fish ( Fig 3D ) . By contrast , GFP staining was absent in the livers of transgenic and WT ( Tü ) zebrafish ( S3A and S3B Fig ) . Together , these results suggest that macrophages populate the kidney to phagocytose RBCs in response to hemolysis , and that these cells are the primary sites for heme-iron recycling in the adult zebrafish . To assess whether hrg1 plays a role in red cell turnover , we compared hrg1 mRNA levels in tissues of adult zebrafish exposed to PHZ . qRT-PCR revealed that the levels of of hrg1a and hrg1b mRNA were significantly upregulated by PHZ in the kidney ( Fig 4A ) . By contrast , only hrg1a was upregulated in the spleen , and neither were altered in the liver ( Fig 4A ) . Consistent with these findings , IHC using anti-HRG1 antibodies showed increased Hrg1 staining after PHZ treatment in the kidney macrophages , while the signal was barely visible in the spleen and liver ( Fig 4B ) . DAB ( 3 , 3'-Diaminobenzidine ) -enhanced Perl’s iron and Prussian blue staining detected iron accumulation in kidney macrophages of WT zebrafish at 1 , 2 and 3 days post PHZ-treatment , but not in the kidneys of DKO zebrafish ( Fig 4D , S4A Fig ) . In contrast to the WT kidneys , o-dianisidine staining revealed accumulation of heme in the kidneys of DKO zebrafish at 2 and 3 days after PHZ treatment ( Fig 4E ) . No Perl’s iron staining above background could be detected in the spleens from WT or DKO zebrafish ( S4B Fig ) . Importantly , the macrophage numbers or morphology in the kidney and spleen were unaltered in response to PHZ-induced hemolysis in the DKO mutants ( S4C and S4D Fig , yellow arrows ) . These results further confirm that loss of Hrg1 results in the accumulation of heme in kidney macrophages due to defects in heme-iron recycling from damaged RBCs . To determine whether heme and iron-dependent gene expression profiles were perturbed in hrg1 mutant fish , we performed an RNA-seq on total RNA extracted from dissected kidneys and spleens of WT and DKO fish with or without PHZ-treatment . MA-plots of all 24 , 220 genes annotated in zebrafish genome GRCz10 [21] showed large numbers of differentially expressed genes in kidney and spleen when pairwise comparisons were performed between DKO and WT samples ( Fig 5A and S5A Fig ) . Gene ontology ( GO ) enrichment analysis between PHZ-treated versus untreated kidney samples revealed that genes related to porphyrin metabolism , heme metabolism , and intracellular sequestering of iron were significantly downregulated , and immune-related genes were upregulated in the DKO kidney after PHZ treatment ( Fig 5B and 5C ) . Strikingly , enrichment analysis to identify connecting pathways and interactomes from the GO data identified hmox1a connecting the heme/porphyrin metabolism sub-network with iron homeostasis ( Fig 5D ) . On contrast to the kidney , GO analysis of the spleen data showed genes related to blood regulation and hemostasis were downregulated and cell-division genes were upregulated ( S5B and S5C Fig ) . To classify iron metabolism genes in zebrafish , we compiled a list of 86 mammalian iron metabolism genes and performed BLAST homology to identify 124 potential orthologs in the zebrafish genome ( S1 and S2 Table ) . Further examination of these genes identified 20 genes that were significantly downregulated ( p<0 . 0001–0 . 038 ) in the DKO kidneys after PHZ treatment ( Fig 5E ) . These genes included ones involved in heme degradation ( hmox1a ) , iron-storage and transport ( fth , ftl , slc40a1/fpn1 ) , heme synthesis ( alas2 , fech ) , and systemic iron regulation ( erfe ) ( Fig 5E ) . By contrast , 42 genes involved in iron metabolism regulation ( hamp1 , tmprss6 ) , macrophage differentiation ( spic ) , and inflammation ( il6 , il10 , il22 ) were upregulated ( p<0 . 0001–0 . 049 ) ( Fig 5E ) . Comparison of the iron and heme metabolism genes from the kidney and spleen revealed 10 downregulated and 13 upregulated genes that were common to both datasets ( Fig 5F , S5C and S5D Fig ) . These results demonstrate that hrg1 deficiency causes significant alterations in gene expression of heme and iron metabolism pathways in zebrafish .
In this study , we demonstrate the in vivo function of Hrg1 in a vertebrate model system . We show that the zebrafish kidney is the primary organ for heme-iron recycling during EP and that zebrafish Hrg1a and Hrg1b are heme transporters that are expressed and upregulated in kidney macrophages after PHZ-induced hemolysis . In agreement with this finding , genetic ablation of hrg1 results in aberrant heme-iron metabolism at the histological and transcriptomic level . In mammals , the spleen and liver are major organs for heme-iron recycling . In zebrafish , the kidney is the organ where adult hematopoiesis occurs , comparable to the bone marrow in mammals . However , a role in RBC turnover has not been ascribed to the zebrafish kidney . It has been postulated that the zebrafish spleen is the major organ for EP , despite a lack of direct experimental evidence [15 , 16] . This rationale is supported by the observation that the majority of the zebrafish spleen is a reservoir of RBCs , with identifiable lymphatic or myeloid cells plus red and white pulps that are comparable to mammals [15 , 16] . However , our studies show that the zebrafish kidney is the major site for EP after hemolysis . Indeed , iron staining reveals active heme-iron recycling in the zebrafish kidney , but not the spleen and liver even though expression of hmox1a , which encodes for the heme degrading enzyme , is upregulated in all three tissues [22] . We have previously used pre-mRNA splice-blocking morpholinos , which targeted the boundaries of hrg1a intron 2 and exon 3 ( hrg1a_I2E3_MO2 ) and exon 2 and intron 2 ( hrg1a_E2I2_MO1 ) . Both morpholinos resulted in anemic embryos which could be partially rescued with expression of hrg1 . Short sequence BLAST searches showed that the morpholino is specific to its targeting site with low off-target sequences in the zebrafish genome . How is it possible for the hrg1a morphants to show anemia while germline CRISPR mutants do not ? It is possible that compensatory pathways can buffer against deleterious mutations , an effect typically not observed in transient morpholino knockdowns . Indeed , Rossi et al showed that egfl7 mutants do not show any obvious phenotypes compared to morphants because Emilin2 and Emilin3 can compensate for loss of Egfl7 [23] . Another study showed that tmem88a-/- mutant embryos partially recapitulated but had a much milder phenotype compared to tmem88a morphants [24] . One way to recapitulate the morphant phenotype would be to delete the morpholino target site by removing intron 2 . Although we attempted to delete the entire hrg1a and hrg1b locus using two CRISPR guide RNAs , only F0 chimeras were recovered with no germline transmission in the F1 progeny even after screening large numbers ( >300 ) of F1 embryos . We speculate that the hrg1a locus may harbor genetic elements which might be essential for embryonic development . In humans , heme-iron recycling from senescent RBCs contributes to more than 90% of daily iron requirement , while dietary iron accounts for the remaining 10% [2 , 3] . In zebrafish , the individual contribution of iron/heme absorption versus recycling is poorly understood . The zebrafish is typically fed a diet of brine shrimp and dry food extracts every day that is likely to be iron-loaded , together with iron absorbed from surrounding water . While regulation of iron absorption by the Fpn1-Hamp1 axis is well-documented in mice , it is noteworthy that zebrafish can absorb iron by an alternate pathway via their gills [25] . Indeed , we measured hamp1 and fpn1 mRNA by qRT-PCR in isolated livers from zebrafish . Our results show that hamp1 levels are significantly elevated in the presence of PHZ but is equivalent in both , WT and DKO fish ( S6 Fig ) . The upregulation of hamp1 in response to PHZ is consistent with published studies in zebrafish embryos [26] . However , changes in fpn1 expression is statistically not significant even though there is a trend of lower fpn1 levels in DKO fish exposed to PHZ . Whether the Fpn1-Hamp1 axis regulates iron absorption from the gills is currently unknown . Therefore , one possible explanation for why the hrg1 mutants lack overt erythropoietic phenotypes could be because zebrafish can obtain sufficient amounts of iron from dietary absorption or from the water [27] . Our RNA-seq results indicate that not only genes involved in heme-iron metabolism , but also immune-related genes are differentially expressed in PHZ-treated DKO zebrafish . One top hit of upregulated genes in DKO mutants is an unannotated gene named si:ch211-201o1 . 1 , which is an NLRP3 homolog in zebrafish . It has been reported that an increase in extracellular heme activates NLRP3 expression , triggering inflammasome activation [28] . One possible explanation for this upregulation would be that the heme accumulating in the kidney macrophages in the absence of Hrg1 triggers an inflammatory response . It will be interesting to determine the immune response of Hrg1 mutant fish in the presence of pathogens and inflammatory agents that causes hemolysis . Little is understood about red cell recycling and turnover in the adult zebrafish as the vast majority of studies related to red cell development and heme and iron metabolism have been confined to embryos . One major advantage of the zebrafish over other vertebrate models is the ex utero transparency of early-stage embryos which are easier to genetically and chemically manipulate . It is therefore noteworthy that hrg1a and hrg1b mRNA are both expressed in the one-cell embryo raising the possibility that Hrg1 and heme may play a hitherto underappreciated role in early embryonic development . Although there are no obvious phenotypes in the DKO mutant embryos , it is possible that these embryos may be more sensitive to maternal iron deficiency or disruption in heme biosynthesis . Here , generating transgenic zebrafish expressing genetically-encoded heme sensors/reporters [29 , 30] or label-free imaging [31] to directly visualize heme trafficking at the maternal-embryonic interface in the transparent embryo , and within the hematopoietic organs during heme recycling in the adult fish will significantly influence our understanding of the role of heme and iron in vertebrate red cell development .
All zebrafish procedures were approved by University of Maryland College Park Animal Care and Use Committee ( #R-NOV-17-52 ) . The Zebrafish Tü strain was used as wild-type . The light providing cycle was maintained at 10 hr light off and 14 hr light on . Embryos were kept in embryo medium prepared following The zebrafish book ( 4th ) . To keep embryos transparent in early developmental stages , 0 . 003% PTU was added around 18-24hpf . For genotyping of adult zebrafish , a small piece of tail was clipped and placed in 50μl 50mM NaOH at 95°C for 30min . For genotyping of embryos , either whole embryos or a small piece of tail was dissolved in 10μl 50mM NaOH at 95°C for 30 min [32] . The lysates were neutralized by adding 1/10 volume of 1M Tris-HCl pH 8 . 0 and 1 μl of crude lysate was used for PCR genotyping . The PCR fragment of hrg1aiq261 and hrg1biq361 alleles were genotyped with the following primers: for hrg1a , forward: 5’- GAATTATCAAGCTTCACATCACAGGCTCTTTCCGAG -3’ , reverse: 5’- AAGCTACACTGCAGCACCGCTGTCTCCAGGTCAAACG -3’; for hrg1b , forward: 5’- actgcataGGATCCCCCTTTAAAGTGTGTTATCATGTG -3’ , reverse: 5’- GCagactcctcgagCTTCCTACTACAGGGCCTGAATC -3’ . For PHZ treatment in adult fish , 2 . 5 μg/ml PHZ was prepared in system fish water . Adult fish was placed in fish water with PHZ for 25 min at 28°C . PHZ was then rinsed off with fresh fish water . Adult zebrafish ( 8–12 months old ) were anesthetized with 0 . 02% tricaine ( MS-222 , Western Chemical Inc . ) in fish water . Adult fish were first fixed by 10% Neutral buffered formalin ( NBF ) after slitting along the ventral abdominal wall to allow the fixative to immerse the gastrointestinal organs . The volume of fixative was at least 10 times of fish volume . The fixed fish were either processed as whole-mount paraffin embedded sections or frozen sections embedded in OCT after dissecting the kidneys , spleens and livers . Perl’s Prussian blue stain was performed to detect ferric iron zebrafish sections [33] . To perform DAB-enhanced staining , endogenous peroxidase activity was quenched by incubating embryos in 0 . 3% H2O2 ( in methanol ) for 20 min at RT . Following 3 times rinse in PBS , sections were incubated in DAB substrate kit ( Pierce , Thermo Fisher ) for 15 min . Ferric ferrocyanide catalyzes oxidation of DAB , producing a reddish-brown color . Whole amount in situ hybridization ( WISH ) was performed following standard protocol as described [34] . To generate probes for hrg1a and hrg1b , 3’UTR regions were amplified using following primers: hrg1a probe: forward , 5’- gcagtcacctcgagACACACAGCAGCACACTAGTGTC -3’ , reverse , 5’- gatctaggatccGTCTGAGCGCAGCTGACAGAC -3’; hrg1b probe: forward , 5’- gcagactcctcgagTTGGCTCCTTCAGCTCTAATGG -3’ , reverse , 5’- gatctcggatccGACTTAAACTGTATATTATTTCC -3’ . The amplified fragments were cloned to pCS2+ vector with BamH1 and Xho1 digestion . Probes were synthesized by using DIG RNA Labeling Kit ( SP6/T7 ) ( Roche ) . Dissection of various adult zebrafish tissue was performed as previously described [35] . For qRT-PCR experiments , dissected tissue was immediately placed in TRIzol and flash-frozen . The CRISPR gRNA was designed using Optimized CRISPR Design ( http://crispr . mit . edu/ ) . The gRNA target sequences for hrg1a ( NM_200006 . 1 ) and hrg1b ( NM_001002424 . 2 ) are listed: hrg1a exon 2: GGTGGATCTGACGACAGGAA TGG; hrg1b exon 3: GGCGGTAGCGGTAGGAGTAC AGG . The gRNA constructs were cloned using pT7-gRNA as backbone [36] . pCS2-Cas9 was used to produce Cas9 capped mRNA by in vitro transcription . Approximately ~300ng Cas9 mRNA and ~100ng gRNA were co-injected to embryos at 1-cell stage . Injected embryos were raised to adulthood as F0 chimeric founders . The founders were subjected to tail-clip genotyping to confirm indels at target sites . The positive chimeric founders were them crossed to WT zebrafish . The F1 embryos with indels at target sites were raised as stable mutant lines . One microgram of total RNA was used for reverse transcription by iScript cDNA synthesis kit ( Bio-Rad ) . The reaction without reverse transcriptase was used for a negative control . cDNA was diluted 2–5 times for following PCR reaction . qRT-PCR was performed with SsoAdvanced Universal SYBR Green Supermix ( Bio-Rad ) . Each reaction was triplicated to avoid possible random variations . O-dianisidine staining was performed to detect hemoglobin in RBCs of whole embryos or histological sections as previously described [37] . Heme catalyzes oxidation of o-dianisidine in the presence of H2O2 , producing a dark brown color in hemoglobin-positive cells . Briefly , collected embryos or sections were placed in 1 ml staining solution ( 0 . 06 ( w/v ) O-dianisidine , 25% Ethanol , 10mM Sodium Acetate and 0 . 02% H2O2 ) for 20 min in dark conditions . Staining was stopped by rinsing with 70% ethanol . FACS analysis was performed as described [38] . Embryos from globinLCR: GFP transgenic background were pooled . The cells were disaggregated and then sequentially filtered through 70 μm and 33 μm cell strainers . The percentage of GFP-positive cells in transgenic embryos was analyzed by FACSCantos II machine ( BD Biosciences ) . Dechlorinated zebrafish embryos or dissected adult tissues were disrupted using a Dounce Homogenizer in appropriate volume of homogenization buffer ( 10mM Tris-HCl , mM EDTA , 1mM PMSF , 1X protease inhibitor cocktail ( Roche ) ) . The homogenized solution was then centrifuged at 800 g at 4°C for 5 min . The supernatant was ultra-centrifuged at 100 , 000 g at 4°C for 90 min . The supernatant is the cytosolic fractionation and the pellet was treated as crude membrane fraction . The pellet was collected and dissolved in lysis buffer ( 2% Triton-X100 , 150mM NaCl , 50mM Tris-HCl , 20mM HEPES , 1mM PMSF , 1mM EDTA , 1X protease inhibitor cocktail ) . The membrane lysate was used Western blot experiments . Polyclonal HRG1 antibody serum was generated in rabbit using the C-terminal 17 amino acid peptide sequence ( YAHRYRADFADIILSDF ) of human Hrg1 as antigen ( Epitomics , Inc . ) . Since the C-terminal 17 amino acid sequence of human Hrg1 has high homology to zebrafish Hrg1a and Hrg1b ( 15/17 ) , it cross-reacts with both Hrg1a and Hrg1b . For western blot analysis of Hrg1 protein in zebrafish , total protein concentration in membrane fractionation lysate was measured using the Pierce BCA assay kit ( Thermo Scientific ) . Equal amount of total protein was mixed with Laemmli sample buffer and were separated on 12% SDS-PAGE and transferred to a 45 μM nitrocellulose membrane with semi-dry transfer apparatus ( Bio-Rad ) . The affinity purified Hrg1 antibody was used at a concentration of 1:1000 , goat anti-rabbit HRP-conjugated secondary was used at 1: 30 , 000 , and blots were developed in SuperWest Femto Chemiluminescent Substrate ( Thermo Scientific ) . Embryos were anesthetized in of 0 . 02% tricaine in embryos medium , 1% bovine serum albumin ( BSA ) in calcium- and magnesium-free PBS . The tails of approximately 30 embryos were cut with surgical scissors to allow red blood cells ( RBCs ) to flow into the tricaine solution . The tricaine solution containing RBCs was loaded into Shandon EZ Single Cytofunnels ( Thermal scientific ) and concentrated onto a slide ( Thermal scientific ) by centrifugation at 450 rpm for 3 mins using a Shandon Cytospin 4 cytocentrifuge ( Thermal scientific ) according to the manufactory’s instructions . Slides were air-dried prior to May-Grunwald Giemsa staining . May-Grünwald staining solution ( May-Grünwald solution ( MG500 , sigma-aldrich ) : methanol = 1:3 ) was gently added onto slides , incubated 5 min at RT and the stain rinsed off with distilled water . Subsequently , the slides were incubated with 1 ml Giemsa staining solution ( Giemsa Stain ( GS500 , Sigma-Aldrich ) : water = 1:20 ) for 15 to 30 mins . The Giemsa staining solution was then washed off with distilled water and slides were air-dried before examining under microscope . The S . cerevisiae strain W303 containing the hem1Δ mutation has been described previously [39 , 40] . The mutant yeast cells were maintained at 30°C in yeast peptone dextrose ( YPD ) media supplemented with 250 μM δ-aminolevulinic acid ( ALA ) ( Frontier Scientific ) . To generate yeast expression plasmids , the zebrafish hrg1a and hrg1b ORFs were cloned with primers containing BamHI and XbaI restriction sites into the pYES-DEST52 vector ( Invitrogen ) . The mutant alleles of hrg1a and hrg1b were amplified with primers containing BamH1 and XbaI sites ( with and without a C-terminal HA tag ) , and cloned into pYES-DEST52 . The dilution spot assays were performed as described previously [39 , 40] . Plasmids containing potential heme transporters were transformed into hem1Δ yeast using the lithium method [41] . To assay aerobic growth exclusively , yeast was induced with 2% galactose in the of ALA , and then spotted onto plates containing indicated heme or ALA concentrations as well as 2% glycerol and 2% lactate as a carbon source . The western blotting experiments were performed as previously reported [19] . The kidneys from 5–6 month old adult zebrafish ( Tü , DKO , non-PHZ and PHZ-treated ) were dissected and flash-frozen in TRIzol before RNA extraction ( To minimize variations from individual fish , tissues from three adult zebrafish were pooled for each of the three biological replicates . 3 fish as a cohort , 3 cohorts per genotype ) . Total RNA was extracted following TRIzol manual ( Invitrogen ) . The extracted RNA was digested with RNase free-DNase to remove remaining genomic DNA and cleaned up using Qiagen RNA mini column ( Qiagen , Germany ) . Quality and quantity of total RNA were checked by Agilent Bioanalyzer 2100 . One microgram of total kidney RNA and 100 ng of total spleen RNA were used for RNA-seq library construction . Purified mRNA was prepared from total RNA following the manufactory’s manual of NEBNext Poly-A ) mRNA Magnetic Isolation Module ( E7490S , New England Biolabs ) . RNA-seq libraries was constructed with NEBNext Ultra RNA Library Prep Kit for Illumina ( E7530L , New England Biolabs ) . The RNA-seq libraries and fragment size was roughly qualified by Agilent Bioanalyzer 2100 . RNA-seq libraries were quantified using NEBNext Library Quant Kit for Illumina ( E7630S , New England Biolabs ) . RNA-seq was performed using Illumina’s HiSeq-2500 . Total of 24 samples with single-end 50 base reads were sequenced , with triplicate libraries of spleens and kidneys . Bioinformatics quality control was done using FastQC , version 0 . 11 . 5 . The reads were aligned to zebrafish GRCz10reference genome using STAR , version 2 . 5 . 2b . The numbers of reads mapped to genes were counted using HTSEQ , version 0 . 6 . 1p1 . Finally , differentially expressed genes were identified via DESeq2 , version 1 . 12 . 3 with the cutoff of 0 . 05 on False Discovery Rate ( FDR ) . R version 3 . 3 . 2 ( 2016-10-31 ) was used , and Bioconductor version 3 . 4 with BioInstaller version 1 . 24 . 0 were used . For gene annotation , we used Ensembl GRCz10 , release 87 . False Discovery Rate ( FDR ) by Benjamini-Hochberg was used to determine the statistical significance with the cutoff value of 0 . 05 . GO enrichment and network analysis were performed using R package clusterProfiler [42] . All the sequencing data including read counts per gene were deposited to GEO with the accession number of GSE109978 . | Total body iron stores in mammals is a composite of iron absorption from diet and iron recycled by macrophages from dying red blood cells ( RBCs ) . Upon erythrophagocytosis of RBCs , the hemoglobin is degraded and heme is imported from the phagosomal compartment into the cytoplasm so that the iron can be released from heme . Defects in these pathways can lead to aberrant iron homeostasis . The Heme Responsive Gene-1 ( HRG1 , SLC48A1 ) was identified previously as a heme importer in the intestine of the roundworm , Caenorhabditis elegans . In cell culture studies , HRG1 was demonstrated to mobilize heme from the erythrophagosome of mouse macrophages into the cytosol . However , the in vivo function of HRG1 remains to be elucidated . The zebrafish is a powerful genetic animal model for studying vertebrate development and ontogeny of hematopoiesis . In zebrafish , the kidney marrow is the adult hematopoietic organ that is functionally analogous to the mammalian bone marrow . In this study , we show that Hrg1 plays an essential in vivo role in recycling of damaged RBCs , and that the kidney macrophages are primarily responsible for recycling heme-iron in the adult zebrafish . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"blood",
"cells",
"heme",
"fish",
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"immune",
"cells",
"spleen",
"immunology",
"rna",
"extraction",
"vertebrates",
"animals",
"animal",
"models",
"osteichthyes",
"developmental",
"biology",
"model",
"organisms",
"experimental",
"organism",
"systems",
"embryos",
"kidneys",
"extraction",
"techniques",
"research",
"and",
"analysis",
"methods",
"embryology",
"white",
"blood",
"cells",
"animal",
"studies",
"proteins",
"animal",
"cells",
"gene",
"expression",
"biochemistry",
"zebrafish",
"eukaryota",
"anatomy",
"post-translational",
"modification",
"cell",
"biology",
"physiology",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"renal",
"system",
"macrophages",
"organisms"
] | 2018 | Hrg1 promotes heme-iron recycling during hemolysis in the zebrafish kidney |
Control and prevention of Chagas disease rely mostly on residual spraying of insecticides . In Argentina , vector control shifted from a vertical to a fully horizontal strategy based on community participation between 1992 and 2004 . The effects of such strategy on Triatoma infestans , the main domestic vector , and on disease transmission have not been assessed . Based on retrospective ( 1993–2004 ) records from the Argentinean Ministry of Health for the Moreno Department , Northwestern Argentina , we performed a cost-effectiveness ( CE ) analysis and compared the observed CE of the fully horizontal vector control strategy with the expected CE for a vertical or a mixed ( i . e . , vertical attack phase followed by horizontal surveillance ) strategy . Total direct costs ( in 2004 US$ ) of the horizontal and mixed strategies were , respectively , 3 . 3 and 1 . 7 times lower than the costs of the vertical strategy , due to reductions in personnel costs . The estimated CE ratios for the vertical , mixed and horizontal strategies were US$132 , US$82 and US$45 per averted human case , respectively . When per diems were excluded from the costs ( i . e . , simulating the decentralization of control activities ) , the CE of vertical , mixed and horizontal strategies was reduced to US$60 , US$42 and US$32 per averted case , respectively . The mixed strategy would have averted between 1 . 6 and 4 . 0 times more human cases than the fully horizontal strategy , and would have been the most cost-effective option to interrupt parasite transmission in the Department . In rural and dispersed areas where waning vertical vector programs cannot accomplish full insecticide coverage , alternative strategies need to be developed . If properly implemented , community participation represents not only the most appealing but also the most cost-effective alternative to accomplish such objectives .
Over the past 15 years , the burden of Chagas disease has been significantly reduced ( from ∼30 million human cases in 1990 to ∼9–11 million in 2006 ) as a consequence of the direct actions promoted by several multinational regional initiatives [1] , [2] . The key for such success was the long term implementation of residual insecticide applications to kill triatomine bugs , the screening of blood donors for the presence of Trypanosoma cruzi , and the treatment of infected infants born to infected mothers [3] . In the Southern Cone , disease transmission by the main vector , Triatoma infestans , was interrupted in Uruguay , Chile and Brazil and in southern Argentina [1] , [3] . However , limited success was obtained in the Gran Chaco region of northern Argentina , Bolivia and Paraguay ( the core of T . infestans distribution ) where Chagas disease is still highly prevalent . Within its 1 . 3 million km2 , the Gran Chaco provides favorable conditions for the development of Chagas and other neglected diseases , including high levels of poverty and social exclusion , low population density , population mostly rural , subsistence economy , and a weak health system [4] , [5] . Recent estimations of Chagas disease prevalence in rural populations of this region show values ranging from 25% to 45% in Argentina , 17% to 49% in Bolivia and 14% to 56% in Paraguay [5] , [6] , [7] , much higher than the overall 1 . 7% estimated for the Southern Cone countries [2] . Furthermore , the lack of effectiveness of pyrethroid insecticides in peridomestic habitats [8] , [9] coupled with the presence of sylvatic populations in Bolivia and Argentina [10] [L . A . Ceballos , unpublished results] and the emergence of insecticide resistance in Argentina and Bolivia [11] , [12] renders the elimination of T . infestans from the Gran Chaco an elusive challenge . In Argentina , Chagas disease vector control began in 1962 with the creation of the National Chagas Service ( NCS ) [13] , [14] . Inspired by the old malaria programs , NCS established a vertical and centralized structure based on the application of insecticides ( mostly HCH and organophosphates ) by qualified personnel . Overall , the program strongly reduced T . infestans infestation and T . cruzi seroprevalence [14] , [15] , but failed to achieve full coverage of insecticide applications ( as late as 1990 , many districts in the Gran Chaco have not yet been sprayed ) and to interrupt disease transmission . As a consequence of decentralization and reduced health budgets , by the end of 1980's NCS did not have enough resources to maintain a vertical structure nor to warrant the continuity and contiguity of vector control actions . Aware of these limitations , NCS started researching on alternative vector control strategies [16] , [17] . Based on promising field results [16] , and under the aegis of the Southern Cone Initiative , in 1992 NCS launched a new vector control program ( “Plan Ramón Carrillo” ) based on community participation and on the incorporation of appropriate technology [14] , [17] , [18] . This new strategy was embedded in the Primary Health Care ( PHC ) system of Argentina , and included the transference of knowledge and practices of control and surveillance of T . infestans to PHC agents , community leaders and rural villagers , who became the first line of T . infestans control [14] , [17] , [19] . During 1993–2001 , 15 , 500 community leaders sprayed with residual insecticides all of the 961 , 500 houses in the endemic area during the attack phase; 85% of such houses were under community-based vector surveillance [14] . As a consequence of the vector control activities , five provinces , all outside the Gran Chaco , were certified free of vector-borne transmission by 2001 [19] . However , a different scenario was observed in the Argentinean Gran Chaco , with 5 of its 9 provinces reporting vector-borne transmission of Chagas disease by the year 2000 [14] . An evaluation of the effects of the horizontal strategy at the district-wide level in this region is lacking . In its conception , the horizontal strategy involved the participation of rural communities only in the surveillance phase [16] . However , budget and personnel constraints forced NCS to implement a fully horizontal strategy ( i . e . , community participation in both the attack and surveillance phases ) , with the consequent loss of quality of insecticide applications targeting the prevailing high bug infestation levels . Although the horizontal strategy was originally thought to increase the coverage and frequency of insecticide applications while saving the costs of salaries due to the incorporation of unpaid personnel [16] , [20] , [21] , no direct comparative cost-effectiveness ( CE ) analysis between the horizontal and the preceding vertical strategy was performed to date . As a part of a larger project on the eco-epidemiology of Chagas disease in northern Argentina , the objectives of the present study were to assess the effects of the horizontal vector control strategy on the prevalence of infestation by T . infestans and on the occurrence of human acute cases over a 12-year period ( 1993–2004 ) in the Moreno department; and to perform a comparative cost-effectiveness analysis between different vector control strategies ( fully horizontal , vertical and mixed ) in a highly endemic district of the Argentine Chaco .
We analyzed longitudinal data from the NCS for the Moreno Department ( centroid at 62° 26′ W , 27° 15′ S ) , located in the Province of Santiago del Estero , northwestern Argentina ( Figure S1 ) . This district was chosen because: a ) it is located in the Gran Chaco region; b ) historically it presented the highest rates of disease incidence and T . infestans infestation; c ) all previous control programs failed to reach full coverage of spraying activities; d ) an ongoing long-term longitudinal study [22] developed in five rural communities of the Department allowed us to derive key parameters for the present study . In 2001 , Moreno had approximately 25 , 000 habitants and 5 , 439 houses , 54% of which were rural houses belonging to 275 communities [23]; most of the rural communities ( 75% ) consisted of 1–10 houses ( Figure S1 ) . Health infrastructure in Moreno is composed of three hospitals located in the three major cities , and approximately 22 PHC centers scattered among rural communities . Rural houses usually have adobe walls and thatched roofs , one or two bedrooms , and a 5–10 m wide veranda in the front . The peridomestic environment includes structures that do not share a roof with the bedrooms , such as storerooms , chicken coops and corrals . Exploitation of forest resources ( hardwood for charcoal and logs , hunting ) , raising goats ( and cattle ) and subsistence agriculture are the main sources of income of rural villagers . Under the horizontal strategy launched in 1992 , NCS activities focused on: a ) training of local villagers in spraying with pyrethroid insecticides and in bug detection activities; b ) spraying of rural communities when a human acute case was detected; c ) evaluating domiciles and peridomiciles for the presence of T . infestans bugs , and d ) the delivery of insecticides , manual compression sprayers , and other supplies to all community leaders . Training workshops for villagers took place at each local school . Workshops provided basic information on Chagas disease epidemiology , and training in insecticide spraying methods and detection of domestic infestation using sensor boxes [18] . At least one resident or PHC agent from every community was selected as a “leader” , and was in charge of storing and distributing the insecticides and sprayers to the villagers who requested them . Each leader was provided with a 5-liter manual compression sprayer , pyrethroid insecticides , and forms to report the spraying activities to NCS personnel on a regular basis [16] . No salary was paid to leaders for their duties . Insecticide was distributed in small bottles ( doses ) with the amount of insecticide necessary to fill a 5-liter manual compression sprayer . Villagers were in charge of spraying all domestic and peridomestic structures in their house . After spraying , villagers had to return the manual compression sprayers to the leader , indicating the number of insecticide doses applied , and whether they found T . infestans bugs before , during , or immediately after spraying . The monthly number of sprayed houses , the amount of insecticide and domestic boxes used , and the number of house compounds infested by T . infestans in domiciles and peridomestic habitats were then reported by leaders to NCS . The program scheme included an attack phase with two spraying rounds of every rural house separated by six months . After the first or second spraying rounds a community was considered under surveillance phase . Suspension concentrate ( SC ) deltamethrin applied at 25 mg/m2 or 20% SC cypermethrin at 125 mg/m2 were the insecticides and doses most commonly used ( Table S1 ) . We performed a generalized CE analysis [24] and compared the observed CE of the fully horizontal vector control strategy with the expected CE of a vertical or a mixed strategy ( i . e . , vertical attack phase followed by a horizontal surveillance phase ) . Generalized CE analysis is based on the evaluation of a suite of interventions against the counterfactual of “doing nothing” , thereby providing a unique framework for evaluating and comparing health interventions , and a gateway for identifying opportunities to improve them . Direct and indirect costs were estimated separately for the attack and surveillance phases . Direct costs included staff ( salaries and per-diems ) , supplies ( consumables used for insecticide spraying and vector surveillance ) and mobility ( fuel and minor vehicle fixes during fieldwork ) ( see Text S1 for more details ) . Straight line depreciation was used to reflect the cost of the use of vehicles ( 10 years ) and manual spraying compressors ( 5 years ) . Indirect costs included the maintenance of vehicles and the payment of salaries during the time in which personnel was not assigned to field activities . Costs in Argentine pesos were inflated to 2004 US dollars . Costs were only estimated for activities performed in rural communities . Observed costs for the implementation of the fully horizontal strategy were obtained from NCS records , whereas for the vertical and mixed strategies costs were estimated based on the number of houses of Moreno and the personnel and supplies needed for each strategy ( see Text S1 for more details ) . The number of Chagas disease human cases ( symptomatic and asymptomatic ) averted by each strategy was chosen as a measure of their effectiveness . Averted cases were estimated as the difference between the number of human cases observed ( horizontal ) or expected ( vertical and mixed ) for each strategy and the number of cases expected in the absence of vector control actions . The number of human cases ( I ) was estimated by applying the following discrete model: I ( t ) = λt*St ; where λt represents the instantaneous incidence rate and St the number of susceptible individuals in year t . Estimation of averted cases was based on the following assumptions: ( 1 ) the acquisition of infection is independent of age and sex . ; ( 2 ) infection is irreversible; ( 3 ) mortality , immigration and emigration are negligible; ( 4 ) on average , each year there were 631 live births [23]; ( 5 ) congenital transmission is negligible; ( 6 ) the susceptible population at year 0 is equivalent to 67 . 7% of total rural population [22]; ( 7 ) in the absence of control actions , the instantaneous incidence rate ( λ ) is constant in time and space and equivalent to the observed value in rural communities of the Moreno Department in 1992 in the absence of control interventions ( 4 . 3 per 100 person-years ) [25]; ( 8 ) reported symptomatic cases are only 7% of total cases [26] . Cost-effectiveness was estimated as the ratio of direct or indirect costs to the number of averted cases , and expressed as 2004 US dollars per averted case . The strategies evaluated were: 1 ) fully horizontal; 2 ) vertical ( assuming interruption of disease transmission after the attack phase ) ; 3 ) mixed with vector-borne transmission ( i . e . , a scenario with persistent transmission throughout the surveillance phase ) , and 4 ) mixed without vector-borne transmission ( i . e . , the attack phase effectively interrupted vector-borne transmission ) . A sensitivity analysis was performed to evaluate the relative effects of individual key parameters on the absolute value of the CE ratio . To assess the long term effects of each strategy , CE ratios were projected over a 25-year period considering a 12 . 8% inflation rate ( the 2002–2006 average for Argentina ) [27] and a 3% discounting rate per year . For the vertical strategy , an optimistic scenario in which T . infestans could be eliminated from Moreno after 10 years of sustained vector control actions was considered . For the 25-year projection , total costs ( and hence , CE ratios ) accrued in the vertical strategy after year 10 were considered zero due to the interruption of NCS visits after the elimination of T . infestans . Because vector-borne transmission of T . cruzi occurs mostly in rural or peri-urban areas , we excluded the three main cities of Moreno ( totaling 2 , 528 houses ) from all the analyses . To compare the prevalence of domestic infestation according to the number of times a community was sprayed , we applied Kruskal-Wallis tests with Dunn contrasts [28] . Multiple lineal regression analysis was applied to test whether spraying coverage ( i . e . , the percentage of houses in the community that were sprayed in the most recent round ) and the number of times the community was sprayed from 1993 to 2000 were significantly associated with the prevalence of domestic infestation in year 2000 ( the year with more simultaneous records of community infestation ) . Statistical analyses were performed using SPSS 14 . 0 ( SPSS Inc . , Chicago , IL ) and STATA 9 . 1 ( Stata Corp , College Station , TX ) .
Of the 275 rural communities found in Moreno in 1993 , 242 ( 88% ) were sprayed with insecticides at least once during 1993–2004 and were thus considered under vector surveillance . The remaining 33 communities were only visited by NCS personnel for community training , insecticide delivery or vector evaluations , but were never registered as sprayed . Most ( 79% ) of these rural communities had an average of 1 to 4 houses . Only 55 ( 23% ) of the rural communities declared under vector surveillance had two insecticide spray cycles under the attack phase . Villagers performed 79% of the 5 , 759 insecticide sprays registered in Moreno ( Table S1 ) . The total average number of insecticide doses per household was 7 . 0 during the attack phase and 5 . 3 during the surveillance phase ( Table S1 ) . A total of 1 , 793 insecticide fumigant canisters were delivered during 1993–2004 , at a rate of 146 and 152 canisters per year in the attack and surveillance phases , respectively . Moreover , a total of 12 , 982 domestic biosensor boxes were delivered to the communities for vector surveillance ( Table S1 ) . The prevalence of infestation by T . infestans in rural domiciles was 77% in 1993 and decreased to 4% by 1996 ( Figure 1A ) , coinciding with the attack phase . After 1996 domestic infestation fluctuated between 10% and 28% . Peridomestic infestation followed the same trend as domestic infestation , with a decline from 78% to 9% by 1996 , and ranging from 22% to 38% during 1997–2004 ( Figure 1A ) . The prevalence of infestation in 25 communities during 1999–2001 was positively correlated with the infestation prevalence assessed by timed manual collections performed by NCS staff in the same communities in 2002 in domiciles ( r = 0 . 45 , P<0 . 02 ) , but not in peridomiciles ( r = −0 . 14 , P>0 . 4 ) . Because leaders' reports likely underestimated peridomestic infestation ( with which they have less contact ) , this information will not be analyzed hereafter . The initial attack phase apparently produced a downward trend in the reported number of human acute cases , from an average of 10 per year during 1988–1993 to 0 in 1997 ( Figure 1B ) . From 1998 to 2004 the annual number of cases fluctuated between 0 and 3 with no clear trend . All reported cases were symptomatic , and referred for standard treatment at Hospital Independencia in Santiago del Estero's capital . Domestic infestation prevalence varied significantly with the number of times each community was reported as sprayed with insecticides by rural villagers ( Kruskal-Wallis χ2 = 17 . 9; g . l . = 4; P = 0 . 003 ) ( Figure 2 ) . In communities reported as never sprayed since 1993 , the median domestic infestation prevalence was 100% ( Figure 2 ) . In communities registered as sprayed once during 1993–2000 , a significant reduction in the median domestic infestation prevalence was observed ( Dunn contrast , Q = 3 . 02; g . l . = 1; P<0 . 05 ) . The increase of insecticide spraying frequency from one to three was not followed by a significant reduction in domestic infestation prevalence ( Q<2 . 93; P>0 . 05 ) . However , when communities were reported as sprayed four or more times during 1993–2000 , a significant reduction in the median domestic infestation prevalence was observed ( Q = 4 . 08; g . l . = 1; P<0 . 05 ) . The prevalence of domestic infestation by T . infestans in 2000 was significantly and positively associated with the time since last insecticide spray ( multiple linear regression coefficient , beta = 0 . 39 , t = 3 . 34 , P<0 . 001 ) and negatively associated with the coverage of the last insecticide spray ( i . e . , beta = −0 . 38 , t = −3 . 28 , P≤0 . 001 ) . On average , communities with a domestic infestation prevalence ≥50% in 2000 were sprayed 5 . 0 ( Standard Deviation , SD , 1 . 8 ) years earlier , whereas communities with domestic infestation prevalences ≤50% were sprayed 3 . 0 ( SD , 2 . 0 ) years earlier . The mean coverage of the last insecticide spray was 79% ( SD , 28% ) for communities with domestic infestation prevalence ≥50% in 2000 , and 84% ( SD , 24% ) for communities with domestic infestation prevalence ≤50% . Total cost ( direct and indirect ) of the fully horizontal strategy implemented in Moreno during 1993–2004 was $309 , 426 , of which 47% corresponded to indirect costs ( Table S2 ) . Indirect costs represented 38% of the total $849 , 625 estimated for the vertical strategy ( Table S3 ) and 42% of the $582 , 885 estimated for the mixed strategy ( Table S4 ) . Annual direct costs of the horizontal strategy were between 3 . 4 ( attack ) and 3 . 2 ( surveillance ) times lower than the annual direct costs of the vertical strategy ( Figure 3 ) . The cost in personnel ( salaries and perdiems ) was the cause of the marked difference between strategies . Personnel costs for the vertical strategy were 8 . 6 ( attack ) and 5 . 6 ( surveillance ) times higher than personnel costs for the horizontal strategy ( Figure 3 ) . The total direct cost of spraying a single house during the attack phase was US$ 15 for the horizontal strategy and US$ 38 for the mixed and vertical strategies , whereas the cost of surveying a single house was US$ 17 for the horizontal strategy , US$ 20 for the mixed strategy and US$ 22 for the vertical strategy . The CE ratio of each strategy ( expressed in 2004 US$ per averted case ) is presented in Table 1 . The lower the coefficient the more cost-effective a strategy ( i . e . , less money would be needed to avert a single case ) . Although the fully horizontal strategy showed direct CE ratios 1 . 9–3 . 3 times lower than the other strategies , the estimated numbers of human cases were 1 . 6 to 4 . 0 times higher than for the remaining strategies ( Table 1 ) . When those strategies that may accomplish the interruption of disease transmission ( i . e . , vertical and mixed WoT ) are compared , it can be seen that the strategy Mixed WoT would be the most cost-effective ( Table 1 ) . Figure 4 shows the results of the sensitivity analysis of CE to various parameters . Changes in the incidence rate ( lambda ) exerted the highest variation of the direct CE ratio for all strategies . At low incidence rates ( 0 . 01 cases per year ) , horizontal and mixed WoT strategies presented similar and lower CE ratios than the vertical strategy ( ΔCE = 201 ) , whereas at high incidence rates ( 0 . 08 cases per year ) the difference in CE ratios between strategies was less marked ( ΔCE = 58 between horizontal and vertical ) . Variations in the acute infection rate or the baseline human infection prevalence did not affect CE values greatly ( range of ΔCE between strategies , 66–109 ) , with the horizontal strategy presenting always lower CE ratios than the mixed and vertical strategies ( Figure 4 ) . The last panel of Figure 4 shows the variation of CE ratios due to changes in perdiems ( from a 50% reduction to the complete elimination ) , a scenario compatible with the decentralization of vector control activities . The elimination of perdiem expenses reduced the CE ratio of the vertical and mixed strategies to values closer to the CE ratio of the horizontal strategy ( ΔCE between horizontal and vertical strategies = 28 ) . The long-term effectiveness of each strategy was evaluated by projecting the annual direct CE ratios over a 25-yr period ( Figure 5 ) . For the vertical strategy , a scenario that assumed the elimination of T . infestans ( and the suppression of the costs associated with vector control ) after 10 years was evaluated . Figure 5 shows that the mixed and fully horizontal strategies would be more cost-effective than the vertical strategy for up to 16–19 years of interventions , and that the CE of the horizontal and mixed strategies would converge after 21 years of interventions .
As with other vector-borne diseases , the incorporation of community participation in Chagas' disease control and prevention evolved in response to the failure of some vertical programs to achieve their main objectives ( originally , vector elimination and interruption of disease transmission ) borne , in part , by the acute limitations in personnel and financial support of the health system [18] , [20] , [21] . The present study represents the first thorough evaluation of the overall performance of a horizontal Chagas' disease vector control program and the first comparative assessment of the CE of different vector control strategies in a highly endemic rural area of Argentina . The results derived from our work may help NCS and other vector control agencies to better plan and design cost-effective control interventions against Chagas' disease vectors . To achieve significant levels of vector control in endemic areas , Chagas disease control actions need to be sustained over time [22] , [29] , [30] . In many Latin American countries , the current scenario of partial decentralization of health services , increased poverty , lack of political interest and declining funding for vector control activities represent a serious challenge for the persistence of vertical control strategies [29] . Furthermore , in rural and dispersed areas where waning vertical vector programs cannot accomplish full coverage , alternative strategies need to be developed . The incorporation of participatory approaches against vector borne diseases not only has proven to be cost-effective but also important for the sustainability of control programs [20] , [21] , [22] , [30] , [31] . Our analysis shows that the implementation of a mixed strategy would have averted between 1 . 6 and 4 . 0 times more human cases than the fully horizontal strategy and , given the realities observed in the ground , would have been the most cost-effective option to interrupt parasite transmission . If properly implemented , community participation represents not only the most appealing but also the most cost-effective alternative to control Chagas disease vectors in resource-constrained settings . When CE ratio projections were compared , it was clear that the main difference between strategies arises with the potential elimination of T . infestans , since vector elimination from a defined region is associated with a significant reduction or even suppression of operational budgets [32] . Although initially it was assumed that 10 years of vector control actions would be enough to accomplish the regional elimination of T . infestans from the Southern Cone [33] , the eco-epidemiologic reality observed in the Gran Chaco region challenges the feasibility of such assertion [5] , [22] , [29] . The impossibility of accomplishing the regional elimination of T . infestans would have a significant effect in the long-term costs of vector control actions since it would be necessary to maintain a sustained and indefinite surveillance phase to prevent domestic reinfestation by T . infestans and interrupt vector-borne transmission of T . cruzi . The success of participatory approaches against tropical diseases is strongly dependent on sustained and continuous collaboration and articulation between external agencies , governments , and communities [30] . In Moreno , such coordination occurred during the attack phase ( evidenced by the significant decrease in bug infestation and disease transmission ) but not during the surveillance phase . In the latter period , the nearly absence of insecticide sprays and the gradual increase in the prevalence of T . infestans infestation were determined by the shortage of insecticide purchases at the central level and by the shift of personnel from the Chagas control program to the recently established dengue control program . Shortage of insecticides , spare parts for compression sprayers and absence of NCS personnel in the field were probably the main obstacles for villagers to continue control activities during the surveillance phase . It is not surprising that new human acute cases of Chagas disease were reported starting in 1998 . However , when a mixed control strategy coordinated and supervised by NCS is implemented , T . infestans infestation can be significantly reduced and vector-borne transmission of T . cruzi successfully interrupted [16] , [22] , [34] , [35] , [36] . One of the direct benefits of the inclusion of rural communities in vector control activities is the offset of the high personnel costs associated with vertical , centralized strategies [20] , [21] . In Moreno , the implementation of fully horizontal or mixed strategies represented a 1 . 6–3 . 5-fold reduction of total direct costs in comparison to a vertical strategy . Such reduction in personnel costs in horizontal strategies , however , came associated with an increase in opportunity costs , because villagers and PHC agents had to divert their available time to control and prevention activities . Given the difficulty to estimate the time villagers devoted to control and surveillance activities , opportunity costs were not included in our cost estimates . Indirect costs represented a significant component of the total cost of each strategy ( range , 38–47% ) , with personnel cost being the most important component . Such high costs were the consequence of NCS centralized structure , since field technicians remained stationed at their central base after long distance travel to the field , devoting their time to activities other than vector control . As shown by the sensitivity analysis , decentralization of NCS structure would be a viable alternative to reduce vector control costs , since perdiem expenses would be sharply reduced and personnel time devoted to vector control , community education and supervision increased . The measure of effectiveness chosen for the present study allowed the estimation of the cost of averting a single vector-borne Chagas disease human case . However , other measures like the reductions of bug infestation levels , disability adjusted life-years ( DALYs ) or the quality-adjusted life-years ( QALYs ) have been proposed for evaluating the effectiveness of control programs [37] . Although vector control actions would have a direct effect on domestic infestation levels by T . infestans , such measure of effectiveness was not used in the present study because vector-borne transmission of T . cruzi seems to occur at even low bug densities [25] , [38] . In addition , DALYs and QALYs were not used because of their known underestimation of the disability weight for chronic parasitic diseases [39] , [40] , and lack of relevant data ( e . g . , age-adjusted infection prevalence and mortality rates for each infection phase ) to parameterize them . As most ( 79% ) of the insecticide sprays in Moreno were conducted by villagers , the data herein presented show how effective such sprays were on bug infestation and disease transmission . The prevalence of domestic infestation in communities with an active surveillance and 4 spraying rounds or more during 1993–2000 was 10% , indicating that control actions performed by villagers were sufficient to maintain low infestations but not to eliminate T . infestans from domiciles . This may be expected from the observed lower effectiveness of insecticide sprays performed by villagers rather than by NCS technicians , and the resulting higher reinfestation rates . As part of an insecticide trial in 400 houses of Moreno department during 2002–2005 , we surveyed heavily infested communities that had been last visited by NCS 4–5 years earlier and found that many villagers did not spray their houses correctly; did not take all the furniture and other items out of the domicile before spraying; changed the dilution of the insecticide to make it last more; and used sprayers in inadequate conditions . Because training workshops had occurred almost 10 years before , most of the young people did not know how to properly spray a house . This demonstrates that community participation cannot be assumed , but has to be systematically supported and promoted [31] . Unit costs of house spraying with pyrethroid insecticides in Moreno were within the cost range estimated for other areas in the Americas , with insecticide costs being a variable ( albeit important ) budget component of the vector control programs ( Table 2 ) . Such costs were much lower and therefore more affordable than the 200–2 , 000 US$ range estimated for housing improvements [41] . The dependence on residual insecticides for the suppression of disease transmission represents an additional burden for Chagas disease vector control programs because insecticide purchases are negotiated at international market prices . The integration of disease programs ( i . e . , Chagas and malaria where both diseases overlap ) and , particularly , the international call for a significant reduction in insecticide prices allocated for vector-borne disease prevention in developing countries represent , in our opinion , some of the integrated , inter-programmatic , inter-sectoral actions [42] needed for reducing the burden of Chagas disease in the Americas . | Despite decreasing rates of prevalence and incidence , Chagas disease remains a serious problem in Latin America , especially for the rural poor . Without vaccines , control and prevention rely mostly on residual spraying of insecticides . Under the aegis of the Southern Cone Initiative , and in agreement with global trends in decentralization of the health systems , in 1992 the Argentinean vector control launched a new vector control program based on community participation . The present study represents the first thorough evaluation of the overall performance of such vector control program and the first comparative assessment of the cost-effectiveness of different vector control strategies in a highly endemic rural area of northwestern Argentina . Supported by results of independent studies , the present work shows that in rural , poor and dispersed areas of the Gran Chaco region , the implementation of a mixed ( i . e . , vertical attack phase followed by horizontal surveillance ) strategy constantly supervised and supported by national or local vector control programs would be the most cost-effective option to interrupt vector-borne transmission of Chagas disease . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"public",
"health",
"and",
"epidemiology/infectious",
"diseases",
"public",
"health",
"and",
"epidemiology",
"public",
"health",
"and",
"epidemiology/health",
"policy"
] | 2009 | Cost-Effectiveness of Chagas Disease Vector Control Strategies in Northwestern Argentina |
Early identification of adverse effect of preclinical and commercial drugs is crucial in developing highly efficient therapeutics , since unexpected adverse drug effects account for one-third of all drug failures in drug development . To correlate protein–drug interactions at the molecule level with their clinical outcomes at the organism level , we have developed an integrated approach to studying protein–ligand interactions on a structural proteome-wide scale by combining protein functional site similarity search , small molecule screening , and protein–ligand binding affinity profile analysis . By applying this methodology , we have elucidated a possible molecular mechanism for the previously observed , but molecularly uncharacterized , side effect of selective estrogen receptor modulators ( SERMs ) . The side effect involves the inhibition of the Sacroplasmic Reticulum Ca2+ ion channel ATPase protein ( SERCA ) transmembrane domain . The prediction provides molecular insight into reducing the adverse effect of SERMs and is supported by clinical and in vitro observations . The strategy used in this case study is being applied to discover off-targets for other commercially available pharmaceuticals . The process can be included in a drug discovery pipeline in an effort to optimize drug leads and reduce unwanted side effects .
Early identification of the adverse effects of preclinical and commercial drugs is crucial in developing highly efficient therapeutics , since unexpected adverse drug effects contribute to one-third of all drug failures in the late stage of drug development [1] . Conventional practices for identifying off-targets rely on a counterscreen of compounds against a large number of enzymes and receptors in vitro [2–4] . Computational approaches could not only save time and costs spent during in vitro screening by providing a candidate list of potential off-targets but also provide insight into understanding the molecular mechanisms of protein–drug interactions . It has been shown that potential off-targets can be identified in silico by establishing the structure–activity relationship of small molecules [5–12] . However , the success of ligand-based methods strongly depends on the availability and coverage of the chemical structures used in training , and few of them directly take the target 3D structure into account . Although the assessment of protein–ligand interactions by docking studies at the atomic level is extremely valuable for understanding the molecular mechanism of adverse therapeutic effects [13 , 14] , protein–ligand docking on a large scale is hindered by the biased structural coverage of the human proteome [15] and a lack of practical methodologies to accurately estimate the binding affinity [16] . Here we approach the problem from a different direction by postulating that proteins with similar binding sites are likely to bind to similar ligands [17] . In this study we test this postulate by predicting potential off-target binding sites for selective estrogen receptor modulators ( SERMs ) . Several commercial drugs targeting estrogen receptor alpha ( ERα ) have been developed to treat breast cancers and other diseases [18] . However , therapy from these drugs such as Tamoxifen ( IUPAC name: ( Z ) -2-[4- ( 1 , 2-diphenylbut-1-enyl ) phenoxy]-N , N-dimethyl-ethanamine ) ( TAM ) is associated with undesirable side effects such as cardiac abnormalities [19] , thromboembolic disorders [20] , and ocular toxicity [21] . To identify off-targets of these SERMs and to attempt to elucidate the molecular mechanisms explaining their adverse effects , we searched for similar ligand binding sites across fold and functional space using a template for the known SERM binding site in ERα ( Protein Data Bank id: 1XPC ) . The search used a robust and scalable functional site prediction and comparison algorithm developed recently in our laboratory [22; Xie and Bourne , submitted] . Consequently , a similar inhibitor site is detected for Sacroplasmic Reticulum ( SR ) Ca2+ ion channel ATPase protein ( SERCA ) . The prediction is further verified with detailed protein–ligand docking and surface electrostatic potential analysis . Our prediction correlates well with clinical and biochemical observations , providing molecular insight into reducing the adverse effect of SERMs . The strategy used in this case study could be applied to discover off-targets for other commercially available pharmaceuticals and to repurpose existing drugs to treat different diseases [Xie , Kinnings , and Bourne , in preparation] . The process could also be included in a drug discovery and development pipeline in an effort to optimize drug leads and reduce unwanted side effects .
We first selected 10 , 730 structures from the RCSB Protein Data Bank ( PDB ) [23] as available structural models from the human proteome by mapping sequences of all PDB structures to Ensembl human proteins using a sequence identity above 95% ( see Methods ) . This resulted in non-human proteins being included , but with this high level of sequence identity , similarity of structure , and binding site in particular , can be assumed . These structures form 2 , 586 sequence clusters using a sequence identity of 30% . Of the 10 , 730 structures , we determined that 1 , 585 belong to the existing druggable proteome and correspond to 929 unique drug targets ( see Methods ) . Using a sequence identity cutoff of 30% , 825 structures are chosen as the representative set of human druggable proteins . It is estimated that the 825 structures represent approximately 40% of the known existing druggable targets and 10% of the human druggable genome ( Table S1 ) . However , we estimate that , based on sequence similarity , and hence structure similarity , six human proteins can be mapped to one drug target . Thus , by taking homology models into account , we estimate that the structural coverage of the druggable human genome is approximately 40% ( see Methods ) . Thus , while no means complete , we have a significant number of secondary protein targets that can be analyzed for off-site binding . In our case study , this was sufficient to find a candidate secondary target . These 825 structures were scanned for similarity to the ligand binding site of ERα ( PDB id: 1XPC ) with our sequence order independent profile–profile alignment algorithm ( SOIPPA ) [Xie and Bourne , submitted] . A Sarcoplasmic Reticulum ( SR ) Ca2+ ion channel ATPase protein ( SERCA ) ( PDB id: 1WPE ) showed the most significant similarity with p-value < 0 . 001 . The significance of SERCA was confirmed using a search with a larger set of 2 , 586 non-redundant human homologous proteins that included both druggable and non-druggable proteins ( Figure S1 ) . Besides SERCA , several other proteins exhibited significant similarity to the ERα ligand binding sites ( Table S2 ) . These sites are the subject of an ongoing investigation . It is noted that the SERCA structure 1WPE is from Oryctolagus cuniculus ( rabbit ) , as the human SERCA is absent from the PDB . A BLAST [24] search against the Ensembl version of the human genome [25] revealed that human and rabbit SERCA share 96% sequence identity without insertion or deletion . Moreover , the transmembrane domains and known ligand binding site residues were found to share 98% and 100% sequence identity , respectively ( Figure S2 ) . Therefore , the rabbit SERCA structure was used as a reasonable structural model for human SERCA throughout this study . SERCA plays a key role in regulating cytosolic calcium levels by accumulating calcium in the lumen [26] . SERCA consists of four SCOP domains [27]: a double-stranded beta-helix; a HAD-like domain; an ATP-binding domain N of metal cation-transporting ATPase; and a transmembrane domain M with an up–down bundle architecture . The predicated binding site is located in the all-helix transmembrane domain . It is noted that the ERα ligand binding domain itself adopts a similar all-helical orthogonal bundle architecture , but its similarity to SERCA cannot be established directly from structural comparison since the rmsd is 5 . 8 Å , the Z-score 3 . 7 , and the sequence identity 7 . 1% as determined by CE alignment [28] . A search through protein–ligand complex structures in the PDB reveals that two co-crystal inhibitors , thapsigargin ( TG1 ) and 2 , 5-ditert-butylbenzene-1 , 4-diol ( BHQ ) ( PDB id: 2AGV ) [29] bind in the vicinity of the predicted binding sites and are in contact with part of the predicted site ( Figure 1 ) . If amino acid residues whose atomic distances to the inhibitors are less than 6 . 0 Å are considered as the binding site , 30% of residues overlap between known and predicated site . Thus SERM is predicted to bind to a site similar to these two inhibitors . It is suggested that the two calcium ions , which bind in the region of the putative binding site ( Figure 2 ) , are prevented from binding by SERM with consequences that are outlined subsequently . In a reverse search , we scanned the set of proteins comprising the druggable proteome against the TG1 and BHQ sites of SERCA . ERα receptors were ranked at the top with a p-value < 0 . 001 for the TG1 site but a p-value of 0 . 052 for the BHQ site ( Figure 3 ) . Figure 4 illustrates that the SERM binding site is part of the predicated site . This complementarity of binding confirms the similarity between the SERCA inhibitor and the SERM binding site with high confidence . As a further test , we compared the electrostatic potential ( ES ) between the binding site in ERα and SERCA , as ES is an important identifier of ligand binding [30] . As seen in Figure 5 , the SERM binding site is relatively negatively charged . Similarly , the binding pocket of SERCA also shows a negative potential . These observations are consistent with the binding site similarity predicated from SOIPPA at the residue level . To understand the molecular mechanism of SERM's inhibition on SERCA , we performed a detailed protein–ligand interaction study by docking a series of SERM molecules to both SERCA and ERα proteins with eHits 6 . 2 [31] and Surflex 2 . 1 [32] docking software . These two free software packages were selected because of their relatively high accuracy , speed , and ease of use in a large-scale study [33 , 34] . Moreover , these two packages adopt different strategies during conformational search , offering independent confirmation of our findings . The conformational search in eHits is performed by breaking molecules into rigid fragments and docking them independently . The final binding pose is determined by linkage and optimization of the reconstructed ligands from the fragments . The conformational search in Surflex relies on generation of an idealized binding site ligand called protomol and alignment of the ligand to the protomol to achieve maximum molecular similarity . The most similar poses are subject to local energy minimization . Both eHits and Surflex use empirical scoring functions but with different terms and parameterization . The molecules studied included TAM and its metabolite 4-hydroxytamoxifen ( OHT ) , raloxifene ( RAL ) , bazedoxifene ( BAZ ) , ormeloxifene ( ORM ) , and lasofoxifene ( LAS ) . Figure 6 depicts their chemical structures . These molecules consist of two moieties: a phenoxy-ethanamine moiety ( N-moiety ) and a more hydrophobic fragment with two benzene rings ( C-moiety ) . Bonds to break these two moieties are marked by the red bar in Figure 6 . Table 1 shows docking scores from eHits . Both of the predicated inhibition sites ( TG1 and BHQ ) from SERCA are able to bind to TAM and its analogs . However , it is more likely that the TG1 site is the preferred off-target binding site for SERMs because its binding affinity is consistently greater than that of the BHQ site . Analysis of their binding poses when bound to the TG1 site indicates that the N-moiety of these molecules adopt similar binding poses with a specific salt interaction between Glu255 and the amine groups , as shown in Figure 7 . This charge neutralization is also observed in ERα when binding to SERMs and is considered the origin of the antiestrogenic effect of SERMs [35] . The binding poses of the C-moiety are more variable due to different conformational constraints . Some of them , such as TAM and BAZ , may have stronger aromatic interactions with the receptor than other SERMs . The predicated binding poses are cross-docked with both eHits 6 . 2 and Surflex 2 . 1 and show consistent patterns in binding poses and affinities . Binding affinity alone may not be conclusive because of the poor accuracy of the scoring function [16 , 36 , 37] . However , binding poses are able to be predicated reasonably accurately by most docking programs [16 , 38] . To leverage the strength and weakness of existing docking algorithms , we docked more than 1 , 000 decoy molecules for both the N-moiety and the C-moiety into the predicated SERCA off-target and primary ERα target sites , respectively . In this way , similar binding sites will show a strong docking score correlation independent of the scoring function , assuming that binding poses are consistent between the two sites . Alternatively , the correlation will be weak if the docking pose is random or the two sites are dissimilar . There were two reasons to break down the molecule into N- and C-moieties . First , the conformational search space of the molecules used in docking will be reduced by using a small fragment and is more likely to predicate their binding pose consistently . Second , the N-moiety has been predicated to have more favorable interactions than the C-moiety . The separate evaluation of their binding affinities will further verify the predicated binding poses and provide insight into designing highly specific SERMs to minimize off-target binding . As shown in Figure 8 , the correlation of the eHits docking scores of the N- and C-moiety analogs between the SERM and SERCA TG1 sites is strong relative to that of the BHQ site ( Figure 8A and 8C ) . Taking into account that the correlation between docking scores and experimental affinity is around 0 . 4–0 . 6 for most of docking programs [16 , 36 , 37] , the N-moiety correlation coefficient of 0 . 46 between SERM and TG1 sites is close to the limit of docking score accuracy . Moreover , the docking score distribution is centered around an optimal correlation line between two identical binding sites ( green line in Figure 8 ) with a standard deviation of 0 . 88 , much less than the 2 . 33 between the SERM and the BHQ site . Docking score correlations from Surflex show the same pattern as those from eHits . The N-moiety correlation coefficients are 0 . 50 and 0 . 44 for the TG1 and BHQ sites , respectively . The corresponding standard deviations are 1 . 51 and 2 . 14 , respectively . These results are consistent with the predicated binding poses and the relative binding affinities , further supporting the notion that the SERCA TG1 site is similar to the SERM binding site . There is experimental evidence to support our theoretical off-target binding site . It has been shown that pretreatment with TAM inhibits TG1′s effect in increasing the intracellular Ca2+ concentration [39–41] . One potential mechanism for this observed effect is that TAM binds to the same site as TG1 , thus blocking its effect , although it is not clear how TG1 inhibition of SERCA leads to an increase in Ca2+ concentration from these experiments . Our findings indicate that TAM is able to bind directly to the TG1 site and for the first time suggests an inhibition mechanism in atomic detail . Moreover , residue 309 ( Glu-309 ) is included in the BHQ binding sites . It is known that this residue acts as a cytoplasmic gate that allows the release of calcium ions [29] . It is postulated that if TAM interacts with Glu-309 or its interacting partner , it can affect the function of Glu-309 and thus the transport function of the calcium pump as a whole . Maintaining the level of calcium in the cell is critical to normal cell function . Previous clinical findings have shown that TAM therapy is associated with undesirable side effects such as cardiac abnormalities [19] , thromboembolic disorders [20] , and ocular toxicity [21] . Recent physiological studies suggest that TAM [42 , 43] , anti-estrogens/β-estradiol [44] , phytoestrogens [45] , and ovarian sex hormones [46] play important roles in regulating calcium uptake activity of cardiac SR . Given that the gradient concentration of calcium ions in SR is important for muscle contraction [26] , it is possible that the cardiac abnormality is caused by its inhibition of SERCA . It has been observed that TAM significantly reduces intracellular calcium concentration and release in the platelets , which is correlated with platelet adhesion and aggregation [47] . The loss of calcium homeostatis in the platelets may originate from inhibition of SERCA by TAM . In addition , there is evidence that diethylstilbestrol increases intracellular calcium in lens epithelial cells by inhibiting SERCA [48] and cataracts result from TG1-inhibited SERCA upregulation [49 , 50] . We believe the evidence for off-site binding of SERMs to SERCA and the proposed impact that it has on calcium homeostatis leads to the reported adverse effects . As shown in Table 1 , in general , the off-target TG1 binding affinity increases with an increase in primary main target binding affinity . However , the binding affinity difference for RAL between target and off-target binding is larger than TAM . Thus , it is expected that RAL exhibits less competitive binding of the SERCA protein , resulting in less adverse effects than TAM . This predication is consistent with clinical studies that show RAL has less adverse effect of thromboembolic disorder and cataract formation than TAM [18] . Among the molecules studied , ORM is predicated as one of the least SERCA competitive binding therapeutics . Clinical studies indeed show that ORM is safe for long-term usage with less adverse effects [51] . LAS and BAZ are newly developed breast cancer therapeutics currently in clinical trial . Although they show the strongest binding to ERα , they also potentially bind strongly to off-targets . It is expected that LAS and BAZ may have similar competitive binding profiles to RAL when binding to SERCA . If indeed this competitive binding and the associated side effects prove to be consistent , the value of this approach to lead optimization and drug development will be further supported .
SERMs are potent anti-cancer drugs . By combining , first , functional site similarity searching on a structural proteome-wide scale , second , small molecule screening , and , finally , protein–ligand docking , a potential mechanism for the adverse effect of SERMs has been established . Specifically , we provide evidence for off-target binding of SERMs , resulting in the inhibition of a SERCA transmembrane domain which leads to a disruption in calcium homeostasis . The computational prediction presented here is supported by experimental observations from in vitro and clinical studies . Our methodology provides opportunities to develop further refined SERMs with fewer side effects . On a larger scale there exists the opportunity to explore off-targets binding for any existing pharmaceutical or compound of pharmaceutical interest for which a 3D structural model is available . At this time we are beginning to systematically analyze all commercially available pharmaceuticals in an effort to explain any observed side effects .
Sequences of all PDB [23] structures are mapped to Ensembl [25] human protein sequences ( 43 , 738 proteins ) using BLAST [24] . A total of 10 , 730 PDB structures map to 3 , 158 Ensembl human proteins with a sequence identity above 95% . These 10 , 730 structures are considered as structural models for the human proteome . They form 2 , 586 sequence clusters when using a sequence identity of 30% . The existing druggable human proteome is determined by mapping Ensembl [25] human protein sequences against all sequences of drug targets from Drugbank [57] using BLAST [24] . Homologous sequences from the human proteome , with e-values less than 0 . 001 , constitute the druggable human proteome—a total of 13 , 865 human proteins corresponding to 2 , 002 unique drug targets . Among the 10 , 730 human protein structural models , 1 , 585 belong to the existing druggable human proteome and correspond to 929 unique drug targets . 825 sequence clusters are formed after clustering the druggable structures with a sequence identity of 30% . One structure is randomly selected from each of the clusters to constitute a representative set of druggable structures . These structures represent approximately 10% of the complete druggable human proteome and 40% of existing drug targets . A flow chart depicting this selection process is given in Figure S3 . Protein structures are represented by Delaunay tessellation of Cα atoms and characterized with geometric potentials [22] . The similar residue clusters for any protein to a ligand binding site are detected with a SOIPPA algorithm [Xie and Bourne , submitted] . To evaluate the p-value for the similarity score calculated from the site comparison method , we estimate the background distribution using a non-parametric method . First , the drug target of interest is compared against the 825 representative sets of human druggable structures . We remove those hits that are in the same fold as the query because they will probably be true positives . Then a kernel density estimator is used to estimate the background probability distribution of the binding site alignment scores . A Gaussian kernel with fixed bandwidth is used . The optimal bandwidth is estimated from the data by using a least square cross-validation approach [58] . Finally , this estimated density function is used to calculate a p-value for the particular pair of ligand sites being compared . Protein–ligand docking is conducted using the eHits [31] and Surflex 2 . 1 [32] software packages . Default parameter settings were applied when using Surflex . For eHits the accuracy is set to the highest ( accuracy = 6 ) during docking . The highest accuracy means that the most extensive conformational search is performed to determine the ligand binding pose and affinity . Structures of ERα and SERCA proteins were downloaded from the RCSB PDB [23] . The PDB ids of ERα proteins are 1xpc , 3ert , 1r5k , 1err , and 2jfa . The PDB ids of SERCA proteins are 2agv , 1xp5 , 1wpg , 1iwo , 2c88 , and 2eat . The decoy molecules are generated by querying the N- and C-moiety of TAM against the ZINC database [59] using a 2D subgraph similarity search through the associated Web site ( http://blaster . docking . org/zinc/ ) . The query molecular structures of the N- and C-moieties are shown in Figure 9 . The query generates 1 , 638 hits to the N-moiety and 1 , 128 hits to the C-moiety , respectively . The electrostatic potential of the molecule is calculated using the Gemstone interface ( http://gemstone . mozdev . org ) to the Adaptive Poisson-Boltzman Solver ( APBS ) [60] . For calculation of the original ERα drug target ( PDB id: 1XPC ) , the dielectric constant is set to 2 . 0 for the protein and 78 . 54 for the solvent . For the off-target SERCA ( PDB id: 2AGV ) , the constant is set to 4 . 0 for the protein and 78 . 54 for the solvent , as this molecule is embedded in the phospholipid bilayer . Other parameters are set to defaults as provided by the Gemstone interface . Visualization of the structures is performed using Chimera [61] .
The protein accession numbers for estrogen receptor alpha are UniProt ( http://www . pir . uniprot . org/ ) P03372 and Protein Data Bank ( http://www . rcsb . org/pdb/home/home . do ) id 1XPC , and for sarcoplasmic/endoplasmic reticulum calcium ATPase 1 are UniProt P04191 and PDB id 2AGV , 1WPE . | Early identification of the side effects of preclinical and commercial drugs is crucial in developing highly efficient therapeutics , as unexpected side effects account for one-third of all drug failures in drug development and lead to drugs being withdrawn from the market . Compared with the experimental identification of off-target proteins that cause side effects , computational approaches not only save time and costs by providing a candidate list of potential off-targets , but also provide insight into understanding the molecular mechanisms of protein–drug interactions . In this paper we describe an integrated approach to identifying similar drug binding pockets across protein families that have different global shapes . In a case study , we elucidate a possible molecular mechanism for the observed side effects of selective estrogen receptor modulators ( SERMs ) , which are widely used to treat and prevent breast cancer and other diseases . The prediction provides molecular insight into reducing the side effects of SERMs and is supported by clinical and biochemical observations . The strategy used in this case study is being applied to discover off-targets for other commercially available pharmaceuticals and to repurpose existing safe pharmaceuticals to treat different diseases . The process can be included in a drug discovery pipeline in an effort to optimize drug leads , reduce unwanted side effects , and accelerate development of new drugs . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods",
"Supporting",
"Information"
] | [
"biotechnology",
"oncology",
"computational",
"biology",
"biophysics",
"homo",
"(human)"
] | 2007 | In Silico Elucidation of the Molecular Mechanism Defining the Adverse Effect of Selective Estrogen Receptor Modulators |
In female fruit flies , Sex-lethal ( Sxl ) turns off the X chromosome dosage compensation system by a mechanism involving a combination of alternative splicing and translational repression of the male specific lethal-2 ( msl-2 ) mRNA . A genetic screen identified the translation initiation factor eif4e as a gene that acts together with Sxl to repress expression of the Msl-2 protein . However , eif4e is not required for Sxl mediated repression of msl-2 mRNA translation . Instead , eif4e functions as a co-factor in Sxl-dependent female-specific alternative splicing of msl-2 and also Sxl pre-mRNAs . Like other factors required for Sxl regulation of splicing , eif4e shows maternal-effect female-lethal interactions with Sxl . This female lethality can be enhanced by mutations in other co-factors that promote female-specific splicing and is caused by a failure to properly activate the Sxl-positive autoregulatory feedback loop in early embryos . In this feedback loop Sxl proteins promote their own synthesis by directing the female-specific alternative splicing of Sxl-Pm pre-mRNAs . Analysis of pre-mRNA splicing when eif4e activity is compromised demonstrates that Sxl-dependent female-specific splicing of both Sxl-Pm and msl-2 pre-mRNAs requires eif4e activity . Consistent with a direct involvement in Sxl-dependent alternative splicing , eIF4E is associated with unspliced Sxl-Pm pre-mRNAs and is found in complexes that contain early acting splicing factors—the U1/U2 snRNP protein Sans-fils ( Snf ) , the U1 snRNP protein U1-70k , U2AF38 , U2AF50 , and the Wilms' Tumor 1 Associated Protein Fl ( 2 ) d—that have been directly implicated in Sxl splicing regulation .
Translation initiation is mediated by the binding of a pre-initiation complex to the 5′ cap of the mRNA ( reviewed in [1] , [2] ) that in turn recruits the small subunit of the 40S ribosome to the mRNA . The pre-initiation complex consists of the cap binding protein , eIF4E , and a scaffolding protein , eIF4G , which mediates interactions with various components of the 40S initiation complex . In many organisms there is also a third protein in the complex , eIF4A , an ATP dependent RNA helicase . Modulating eIF4E activity appears to be a key control point for regulating translation . One of the most common mechanisms of regulation is by controlling the association eIF4E with eIF4G . Factors such as poly-A binding protein that promote the association between eIF4E and eIF4G activate translation initiation , while factors such as the 4E-binding proteins ( 4E-BPs ) that block their association , inhibit initiation [3] , [4] . Although eIF4E's primary function in the cell is in regulating translation initiation , studies over the past decade have revealed unexpected activities for eIF4E at steps prior to translation . Among the more surprising findings is that there are substantial amounts of eIF4E in eukaryotic nuclei [5]–[9] . One role for eIF4E in the nucleus is the transport of specific mRNAs , like cyclin D1 , to the cytoplasm [10] . This eIF4E activity is distinct from translation initiation since an eIF4E mutation that prevents it from forming an active translation complex still allows cyclin D1 mRNA transport [8] . The transport function of eIF4E is modulated by at least two other proteins , PML and PRH [11] , [12] . While PML seems to be ubiquitously expressed , PRH is found only in specific tissues [13] . In addition , the intracellular distribution of eIF4E exhibits dynamic changes during Xenopus development [9] . These observation raise the possibility that eIF4E might have additional functions in the nucleus during development . Consistent with this idea , we show here that eIF4E plays a novel role in the process of sex determination in Drosophila melanogaster . Sex determination in the fly is controlled by the master regulatory switch gene Sex-lethal ( Sxl ) ( reviewed in [14]–[16] ) . The activity state of the Sxl gene is selected early in development by an X chromosome counting system . The target for the X/A signaling system is the Sxl establishment promoter , Sxl-Pe [17] . When there are two X chromosomes , Sxl-Pe is turned on , while it remains off when there is a single X chromosome . Sxl-Pe mRNAs encode RRM type RNA binding proteins which mediate the transition from the initiation to the maintenance mode of Sxl regulation by directing the female-specific splicing of the first pre-mRNAs produced from a second , upstream promoter , the maintenance promoter , Sxl-Pm [18] , [19] . Sxl-Pm is turned on before the blastoderm cellularizes , just as Sxl-Pe is being shut off . In the presence of Sxl-Pe proteins , the first Sxl-Pm transcripts are spliced in the female-specific pattern in which exon 2 is joined to exon 4 ( see Figure 1A ) . The resulting Sxl-Pm mRNAs encode Sxl proteins that direct the female specific splicing of new Sxl-Pm pre-mRNAs and this establishes a positive autoregulatory feedback loop that maintains the Sxl gene in the “on” state for the remainder of development . In male embryos , which lack the Sxl-Pe proteins , the Sxl-Pm pre-mRNAs are spliced in the default pattern , incorporating the male specific exon 3 ( Figure 1A ) . This exon has several in-frame stop codons that prematurely truncate the open reading frame so that male specific Sxl-Pm mRNAs produce only small non-functional polypeptides . As a consequence the Sxl gene remains off throughout development in males . In females , Sxl orchestrates sexual development by regulating the alternative splicing of transformer ( tra ) pre-mRNAs [20]–[23] . Like Sxl , functional Tra protein is only produced by female-specific tra mRNAs , while mRNAs spliced in the default , male pattern encode non-functional polypeptides . Sxl also negatively regulates the dosage compensation system , which is responsible for hyperactivating X-linked transcription in males , by repressing male-specific lethal-2 ( msl-2 ) . Sxl represses msl-2 by first blocking the splicing of an intron in the 5′ UTR of the msl-2 pre-mRNA ( see Figure 1B ) , and then by inhibiting the translation of the mature mRNA [24]–[31] . In addition , there are two other known targets for Sxl translational repression . One is the Sxl mRNA itself . Sxl binds to target sequences in the Sxl 5′ and 3′ UTRs and downregulates translation . It is thought that this negative autoregulatory activity provides a critical homeostasis mechanism that prevents the accumulation of excess Sxl protein . This is important as too much Sxl can disrupt development and have female lethal effects [32] . The other known target is the Notch ( N ) mRNA [33] . Sxl-dependent repression of N mRNA translation is important for the elaboration of sexually dimorphic traits in females . Like msl-2 and Sxl , translational repression appears to be mediated by Sxl binding to sites in the N UTRs . Translational repression of msl-2 mRNA by Sxl is thought to involve two separate mechanisms acting coordinately . Binding sites for Sxl in the unspliced intron in the 5′ UTR and in the 3′UTR of msl-2 are required for complete repression [25] , [26] . Sxl binding to the 5′UTR blocks recruitment of the 40S pre-initiation complex [31] , [34] . While factors that act with Sxl at the 5′UTR of msl-2 have yet to be identified , repression by the 3′UTR requires Sxl , PABP and a co-repressor UNR [35]–[37] . Somewhat unexpectedly , this complex does not affect recruitment of eIF4E or eIF4G to the 5′ end . Instead it prevents ribosomes that do manage to attach to the msl-2 mRNA from scanning [31] , [38] . Although eIF4E does not appear to be a key player in the translational repression of msl-2 mRNAs , we report here that it has an important role in the process of sex determination in Drosophila . We find that eIF4E activity is required in females to stably activate and maintain the Sxl positive autoregulatory feedback loop and to efficiently repress msl-2 . Surprisingly , this requirement for eIF4E activity in fly sex determination is in promoting the female-specific splicing of the Sxl and msl-2 transcripts , not in translational regulation .
In previous studies we examined the biological properties of a truncated Sxl protein , Sx-N , that contains both RRM RNA binding domains , but is missing 40 amino acids from the N-terminus [39] . We found that the splicing activity of Sx-N is impaired; it can not direct the female-specific splicing of tra and has substantially reduced autoregulatory activity . However , the truncated protein is able to inhibit the translation of msl-2 mRNA and kills males even in the absence of a wild type Sxl gene . As would be expected if the male lethal effects of Sx-N are due to repression of msl-2 mRNA translation , hsp83:Sx-NΔ males can be fully rescued by an hsp83:msl-2 transgene that lacks the Sxl binding sites in the 5′ and 3′ UTRs . With the aim of discovering factors important for Sxl dependent repression of msl-2 we screened for deletions that dominantly suppress the male lethal effects ( in a Sxl− background ) of a transgene , hsp83:Sx-NΔ , that constitutively expresses the truncated Sx-N protein . We then identified the interacting locus by testing mutations mapping to the suppressing deletion . We anticipated that genes recovered in this screen would fall into two general classes . In the first would be genes required for efficient expression of Sx-N by the transgene . Consistent with this expectation , one of the suppressing mutations was the heat shock factor , hsf . Genes in the second class would be required for efficient repression of msl-2 by the truncated Sx-N protein . In this group we expected to find factors required by Sxl to inhibit msl-2 translation; however , since the Sxl binding sites in the msl-2 5′ UTR intron are needed to completely repress translation , we anticipated that we might also recover genes that collaborate with Sxl to block the removal of this intron [25] , [26] , [28] , [31] . One of the candidate translation factors recovered in the screen was the eif4e gene , which encodes the cap binding protein . Three independent alleles of eif4e were tested . In an otherwise wild type background less than one in 103 Sxl− males carrying the hsp83:Sx-NΔ transgene survive . By contrast , when the hsp83:Sx-NΔ; Sxl− males were also heterozygous for an eif4e mutation , between 2% and 9% of the transgenic males survived depending upon the allele . Since Sxl-dependent repression of msl-2 translation in vitro is independent of the cap and does not seem to be mediated through interactions with eIF4E [34] , [38] , it was surprising that eif4e was recovered in our screen . However , it seemed possible that an in vivo requirement for eif4e activity might be bypassed in in vitro translation systems . In this case , the levels of Msl-2 should increase in hsp83:Sx-NΔ transgene males when they are heterozygous for one of the eif4e mutations . However , testing whether eif4e mutations perturb Sx-N dependent translational repression of msl-2 mRNA in adults or at earlier stages of development is complicated by the male-lethal effects of the truncated Sxl protein . To circumvent this complication , we tested the effects eif4e on Sxl negative autoregulation as this can be done in females where Sx-N doesn't have such deleterious consequences . The endogenous Sxl-Pm mRNAs have one Sxl binding site in the 5′ UTR , while there can be eight or more in the 3′ UTR . Sxl binds to these sites and downregulates translation . Though the truncated Sx-N protein can also repress translation of Sxl-Pm mRNAs , its inhibitory effects are somewhat weaker than the full-length protein [39] . However , it is possible to detect Sx-N repression of endogenous Sxl mRNAs using the hsp83:Sx-NΔ transgene . This transgene expresses Sxl mRNAs that lack the 5′ Sxl binding site and most of the 3′ UTR binding sites , and as a consequence are less sensitive to repression than the endogenous mRNAs [39] . For this reason , Sx-N protein produced by the transgene preferentially represses translation of the endogenous mRNAs and in hsp83:Sx-NΔ transgenic females the amount of Sx-N is typically greater than the two major endogenous Sxl proteins . We compared the repression of the endogenous Sxl in hsp83:Sx-NΔ transgene females either wild type or heterozygous for eif4e . Figure 1C shows that in transgenic , wild type females the level of endogenous Sxl is less than Sx-N . Consistent with the results of the in vitro translation experiments , reducing eif4e activity does not have an obvious effect on repression of Sxl-Pm mRNAs by Sx-N and the ratio of the endogenous protein to Sx-N in eif4e/+ females remains similar to that in wild type females . With the caveat that Sxl may require a different set of accessory proteins to repress the translation of each of its target mRNAs , this finding does not support the idea that eIF4E functions as a co-factor in Sxl inhibition of msl-2 translation in vivo . The alternative possibility is that eif4e rescues the male lethal effects of Sx-N because Sxl requires eif4e activity to effectively prevent the splicing of the intron in the 5′ UTR of msl-2 pre-RNA . To test this idea , we examined the splicing pattern of msl-2 mRNA in three surviving Sxl-;eif4e/+; hsp83:Sx-NΔ males . In wild type females , Sxl efficiently blocks the splicing of the msl-2 5′ UTR intron and in most female mRNAs the intron is unspliced . In wild type males the 5′ intron is spliced out of most msl-2 mRNAs . As expected , we found that ectopically expressed Sx-N protein blocks the splicing of the 5′ intron and as shown for one of the surviving Sxl−;eif4e/+; hsp83:Sx-NΔ males in Figure S1 , msl-2 mRNA spliced in the female pattern is readily detected . However , we found that Sx-N wasn't able to fully inhibit the splicing of the 5′ intron , and roughly similar quantities of male spliced msl-2 mRNAs were also observed ( Figure S1 ) . Equivalent levels of male spliced msl-2 mRNAs were also found in both of the other Sxl−;eif4e/+; hsp83:Sx-NΔ males . Since the Sxl binding sites in the 5′ UTR are essential for efficient translational repression , Sx-N would not be able to completely block the translation of these male spliced msl-2 mRNAs . Though the results described in the previous section could explain why a small percentage of eif4e/+ males escape the lethal effects of Sx-NΔ , it is not possible to determine if the relative amount of male spliced msl-2 mRNA is increased compared to eif4e+ males because the controls don't survive . However , as it seemed possible that the effects of eif4e on Sxl dependent splicing might not be limited to msl-2 , we took advantage of a simple genetic test for genes involved in Sxl positive autoregulation . The initial activation of the positive Sxl autoregulatory loop in female embryos is sensitive to alterations in the dose of gene products that play a critical role in promoting the female specific splicing of Sxl-Pm pre-mRNAs . Because of this sensitivity , mutations in splicing factors like the U1A/U2B” snRNP protein Snf often show dominant female lethal interactions with Sxl [40]–[46] . If eif4e is required for female specific splicing , then dominant female lethal interactions with Sxl might be observed . In contrast , if eif4e is needed to help repress the translation of Sxl target mRNAs , then reducing eif4e activity should increase the translation of Sxl mRNAs and would be expected to suppress rather than enhance any female specific lethality . The results in Table 1 show that the former prediction is correct . All three of the eif4e alleles we tested , eif4e568 , eif4e587/11 , and eif4e715 , showed dominant female lethal interactions with the null mutation Sxlf1 ( Table 1 ) [47] . These eif4e alleles are P-element insertions and are thought to be hypomorphic mutations [48]–[49] . The weakest allele , eif4e568 , reduces female viability by a quarter , while female viability is reduced by a third to nearly a half for the two stronger alleles eif4e587/11 and eif4e715 . Although the reductions in female viability seen for the three eif4e mutations are not as great as that observed for the snf null allele J210 or the dominant negative allele 1621 , they are roughly equivalent to that seen for the hypomorphic allele JA2 ( Table 1 ) . In the experiments described above the eif4e/+ females were crossed to Sxlf1 males giving two classes of Sxlf1 progeny , those carrying the eif4e mutation and those with the wild type chromosome . We noticed that the viability of both classes of Sxlf1 progeny were affected equally ( data not shown ) suggesting that the lethality is predominantly the result of a lowered maternal contribution of eIF4E rather than a reduction in zygotic eIF4E . Consistent with this conclusion , when we did the reciprocal cross in which the eif4e mutation was introduced from the father and the Sxl mutation introduced from the mother , we found that the viability of Sxl−/+ females was close to that of wild type females ( not shown ) . To confirm that the female lethal interactions are due to a reduction in eif4e activity , we tested whether they can be rescued by an eif4e transgene . Two isoforms of eIF4E are expressed Drosophila . We introduced transgenes expressing each isoform into eif4e715/+ females and mated them to Sxlf1 males . We found that both could suppress the maternal effect lethal interactions between eif4E and Sxl ( data not shown ) . We also tested a second independent Sxl allele , Sxl7B0 [50] . Like Sxlf1 , Sxl7B0 exhibited dominant female lethal interactions with eif4e ( Table 1 ) . The null mutations Sxlf1 and Sxl7B0 discussed above eliminate both early Sxl initiation functions provided by Sxl-Pe mRNAs and late Sxl sex determination functions ( maintenance , sexual differentiation , and dosage compensation ) provided by the Sxl-Pm mRNAs [47] , [50] . While there are no known mutations that specifically eliminate only the late Sxl functions , the Sxlf9 mutation disrupts the initiation function of the Sxl-Pe transcripts [51]–[52] . If the reduction in eif4e activity impairs the female-specific splicing of Sxl-Pm pre-mRNAs , then eif4e mutations should have a smaller effect on the viability of flies carrying a Sxl mutation that only affects the Sxl-Pe pre-mRNAs as these transcripts do not require Sxl for proper splicing [53]–[54] . As can be seen in Table 1 , Sxlf9 differs from Sxlf1 and Sxl7B0 in that it shows only a weak female lethal interaction with eif4e mutations . It also interacts much less strongly with snf1621 than either of the Sxl null alleles ( data not shown ) . The female lethal interactions between Sxl and co-factors like snf that are critical for the female splicing of Sxl-Pm pre-mRNAs arise because the positive autoregulatory feedback loop is not properly set in motion [43]–[45] . However , there are no special requirements for these co- factors in the activation of Sxl-Pe by the X chromosome counting system or the splicing and translation of Sxl-Pe transcripts [53]–[54] . For these reasons , defects in Sxl accumulation are not observed in blastoderm stage embryos compromised for a sex-specific splicing co-factor . However , later in development , when protein expression depends upon female spliced Sxl-Pm mRNAs , the pattern of Sxl accumulation becomes abnormal . To determine if this is true for eif4e as well , we examined the expression of Sxl in blastoderm and post-blastoderm stage embryos . Consistent with the idea that eif4e functions downstream of Sxl-Pe , eif4e mutations have no apparent effect on the expression of Sxl from the Sxl-Pe mRNAs . As shown in Figure 2 and Table S1 , blastoderm stage progeny from eif4e−/+ and snf−/+ mothers crossed to Sxl−f1 fathers resemble wild type in that about 50% of the embryos ( females ) express Sxl protein ( compare panels A & B with C & D ) . While reducing eif4e activity does not perturb activation of Sxl by the X chromosome counting system , it does have a significant effect on the expression of Sxl in older embryos . In the wild type controls ( either w x w or w x Sxlf1 ) , high uniform levels of Sxl protein are observed in about 50% of the embryos , while a equal number show no staining ( panels E & F ) . For the dominant negative snf1621 allele only 11% of the embryos show the expected high uniform level of Sxl while Sxl expression in the remaining female embryos is either irregular or quite low ( Table S1 ) . As would be expected from the relative severity of the synthetic lethal interactions , the effects of the hypomorphic eif4e alleles on Sxl expression in post-cellular blastoderm embryos are not as strong as snf1621 . For both eif4e587/11 and eif4e713 about one third of the embryos ( or about two thirds of the females ) show a high uniform level of Sxl accumulation ( Table S1 ) . The remaining female embryos show either a patchy pattern of Sxl protein accumulation or only low levels of protein ( Figure 2G and 2H ) . These defects in Sxl expression in post-blastoderm embryos indicate that the Sxl autoregulatory feedback loop is not properly established in the female progeny of eif4e−/+ mothers . To confirm that the female lethal effects of eif4e are due to a failure to activate the Sxl positive autoregulatory loop we tested whether Sxl−/+ female progeny of eif4e−/+mothers can be rescued by three different gain-of-function Sxl alleles , SxlM1 , SxlM4 , and SxlM6 , that constitutively splice Sxl-Pm transcripts in the female mode [55] . As a positive control we generated an equivalent combination of SxlM1 and snf1621 . Females trans-heterozygous for each combination were mated with Sxlf1 males . As can be seen in Table S2 for the positive control , SxlM1 suppresses the maternal effect female lethal interactions between snf and Sxlf1 . Similarly , SxlM1 and both of the other gain-of-function alleles also suppress the maternal effect lethal interactions between eif4e587/11 and Sxlfl . In these crosses only half of the female progeny inherit the Sxl gain-of-function allele . As expected , most of the surviving females are the ones that carry the gain-of-function allele . If the positive autoregulatory loop is not properly activated when eif4e is compromised , we would expect to find male spliced Sxl transcripts in female blastoderm/early gastrula embryos . To examine the splicing pattern of Sxl-Pm transcripts specifically in female embryos during this period we took advantage of an X-linked Sxl-Pm splicing reporter . The splicing reporter has a Sxl genomic fragment extending across the regulated splice sites from exon 2 to exon 4 while exon 4 is fused to β-galactosidase sequences ( see Figure 3A: [56] ) . Expression of the fusion gene is driven by the hsp83 promoter . This promoter is activated in the zygote during the late syncytial blastoderm stage around the time when Sxl-Pm transcription commences [57] . Figure 3B shows that the transcripts spanning the regulated Sxl exon2-exon3-exon4 splicing cassette are spliced in the appropriate sex-specific pattern in control adult flies collected from a stock homozygous for the transgene: exon 2–4 in females and exons 2–3–4 in male . Sxlf1 or Sxl+ males carrying the splicing reporter were crossed to eif4e587/11/+ or control wild type females . To visualize the splicing of the regulated exon2-exon3-exon4 cassette when the autoregulatory feedback loop is first activated , we isolated RNA from 1–3 hr embryos and analyzed the structure of the transcripts expressed from the reporter by RT-PCR . When the mother is wild type we find that transcripts spanning the exon2-exon3-exon4 cassette are spliced exclusively in the female pattern ( Figure 3B ) . This is true not only for female embryos that have two wild type copies of Sxl ( fathers are Sxl+/Y ) , but also for female embryos that are heterozygous for the Sxlfl mutation ( fathers are Sxlf1/Y ) . A different result is obtained when the mother is heterozygous for eif4e587/11 ( Figure 3B ) . In this case , we detect not only female but also male spliced reporter RNAs . With this allele , male spliced RNAs are observed in both Sxlfl/+ embryos and in embryos that are wild type for Sxl . Similar results were obtained for snf1621 ( not shown ) . We also observed male spliced reporter RNAs in the female progeny of mothers heterozygous for two other eif4e alleles . However , for both of these eif4e alleles the male transcripts were only present when the female embryos were heterozygous for the Sxl mutation ( not shown ) . Two general mechanisms , one direct and the other indirect , could potentially account for the effects of eif4e on Sxl activation . In the direct mechanism , eif4e would function as a Sxl co-factor in the female specific processing of Sxl-Pm pre-mRNAs . In this case , reducing eif4e activity would compromise the female specific splicing of Sxl-Pm pre-mRNAs and prevent full activation of the positive autoregulatory feedback loop when the loop is first being initiated . In the second , eif4e would be required at a point subsequent to the splicing of the Sxl-Pm pre-mRNAs . For example , it may be needed in the cytoplasm for the efficient translation of Sxl-Pm mRNAs , or it might function in their nuclear export . In this scenario , the expression of Sxl proteins from the newly synthesized Sxl-Pm mRNAs would be impaired and sub-optimal levels of Sxl-Pm proteins would be produced . As a consequence , when the Sxl-Pe proteins decay , there would be an insufficient amount of Sxl remaining to stably maintain the positive autoregulatory feedback loop , and splicing would gradually switch from the female to the male pattern . Though our experiments with the splicing reporter suggest an immediate rather than a gradual effect on splicing of the Sxl-Pm transcripts , we cannot rule out the possibility that there is some disruption in the export or translation of Sxl-Pm mRNAs during the initial activation of the positive autoregulatory feedback loop . Moreover , consistent with the possible importance of post-splicing steps in Sxl activation , Stitzinger et al [58] found female lethal interactions with Sxl when mothers are simultaneously heterozygous for mutations in aspartyl tRNA synthetase and snf . Although the aspartyl tRNA synthetase mutants differ from eif4e in that they do not show female lethal interactions with Sxl on their own , the fact that reductions in the maternal dose of this synthetase can affect the activation of the autoregulatory loop lends credence to a post-splicing function . For these reasons we sought experimental paradigms in which we could assay for eif4e induced perturbations in Sxl dependent female-specific splicing under conditions in which the autoregulatory loop had already been “fully” activated and Sxl proteins were present at wild type levels . In previous studies on snf we found that though there is substantial female lethality when snf1621/+ mothers are mated to Sxl− fathers , the surviving snf1621/Sxl− trans-heterozygous females are morphologically normal , fertile , and express wild type levels of Sxl protein . When we examined the splicing of the Sxl-Pm mRNAs in these surviving females using RT-PCR primer sets that give products spanning the regulated exon2-exon3-exon4 cassette , we found that unlike wild type females ( which give only female spliced transcripts: exons 2–4 ) we could often detect a very low level of male spliced transcripts ( exons 2–3–4 ) in these snf1621/Sxl− trans-heterozygous adult females ( not shown: see snf1621 Sxlf1/++ in Figure 4B ) . We reasoned that the snf1621Sxlf1/++ heterozygous mutant combination might provide a suitable sensitized background to test whether eif4e activity is required for Sxl dependent pre-mRNA splicing . Before assaying the splicing of Sxl-Pm transcripts in adult females triply heterozygous for snf1621 , Sxlf1 , and eif4e , we examined Sxl protein expression in these females . We anticipated that as long as the level of female spliced Sxl mRNAs remained well above some threshold critical for maintaining the positive autoregulatory feedback loop , the homeostasis mechanism provided by Sxl negative autoregulation of Sxl mRNA translation would ensure that Sxl levels would be maintained close to that in wild type . With the possible caveat that there may be tissue specific variations in Sxl levels that can't be detected by this assay , Figure 4A shows that this expectation is correct . We find that the level of Sxl protein in the triple mutant combinations with two different eif4e alleles is equivalent to that seen in control snf1621 Sxlf1/++ ( ane + ) adult females . We next asked if a reduction in eif4e activity in the sensitized snf1621Sxlfl/++ background had any effect on the splicing of Sxl-Pm pre-mRNAs . For this purpose , we used a primer set that simultaneously amplifies both the male ( exon 2–3–4 ) and female ( exon 2–4 ) spliced Sxl mRNAs . This allows us to directly compare the relative ratio of female to male spliced mRNAs in each genetic background . Figure 4B shows that the very modest defects in the female specific splicing of Sxl-Pm pre-mRNAs evident in snf1621Sxlf1/++ females are clearly exacerbated when eif4e activity is reduced . For both eif4e alleles there is a marked increase in the amount of male-spliced Sxl-Pm mRNAs compared to the snf1621Sxlf1/++ control . We used this same sensitized background to examine the effects of reducing eif4e activity on the splicing of the intron in the 5′ UTR of msl-2 mRNAs . As illustrated in Figure 4C , Sxl blocks the splicing of the 5′ UTR intron so that it is retained in most msl-2 mRNAs in females , while this intron is spliced out efficiently in males . In control snf1621Sxlf1/++ females the female-specific splicing of the msl-2 mRNA is partially compromised and , we observe a nearly equal mixture of female and male spliced transcripts . As observed for Sxl-Pm splicing , reducing eif4e activity in this sensitized background further disrupts the female specific splicing of msl-2 mRNAs . In addition to demonstrating a role for eif4e in the splicing of a second Sxl target pre-mRNA , these findings provide additional evidence that the male lethal effects of the hsp83:Sx-NΔ transgene are suppressed because eif4e mutations perturb the female specific splicing of msl-2 mRNAs . The results in the previous sections demonstrate that the modest defects in Sxl and msl-2 pre-mRNA splicing evident in a sensitized snf1621 Sxlf1/++ background are significantly enhanced by reducing eif4e activity . We wondered whether splicing defects are also observed in eif4e/+ females that are wild type for both snf and Sxl . To test this possibility , we examined the splicing of transcripts from the endogenous Sxl gene and the Sxl splicing reporter in females heterozygous for two different eif4e alleles . When we used primers that allow us to visualize simultaneously both the male and female spliced Sxl mRNAs from either the endogenous gene ( Figure 4D ) or from the splicing reporter ( not shown ) , only female spliced Sxl mRNAs were observed in wild type females . In contrast , a very small amount of Sxl mRNA spliced in the male pattern could be detected from the endogenous gene ( Figure 4D ) and also from the splicing reporter ( not shown ) in females heterozygous for eif4e568 or for eif4e587/11 . To confirm that male spliced Sxl mRNAs from the endogenous gene are present in these eif4e/+ females we used RT primers from exon 5 and then PCR amplified using a primer from the male exon and a primer from exon4 . Figure 4E shows that male spliced Sxl mRNAs from the endogenous gene are readily evident in both eif4e568/+ and eif4e587/1/+ females , but not in wild type . Figure 4F shows that male spliced Sxl mRNAs from the reporter are also present in these eif4e heterozygous females , while there is little male spliced reporter mRNAs in control wild type females . To determine whether the effects of eif4e on sex-specific splicing are general or only restricted to Sxl dependent alternative splicing we examined the splicing of doublesex ( dsx ) mRNAs . The dsx gene is downstream of Sxl and like Sxl its transcripts are sex-specifically spliced . However , female-specific splicing of dsx mRNA is dependent upon tra and tra-2 , not Sxl ( reviewed in [14]–[16] ) . We used primer sets that would RT-PCR amplify either female or male spliced dsx mRNAs isolated from either wild type or eif4e/+ females . As expected , wild type females produce only female , not male products ( Figure S2 ) . Significantly , females heterozygous for eif4e also produce only female dsx mRNAs . The results described in the previous sections show that eif4e is required for Sxl splicing . Since eif4e is known to function in translation initiation , it might be needed for the synthesis of some limiting Sxl co-factor . In this scenario , the amount of this splicing co-factor would drop below some critical threshold when eif4e activity is reduced , and this would impair the ability of Sxl to regulate splicing . Alternatively , eif4e itself could be the Sxl splicing co-factor . This latter model makes several predictions that we have tested below .
The RNA binding protein Sxl orchestrates sexual development by controlling gene expression post-transcriptionally at the level of splicing and translation . To exert its different regulatory functions Sxl must collaborate with sex-non-specific components of the general splicing and translational machinery . In the studies reported here we present evidence that one of the splicing co-factors is the cap binding protein eIF4E . We initially identified eif4e in a screen for mutations that dominantly suppress the male lethal effects induced by ectopic expression of a mutant Sxl protein , Sx-N , which lacks part of the N-terminal domain . The Sx-N protein is substantially compromised in its splicing activity , but appears to have closer to wild type function in blocking the translation of the Sxl targets msl-2 and Sxl-Pm . As the male lethal effects of Sx-N ( in an Sxl- background ) are due to its inhibition of Msl-2 expression [39] we anticipated that general translation factors needed to help Sxl repress msl-2 mRNA would be recovered as suppressors in our screen . Indeed , one of the suppressors identified was eif4e . However , consistent with in vitro experiments , which have shown that Sxl dependent repression of msl-2 mRNA translation is cap independent [34] , we found that eif4e does not function in Sxl mediated translational repression of at least one target mRNA in vivo . Instead , our results indicate that eif4e is needed for Sxl dependent alternative splicing and argue that it is this splicing activity that accounts for the suppression of male lethality by eif4e mutations . In wild type females , Sxl protein blocks the splicing of a small intron in the 5′ UTR of the msl-2 pre-mRNA . This is an important step in msl-2 regulation because the intron contains two Sxl binding sites that are needed by Sxl to efficiently repress translation of the processed msl-2 mRNA . When this intron is removed repression of msl-2 translation by Sxl is incomplete [25]–[28] and this would enable eif4e/+ males to escape the lethal effects of the Sx-N transgene . Several lines of evidence support the conclusion that eif4e is required for Sxl dependent alternative splicing . One comes from our analysis of the dominant maternal effect female lethal interactions between eif4e and Sxl . The initial activation of the Sxl positive autoregulatory feedback loop in early embryos can be compromised by a reduction in the activity of splicing factors like Snf , Fl ( 2 ) d , and U1-70K , and mutations in genes encoding these proteins often show dose sensitive maternal effect , female lethal interactions with Sxl . Like these splicing factors , maternal effect female lethal interactions with Sxl are observed for several eif4e alleles . Moreover , these female lethal interactions can be exacerbated when the mothers are trans-heterozygous for mutations in eif4e and the splicing factors snf or fl ( 2 ) d . Genetic and molecular experiments indicate that female lethality is due to a failure in the female specific splicing of Sxl-Pm mRNAs . First , female lethality can be rescued by gain-of-function Sxl mutations that are constitutively spliced in the female mode . Second , transcripts expressed from a Sxl-Pm splicing reporter in the female Sxl−/+ progeny of eif4e/+ mothers are inappropriately spliced in a male pattern at the time when the Sxl positive autoregulatory loop is being activated by the Sxl-Pe proteins . While splicing defects are evident in these embryos at the blastoderm/early gastrula stage , obvious abnormalities in expression of Sxl protein are not observed until several hours later in development . Though this difference in timing would favor the idea that eif4e is required for splicing of Sxl-Pm transcripts rather than for the export or translation of the processed Sxl-Pm mRNAs , we can not exclude the possibility that there are subtle defects in the expression of Sxl protein at the blastoderm/early gastrula stage that are sufficient to disrupt splicing regulation during the critical activation phase yet aren't detectable in our antibody staining experiments . However , evidence from two different experimental paradigms using adult females indicates that this is likely not the case . In the first , we found that reducing eif4e activity in a sensitized snf1621 Sxlf1/++ background can compromise Sxl dependent alternative splicing even though there is no apparent reduction in Sxl protein accumulation . In this experiment we took advantage of the fact that once the positive autoregulatory feedback loop is fully activated a homeostasis mechanism ( in which Sxl negatively regulates the translation of Sxl-Pm mRNAs ) ensures that Sxl protein is maintained at the same level even if there are fluctuations in the amount of female spliced mRNA . While only a small amount of male spliced Sxl-Pm mRNAs can be detected in snf1621 Sxlf1/++ females , the level increases substantially when eif4e activity is reduced . Since these synergistic effects occur even though Sxl levels in the triply heterozygous mutant females are the same as in the control snf1621 Sxlf1/++ females , we conclude that the disruption in Sxl dependent alternative splicing of Sxl-Pm transcripts in this context ( and presumably also in early embryos ) can not be due to a requirement for eif4e in either the export of Sxl mRNAs or in their translation . Instead , eif4e activity must be needed specifically for Sxl dependent alternative splicing of Sxl-Pm pre-mRNAs . Consistent with a more general role in Sxl dependent alternative splicing , there is a substantial increase in msl-2 mRNAs lacking the first intron when eif4e activity is reduced in snf1621 Sxlf1/++ females . In the second experiment we examined the splicing of pre-mRNAs from the endogenous Sxl gene and from a Sxl splicing reporter in females heterozygous for two hypomorphic eif4e alleles . Male spliced mRNAs from the endogenous gene and from the splicing reporter are detected the eif4e/+ females , but not in wild type females . Moreover , the effects on sex-specific alternative splicing seem to be specific for transcripts regulated by Sxl as we didn't observe any male spliced dsx mRNAs in eif4e/+ females . Two models could potentially explain why eif4e is needed for Sxl dependent alternative splicing . In the first , eif4e would be required for the translation of some critical and limiting splicing co-factor . When eif4e activity is reduced , insufficient quantities of this splicing factor would be produced and this , in turn , would compromise the fidelity of Sxl dependent alternative splicing . In the second , the critical splicing co-factor would be eif4e itself . It is not possible to conclusively test whether there is a dose sensitive requirement for eif4e in the synthesis of a limiting splicing co-factor . Besides the fact that the reduction in the level of this co-factor in flies heterozygous for hypomorphic eif4e alleles is likely to be rather small , only a subset of the Sxl co-factors have as yet been identified ( unpublished data ) . For these reasons , the first model must remain a viable , but in our view , unlikely possibility . As for the second model , the involvement of a translation factor like eif4e in alternative splicing is unexpected if not unprecedented . For this to be a viable model , a direct role for eif4e must be consistent with what is known about the dynamics of Sxl pre-mRNA splicing and the functioning of the Sxl protein . The evidence that the second model is plausible is detailed below . Critical to the second model is both the nuclear localization of eIF4E and an association with incompletely spliced Sxl pre-mRNAs . Nuclear eIF4E has been observed in other systems , and we have confirmed this for Drosophila embryos . We also found that eIF4E is bound to Sxl transcripts in which the regulated exon2-exon3-exon4 cassette has not yet been spliced . In contrast , it is not associated with incompletely processed transcripts from the tango gene , which are constitutively spliced . With the caveat that we have only one negative control , it is not surprising that Sxl transcripts might be unusual in this respect . There is growing body of evidence that splicing of constitutively spliced introns is co-transcriptional [78]–[83] . However , recent in vivo imaging experiments have shown that the splicing of the regulated Sxl exon2-exon3-exon4 cassette is delayed until after the Sxl transcript is released from the gene locus in female , but not in male cells [84] . These in vivo imaging studies also show that , like bulk pre-mRNAs , the 1st Sxl intron is spliced co-transcriptionally in both sexes . Consistent with a delay in the splicing of the regulated cassette , we've previously reported that polyadenylated Sxl RNAs containing introns 2 and 3 can be readily detected by RNase protection , whereas other Sxl intron sequences are not observed [19] . The delay in the splicing of the regulated Sxl cassette until after transcription is complete and the RNA polyadenylated could provide a window for exchanging eIF4E for the nuclear cap binding protein . To function as an Sxl co-factor , eIF4E would have to be associated with the pre-mRNA-spliceosomal complex before or at the time of the Sxl dependent regulatory step . There is still a controversy as to exactly which step in the splicing pathway Sxl exerts its regulatory effects on Sxl-Pm pre-mRNAs and two very different scenarios have been suggested . The first is based on an in vitro analysis of Sxl-Pm splicing using a small hybrid substrate consisting of an Adenovirus 5′ exon-intron fused to a short Sxl-Pm sequence spanning the male exon 3′ splice site [85] . These in vitro studies suggest that Sxl acts very late in the splicing pathway after the 1st catalytic step , which is the formation of the lariat intermediate in the intron between exon 2 and the male exon . According to these experiments Sxl blocks the 2nd catalytic step , the joining of the free exon 2 5′ splice site ( or Adeno 5′ splice site ) to the male exon 3′ splice site ( see Figure 1A ) . It is postulated that this forces the splicing machinery to skip the male exon altogether and instead join the free 5′ splice site of exon 2 to the downstream 3′ splice site of exon 4 . Since we have shown that eIF4E binds to Sxl-Pm pre-mRNAs that have not yet undergone the 1st catalytic step ( Figure 6 ) , it would be in place to influence the splicing reaction if this scenario were correct . The second scenario is more demanding in that it proposes that Sxl acts during the initial assembly of the spliceosome . Evidence for Sxl regulation early in the pathway comes from the finding that Sxl and the Sxl co-factor Fl ( 2 ) d show physical and genetic interactions with spliceosomal proteins like U1-70K , Snf , U2AF38 and U2AF50 that are present in the early E and A complexes and are important for selecting the 5′ and 3′ splice sites [45] , [61] , [64]–[71] . In addition to these proteins , Sxl can also be specifically cross-linked in nuclear extracts to the U1 and U2 snRNAs [43] . Formation of the E complex depends upon interactions of the U1 snRNP with the 5′ splice site , and this is thought to be one of the first steps in splicing . The other end of the intron is recognized by U2AF , which recruits the U2 snRNP to the 3′ splice site . After the base pairing of the U2 snRNP with the branch-point to generate the A complex the next step is the addition of the U4/U5/U6 snRNPs to form the B complex . However , Sxl and Fl ( 2 ) d are not found associated with components of the splicing apparatus like U5-40K , U5-116K or SKIP that are specific for complexes B and B* , or the catalytic C complex [70]–[71] , [74]–[75] , [86]–[88] . Nor can Sxl be cross-linked to the U4 , U5 or U6 snRNAs [43] . If Sxl and Fl ( 2 ) d dissociated from the spliceosome before U4/U5/U6 are incorporated into the B complex , then they must influence splice site selection during the formation/functioning of the E and/or A complex . ( Since the transition from the E to the A complex has been shown to coincide with an irreversible commitment to a specific 5′—3′ splice site pairing , Sxl would likely exerts its effects in the E complex when splice site pairing interactions are known to still be dynamic [89] . ) If this is scenario is correct , eIF4E would have to be associated with factors present in the earlier complexes in order to be able to promote Sxl regulation . This is the case . Thus , eIF4E is found in complexes containing the U1 snRNP protein U1-70K , the U1/U2 snRNP protein Snf , and the two U2AF proteins , U2AF38 and U2AF50 . With the exception of the Snf protein bound to the U2 snRNP , all of these eIF4 associated factors are present in the early E or A complexes , but are displaced from the spliceosome together with the U1 and U4 snRNPs when the B complex is rearranged to form the activated B* complex . This would imply that eIF4E is already in place either before or at the time of B complex assembly . Arguing that eIF4E associates with these E/A components prior to the assembly of the B complex is the finding that eIF4E is also in complexes with both Sxl and Fl ( 2 ) d . Thus , even in this more demanding scenario for Sxl dependent splicing , eIF4E would be present at a time when it could directly impact the regulatory activities of Sxl and its co-factor Fl ( 2 ) d . Taken together these observations would be consistent with a Sxl co-factor model . While further studies will be required to explain how eIF4E helps promote female specific processing , an intriguing possibility is suggested by the fact that hastening the nuclear export of msl-2 in females would favor the female splice ( which is no splicing at all ) . Hence , one idea is that eIF4E binding to the pre-mRNA provides a mechanism for preventing the Sxl regulated splice sites from re-entering the splicing pathway , perhaps by constituting a “signal” that blocks the assembly of new E/A complexes . A similar post-transcriptional mechanism could apply to female-specific splicing of the regulated Sxl exon2-exon3-exon4 cassette . The binding of eIF4E ( and PABP ) to incompletely processed Sxl transcripts after transcription has terminated in females would prevent the re-assembly of E/A complexes on the two male exon splice sites , and thus promote the formation of an A complex linking splicing factors assembled on the 5′ splice sites of exons 2 and on the 3′ splice site of exon 4 .
Flies were raised at room temperature on a standard Drosophila media . Crosses were performed at 29°C unless otherwise indicated with 3–7 females and 2–4 males per vial . Crosses were transferred to new vials every 2–3 days . Similar crosses were performed at 25°C , but the effects were significantly weaker . Unless otherwise noted stocks are referenced by Lindsley & Zimm [89] . w; eif4eSO587/11/TM3Sb ( eif4e587 , FBal0129763 ) , w;eif4eEP568/TM3Sb ( eif4e568 , FBal0122994 ) , w;eif4eSO715/TM3Sb ( eif4e715 , FBal0175695 ) , y1w67c23 , w cm Sxlf1 ct/Bincinscy , y w ( FBal0016680 ) , Sxl7BO/Bincinscy ( FBal0016694 ) , y pn SxlM1/Bincinscy ( FBal0016703 ) , y pn SxlM4/Bincinscy ( FBal0016710 ) , y pn SxlM6/Bincinscy ( FBal0103944 ) , cm Sxlf9/Bincinscy ( FBal0016686 ) , y w snf1621 ct/Bincinscy , y w snf1621 Sxlf1 ct/Bincinscy . To identify suppressors of the dominant male lethality conferred by Sx-N , we crossed w Sxl7B0/Bin; hsp83:Sx-NΔ transgene mothers to Deficiency/Balancer fathers and scored for viable , non-Balancer males containing the transgene . The 67A8-A9 region was one of the chromosomal intervals that was found to contain a suppressor . The eif4e gene mapped to this region and was a strong candidate gene for the dominant suppressor . Four independent eif4e alleles suppressed the male lethal effects of hsp83:Sx-NΔ transgene as indicated in the text . All crosses for both screens were conducted in vials with five females and three males of the appropriate genotype . Matings were allowed to occur for three days at 25°C , at which time the parents were transferred to new vials to ensure that larvae were not crowded . Embryos were collected on apple juice plates sprinkled with yeast at 29°C . They were dechorionated in bleach and fixed in 4% paraformaldehyde:heptane for 20–25 minutes . The fix was removed and embryos devitilinized and stored in methanol at −20°C . To stain , embryos were stepped into PBS , incubated for 1 hour in PAT ( PBS with 1% BSA , 1%Triton-X100 ) and blocked for 30 minutes in PBT ( PBS with 5% BSA ) . Embryos were incubated overnight at 4°C with primary antibody at the appropriate concentration in PBT . The next day the embryos were washed with PBS-T ( PBS with 1% Triton-X100 ) then , incubated for 2 hours at room temperature with secondary antibody at the appropriate concentration in PBT . Embryos were washed with PBS-T , then with PBS . For embryos with fluorescently tagged secondary antibodies , the embryos were incubated for 5 minutes with a 1∶1000 dilution of Hoescht , rinsed twice with PBS , then mounted in Aquamount ( Polysciences , Inc . ) . For embryos with HRP conjugated secondary antibodies , embryos were incubated with 400 ul of 0 . 4 mg/ml DAB in PBS , 1 ul of 3% hydrogen peroxide and 0 . 6 ul of 1 M NiCl2 until the embryos appeared fully stained . To prepare for mounting embryos were stepped into 100% ethanol , then incubated overnight in methyl salicylate . The following morning , embryos were mounted in Permount ( Fisher Scientific ) . Primary antibodies used were: anti-Sxl18 monoclonal at 1∶10 , anti-snf 9G3 monoclonal at 1∶10 and anti-eIF4E polyclonal at 1∶500 ( gift from Paul Lasko ) . Secondary antibodies used were: HRP conjugated goat anti-mouse ( Jackson ImmunoResearch ) at 1∶500 , rhodamine conjugated goat anti-rabbit ( Alexa ) at 1∶500 , fluorescence conjugated goat anti-mouse ( Alexa ) at 1∶500 . Embryonic RNA was prepared as described by Bell et al [90] . Adult RNA from 33 flies was prepared using GE Healthcare mini-spin columns . Reverse transcription was performed according to the procedure of Frohman et al . [91] . 1 . 5–3% of the cDNA was used as template . PCR cycles for embryonic cDNAs were 1X 95°C 4 minutes , 30X 95°C 1 minute , 60–65°C 45 seconds , 72°C 30 seconds , 1X 72°C 10 minutes . If re-amplification was needed , only 10 cycles were performed in the first PCR . Up to 40% of the first PCR was used as template for the second PCR . Primers and temperatures were the same for the second reaction as in the first and 10–30 cycles were performed as needed . Number of cycles needed was evaluated by examining 10 ul samples with EtBr . For adult cDNAs PCR cycles were as follows: 2X 95°C 1 minute , 70–72°C 45 seconds , 7°C 1 minute , 2-4X 95°C 1 minute , 68–70°C 45 seconds , 72°C 1 minute , 2-4X 95°C 1 minute , 66–67°C 45 seconds , 72°C 1 minute , 2-4X 95°C 1 minute , 65–66°C 45 seconds , 72°C 1 minute , 10X ( first PCR ) or 5-30X ( second PCR ) 95°C 1 minute , 65–67°C 45 seconds , 72°C 1 minute . 5 ul of the first PCR diluted 1/100 was used as template for the second PCR . For Southern blotting DNA was run on 1–1 . 2% agarose gels , and Southern blotted to Zeta-Probe membrane or nitrocellulose . For Sxl reactions blots were hybridized with randomly primed Sxl 3B1Δ cDNA [39] . For msl-2 mRNAs the membrane was hybridized to randomly primed msl-2 5′UTR PCR product . Primers used are described in Figure 6 and listed in Table 2 . Nuclear extract was prepared by collecting embryos laid by w1 stock overnight ( <24 hours ) . Embryos were washed with distilled water and 0 . 12 M NaCl , 0 . 04% Triton-X 100 , then dechorionated in 100% bleach for 3 minutes . Dechorionated embryos were rinsed with NaCl , Triton , then NaCl , blotted dry and collected . Embryos were homogenized at 4°C in buffer 1 ( 15 mM HEPES-KOH pH 7 . 6 , 10 mM KCl , 5 mM MgCl2 , 0 . 1 mMEDTA , 0 . 5 mM EGTA , 0 . 35 M sucrose , with 1 mM DTT , 1 mMNa2S2O5 , protease inhibitors , benzamidine and 1mMPMSF ) , using 4 ml buffer/ml lightly packed embryos . The homogenate was filtered through three layers of Mira-cloth , then centrifuged at 2000 xg for 10 minutes at 4°C . Supernatant was removed with a pipet . The pellet was re-suspended in 4 ml buffer/ml embryos , and overlaid onto an equal volume of buffer 2 ( same as buffer 1 except 0 . 8 M sucrose ) , then spun 10 minutes at 2000 xg , at 4°C . The supernatant was removed . The pellet was resuspended in TEN ( 10 mM Tris-HCl pH 7 . 8–8 , 1 . 5 mM EDTA , 100 mM NaCl ) , 2 ml TEN/ml embryos , and sonicated . 20 ul–40 ul of 50% antibody linked protein AG beads , 350 ul co-immunoprecipitation buffer ( 20 mM Hepes , pH 7 . 5 , 150 mM NaCl , 250 mM sucrose , 0 . 05% ( w/v ) Tergitol NP-40 , 0 . 5% ( v/v ) Triton-X 100 plus 1 mM DTT , 1 mMNa2S2O5 , protease inhibitors , benzamidine and 1mMPMSF ) and 12 . 5 ul RNAsin were added to a 150 ul aliquot of sonicate . The mixture was rocked at 4°C overnight , then washed 5 times with co-IP buffer . The beads were boiled for 5–10 minutes with 20 ul protein sample buffer , then spun for 5–10 minutes . 5–10 ul of sample was loaded onto a 12% polyacrylimide gel . The proteins were transferred to Immobilon-P or nitrocellulose . Blots were prehybridized in PBS-5% nonfat dry milk and probed with primary antibody overnight at 4°C . Antibodies used include: mouse anti-Sxl 104 and 114 , mouse anti snf 9G3 [41] , rabbit anti-eIF4e antibody at 1∶1000 [92] , rabbit anti-U170K ( gift of Helen Salz; [45] ) at 1/5000 , rabbit anti-U2AF50 ( gift of Don Rio; [93] ) at 1/5000 , rabbit anti-U2F38 1/5000 ( gift of Don Rio; [94] ) or mouse anti-Fl ( 2 ) d9G2 [60] at 1/10 , mouse anti-scute 5A10 [95] . Blots were washed three times for 10 minutes each in PBST and hybridized with horseradish peroxidase-conjugated secondary antibody ( Goat anti-rabbit ( 1∶10 , 000 ) or Goat anti-mouse ( 1/1000−1/10 , 000 ) from Jackson ImmunoResearch ) in PBST-5% milk for two hours at room temperature . Blots were again washed three times for 10 minutes each in PBST and visualized with an enhanced chemiluminescent agent . Nuclear extract was prepared essentially as above except , after the first centrifugation , the pellet was resuspended in 1 ml buffer/ml embryos and sonicated . 20 ul of 50% antibody linked protein A beads were added to a 150 ul aliquot of sonicate . The mixture was allowed to rock 1 hour at room temperature , then washed as above . RNA was isolated using TRIreagent ( Molecular Research Center , Inc . ) then , treated with DNAse 1 . Reverse transcription , PCR and Southern blotting conditions were as described above with primers as indicated in Figure 6 and Table 2 . Southern blotting conditions were as described above using randomly primed Sxl 3B1Δ cDNA [39 ) as the probe . Antibodies used for immunoprecipitation were; anti-scute SA10 , anti-Sxl 104 and 114 mixed 1∶1 , and anti-eIF4E . 2–5 flies of each genotype were collected and frozen at −80°C . 10 ul of 2x Laemmli sample buffer per fly was added to the flies , which were then homogenized with a hand held Dounce homogenizer . Samples were boiled for 5 minutes and spun for three minutes at 14 , 000 rpm . Samples were diluted as needed with 2x Laemmli sample buffer and up to 10 ul of sample were loaded onto sodium dodecyl sulfate ( SDS ) -12% acrylamide gels , run out and transferred to Immobilon-P or nitrocellulose . Blots were incubated for 60 minutes in PBST ( PBS with 1% Triton-X100 ) , with 10% dry milk then , incubated overnight at 4°C with primary antibody at the appropriate concentration in PBST with 10% dry milk . The next day the blots were washed with PBST for at least an hour , then incubated for 2–4 hours at room temperature with secondary antibody at the appropriate concentration in PBST with 10 mg/ml BSA . Blots were washed with PBST then , developed with ECL Plus ( Amersham ) . Primary antibodies used were: a 1∶1 mixture of anti-SXL104 and 114 at 1/10−1/1000 , anti-eIF4E 1739 at 1/1000 , anti-U2AF50 at 1/50 , 000 , and anti-dFMR J11 at 1/1000 . HRP conjugated goat anti-mouse and goat anti-rabbit ( Jackson ImmunoResearch ) secondary anti-bodies were used at 1/2500 or 1/5000 . | Gene expression in eukaryotes is a complex process that occurs in several discrete steps . Some of those steps are separated into different sub-cellular compartments and thus might be expected to occur independently of one another and involve entirely distinct factors . For example pre-mRNA splicing takes place in the nucleus where it is coupled with transcription , while mRNA translation requires export to the cytoplasm and ribosome loading . We describe studies on the fruit fly Drosophila which indicate that a cytoplasmic translation initiation factor , the cap binding protein eIF4E , plays a key role in alternative splicing in the nucleus . When eIF4E activity is compromised , we observe defects in sex-specific splicing of pre-mRNAs that are regulated by the sex determination master switch gene Sex-lethal . Our data argue that eIF4E likely plays a direct role in the regulation of alternative splicing by Sex-lethal . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genetics",
"biology",
"molecular",
"cell",
"biology",
"genetics",
"and",
"genomics"
] | 2011 | The Translation Initiation Factor eIF4E Regulates the Sex-Specific Expression of the Master Switch Gene Sxl in Drosophila melanogaster |
We have established that HCMV acts as a specific ligand engaging and activating cellular integrins on monocytes . As a result , integrin signaling via Src activation leads to the functional activation of paxillin required for efficient viral entry and for the biological changes in monocytes needed for viral dissemination . These biological/molecular changes allow HCMV to use monocytes as “vehicles” for systemic spread and the establishment of lifelong persistence . However , it remains unresolved how HCMV specifically induces this observed monocyte activation . It was previously demonstrated that the HCMV gH/gL/UL128-131 glycoprotein complex facilitates viral entry into biologically relevant cell types . Nevertheless , the mechanism by which the gH/gL/UL128-131 complex promotes this process is unknown . We now show that only HCMV virions possessing the gH/gL/UL128-131 complex are capable of activating integrin/Src/paxillin-signaling in monocytes . In fibroblasts , this signaling is reversed , such that virus lacking the gH/gL/UL128-131 complex is the only virus able to induce the paxillin activation cascade . The presence of the gH/gL/UL128-131 complex also may have an inhibitory effect on integrin-mediated signaling pathway in fibroblasts . Furthermore , we demonstrate that the presence of the gH/gL/UL128-131 complex on the viral envelope , through its activation of the integrin/Src/paxillin pathway , is necessary for efficient HCMV internalization into monocytes and that appropriate actin and dynamin regulation is critical for this entry process . Importantly , productive infection in monocyte-derived macrophages was seen only in cells exposed to HCMV expressing the gH/gL/UL128-131 complex . From our data , the HCMV gH/gL/U128-131 complex emerges as the specific ligand driving the activation of the receptor-mediated signaling required for the regulation of the actin cytoskeleton and , consequently , for efficient and productive internalization of HCMV into monocytes . To our knowledge , our studies demonstrate a possible molecular mechanism for why the gH/gL/UL128-131 complex dictates HCMV tropism and why the complex is lost as clinical isolates are passaged in the laboratory .
Human cytomegalovirus ( HCMV ) is a betaherpesvirus characterized by worldwide prevalence in the human population . Although infection of immunocompetent individuals is usually mild or asymptomatic , increasing evidence shows that HCMV infection is a strong risk factor in the development of several cardiovascular diseases ( CVDs ) [1]–[5] , and that the infection may lead to the development of some cancers [6] , [7] . In immunocompromised individuals , viral infection can lead to significant morbidity and mortality [8] , [9] . HCMV is the leading viral cause of congenital central nervous system damage and a leading opportunistic pathogen in AIDS and transplant patients [8] , [9] . The virus is shed in nearly all body fluids illustrating HCMV's broad cellular tropism and capacity to spread to and infect most organ systems . It is this broad tropism and multiple organ system involvement that lead , in susceptible individuals , to the hallmark of HCMV pathogenesis - multiorgan failure [9]–[15] . It is thought that , for HCMV to cause broad-organ pathogenesis , infected circulating cells in the blood act as viral-carriers allowing for dissemination of the virus to multiple target tissues . In support , HCMV infection is characterized by a cell-associated viremia , in particular a monocyte-associated viremia prior to the onset of viral pathogenesis [9] , [15]–[18] . As a cell type , monocytes are characterized by high motility and the capacity to migrate to all host organ systems making them an ideal cell type for viral dissemination [19]–[21] . We have previously shown that HCMV infection of monocytes leads to a wide range of biological changes that shape the behavior of target monocytes . HCMV-infected monocytes are characterized by the overexpression and secretion of inflammatory cytokines , an enhanced cellular motility , the increased expression of adhesion molecules allowing for tight adhesion of infected monocytes to endothelial cells , an increase in transendothelial migration , and the promotion of cellular differentiation [16]–[18] . Importantly , monocytes are not permissive for HCMV gene expression and replication upon initial infection and have to differentiate into monocyte-derived macrophages to support productive infection [17] , [22]–[24] . This wide range of molecular changes in monocytes during HCMV infection and lack of productive infection in HCMV-infected monocytes sets this cell type apart from other cell types , underlying the unique biological processes hijacked by HCMV during infection of monocytes . Molecular changes in monocytes begin to occur within minutes post infection , suggesting that a receptor/ligand process initiates changes during the early steps of HCMV infection [16] , [17] . In support , UV-irradiated , virus-treated monocytes showed the same changes in transcription factor regulation , cellular signaling and motility as “live” virus [16]–[18] . Additionally , studies demonstrated that HCMV glycoproteins gB and gH also induced rapid activation of Specificity Protein 1 ( Sp1 ) and nuclear factor κ-light-chain-enhancer of activated B cells ( NF-κB ) transcription factors [16] , [25] , indicating that the act of HCMV binding to target cells triggers biological changes . Recently , we demonstrated that HCMV engages the epidermal growth factor receptor ( EGFR ) , and the β1 and β3 integrins on the surface of monocytes to initiate the receptor-mediated signaling pathways found to be critical for efficient HCMV internalization and virus-induced “hyper” cellular motility [26] , [27] . Furthermore , our results showed that HCMV engagement of each individual receptor ( EGFR vs . β-integrins ) initiated specific changes in HCMV-infected monocytes , such that viral binding of each receptor directed distinct events that modulated viral entry and pathogenic cellular motility [22] , [26] , [27] . Specifically for β1 and β3 integrins , HCMV engagement triggers the activation of the Src/paxillin-signaling axis [27] . Thus , HCMV appears to utilize receptor-initiated signal transduction pathways in monocytes to shape the biology of these cells despite the initial lack of viral gene expression and replication . We propose that the ability of HCMV to induce distinct combinations of signal transduction pathways resulting in HCMV-specific functional changes in monocytes is determined by the nature of the viral glycoproteins expressed on the mature viral envelope . HCMV possesses several major envelope glycoprotein complexes with the glycoprotein B ( gB ) and glycoprotein H ( gH ) complexes being the apparent dominant signaling complexes shown to interact with their cognate cellular receptors , EGFR and β-integrins , respectively [28] , [29] . HCMV gB may also interact with integrins on fibroblasts [29] . Nevertheless , at least on fibroblasts , it appears that the gH complex is the dominant glycoprotein complex responsible for binding to integrins - a receptor-ligand engagement implicated in HCMV entry [28] . With our new data showing that appropriate integrin signaling is required for efficient HCMV entry into blood monocytes , we have now focused our studies on understanding the biological consequences of the gH-integrin interaction . HCMV expresses several types of gH complexes: a dimeric gH/gL complex , a trimeric gH/gL/gO complex , which is sufficient for attachment to and infection of fibroblasts [30] , [31] , and a multimeric gH/gL/UL128-131 complex , which is required for infection of dendritic , endothelial , and epithelial cells [31]–[34] , virus transfer to leukocytes [30] and , according to a recent report , the infection of monocytes [35] . The gH/gL/UL128-131 complex is made of five proteins: gH , gL , the UL128 protein ( pUL128 ) , pUL130 and pUL131 . It is only found expressed on clinical strains of HCMV and is not present or is non-functional in laboratory-adapted strains ( i . e . AD169 ) that have been extensively cultured in fibroblasts [30] , [36]–[39] . Moreover , coding sequences of the UL128 , UL130 and UL131A genes are conserved between clinical isolates , suggesting the importance these three proteins play during in vivo infection [40] . Nevertheless , the mechanism by which the gH/gL/UL128-131 complex promotes viral attachment and entry is unknown . It has been argued that the presence or absence of the UL128-131 complex dictates tropism due to the capacity of this complex to bind to different cell types [41] . We postulate that this idea is an oversimplification of the role the UL128-131 complex plays during infection; that is , this region not only dictates binding , but it also dictates the type and/or levels of receptor-mediated signaling in target cells . To investigate the role the gH/gL/UL128-131 complex plays in the ability of HCMV to induce signal transduction pathways in target monocytes , we used an AD169 clone ( BADwt ) produced from a bacterial artificial chromosome ( BAC ) , containing a frameshift mutation in UL131A , and the virus ( BADrUL131 ) containing a repaired and functional gH/gL/UL128-131 complex [31] . Here , we demonstrated that in monocytes only BADrUL131 is able to induce the integrin/Src/paxillin-signaling pathway – a signaling axis we previously showed to be critical for efficient HCMV entry into and enhanced motility of target monocytes [27] . We also show in this current report that this signaling is reversed in fibroblasts , such that BADwt is the only virus able to induce the paxillin activation cascade . In addition , the presence of the gH/gL/UL128-131 complex appears to have an inhibitory effect on integrin-mediated signaling in fibroblasts . Furthermore , our studies reveal that the presence of the gH/gL/UL128-131 complex on the viral envelope , through its activation of the integrin/Src/paxillin pathway , is necessary for efficient HCMV internalization into monocytes and productive infection in monocyte-derived macrophages . Additionally , the entry efficiency of these viruses ( with and without the UL128-131 complex ) was unchanged during infection of wild type or paxillin-deficient fibroblasts , strongly suggesting contrasting mechanisms of entry into monocytes vs . fibroblasts . Our results indicate that the gH/gL/UL128-131 complex promotes viral internalization through the regulation of actin rearrangement and dynamin , suggesting a macropinocytosis-like route of entry into target monocytes [42]–[44] . From our data , the HCMV gH/gL/U128-131 complex emerges as the specific ligand that is necessary for the activation of the receptor-mediated signaling pathways required for the regulation of the actin cytoskeleton and , consequently , for efficient and productive internalization of HCMV into monocytes . Together , our new studies are the first to document a possible molecular mechanism for why the gH/gL/UL128-131 complex dictates HCMV tropism and why there is a selective pressure to lose this complex as clinical isolates are passaged in the laboratory .
Our results demonstrated that the virus functions as a specific ligand that , through the engagement of integrins and activation of downstream signal transduction pathways , is able to modulate monocyte biology . However , the specific ligand on the viral lipid membrane responsible for inducing the integrin/Src-signaling pathway has not been revealed . In an attempt to answer this question , we began to focus on the HCMV gH complexes , as they were shown to be the predominant HCMV glycoprotein complexes reported to interact with integrins on fibroblasts [28] , and because the gH/gL/UL128-131 complex was documented to be required for endothelial/epithelial and monocyte tropism [30] , [31] , [34] , [35] , [45] . We first investigated if there was a correlation between the presence of the gH/gL/UL128-131 complex on the viral envelope and the ability of several HCMV strains to induce the key integrin/Src-signaling pathway . In our studies , we used HCMV strains containing the gH/gL/UL128-131 complex [our low fibroblast-passaged Towne ( Towne/E p . 40 ) , a moderately fibroblast-passaged Towne/E ( Towne/E p . 51 ) and TB40/E [39]] , as well as viral strains lacking this complex [a highly fibroblast-passaged Towne ( Towne/F p . 57 ) , AD169 and TB40/F ( a high passage TB40/E [39] ) ] . Using a monoclonal antibody recognizing pUL130 ( utilized as a surrogate marker for the UL128-131 complex ) , we found in virus lysates that the TB40/E strain possessed the highest level of pUL130 and that the continuous propagation of the Towne/E strain in fibroblasts caused a stepwise reduction in the expression of pUL130 , such that the protein was no longer detectable in our Towne strain by passage 57 ( Towne/F p . 57; Figure 1A , lower panel ) . pUL130 was also not detected in fibroblast-adapted strains ( AD169 and TB40/F ) . These results are in accord with previous reports demonstrating genetic changes in the UL128-131 region in clinical isolates propagated in fibroblasts [36] , [39] . As an internal control , we examined levels of the HCMV tegument protein , pp65 ( pUL83 ) , in viral lysates ( Figure 1A , upper panel ) . Similar amounts of pp65 were detected in each of the viral strains , suggesting similar particle to PFU ratios were used during infection of cells with the different viral strains . Monitoring of infected fibroblasts with the different virus strains supports our suggestion of a similar number of infectious particles per strain . Next , we tested our panel of HCMV strains for their ability to induce integrin-mediated signaling . Src tyrosine kinases are established regulators of cellular signaling mediated by integrins [46]–[49] , however their activity can also be initiated by other cellular receptors [50]–[53] . Similarly to our previous results [27] , here we show that infection of monocytes with Towne/E p . 40 caused an increase in the phosphorylation of Src at Tyr416 ( ∼2 fold increase ) , when compared to mock-infected monocytes ( Figure 1B ) . Infection with TB40/E had a stronger effect on Src activation ( ∼3 fold increase over mock ) ; than that observed following infection with Towne/E p . 40 ( Figure 1B ) , which correlated with a higher amount of pUL130 in the TB40/E viral particle vs . that seen in the Towne/E p . 40 viral particle ( Figure 1A ) . The highly passaged HCMV strains lacking a functional gH/gL/UL128-131 complex , however , were unable to initiate Src phosphorylation ( TB40/F ) or were inhibitory to Src activation ( Towne p . 57 and AD169 ) upon infection of monocytes ( Figure 1B ) . In addition , the initial activation of Src by Towne/E p . 40 and TB40/E resulted in the activation of a downstream signaling cascade , measured by phosphorylation of p70 S6 kinase ( p70 S6K ) at Thr389 ( 4 . 6- and 6 . 2-fold increase , respectively , over those levels seen in mock-infected cells; Figure 1B ) . Because the activation of the integrin/Src-signaling pathway was demonstrated to be critical for efficient HCMV entry into monocytes [27] , we next investigated the ability of HCMV strains expressing different levels of the gH/gL/UL128-131 complex to efficiently enter monocytes . We only observed efficient entry of Towne/E p . 40 and TB40/E into monocytes , as indicated by the presence of internalized viral genomic DNA ( Figure 1C ) . In contrast , the signal from the internalized HCMV genomic DNA was significantly lower in cells infected with Towne/F p . 57 , AD169 and TB40/F ( Figure 1C ) . Equal loading of samples was ensured by examining the level of GAPDH expression ( Figure 1C ) . Together , our results suggest a link between the presence of the functional gH/gL/UL128-131 complex on the viral envelope and the ability of HCMV to induce the integrin/Src-mediated signaling pathway required for efficient viral entry into target monocytes . In addition , because we used viral strains differentiated only by passage length ( which in turn relates to cell tropism ) , our data provides support for the specific role that UL128-131 plays in these specific strains ( TB40/E vs . TB40/F and Towne/E vs . Towne/F ) ; at present we are not aware of other mutations that may be present in these different passaged strains . These data also support our previous observation of the critical role for the integrin/Src-signaling in efficient HCMV internalization into monocytes [27] . The aforementioned results suggest that the gH/gL/UL128-131 complex may physically engage cellular integrins on the surface of monocytes to initiate receptor-mediated signaling . Furthermore , we speculated that the gH/gL/UL128-131 complex not only bound integrins , but that it might engage only select integrins differentiating the biological effect caused by the gH/gL complex without the UL128-131 trimer from that of the gH/gL complex with the UL128-131 trimer . Thus , we next immunoprecipitated from HCMV-infected monocytes β1 or β3 integrins , receptors previously found to interact with the HCMV virion on these cells [27] , as well as on other cell types [28] , [54] , [55] , and investigated if the UL128-131 complex was capable of engaging these integrins . To pull down sufficient amounts of interacting proteins for their visualization using western blot analysis , we utilized the DTSSP crosslinker to stabilize the interactions . We chose this crosslinker as it allows the dissociation of crosslinked complexes by 5% β-mercaptoethanol and a separation of the individual proteins in those complexes using a standard SDS-PAGE analysis [56] . We found that pull down of β1 and β3 integrins from TB40/E- and AD169-infected monocytes resulted in the finding that the gH protein only interacts with β1 integrins ( Figures 1D and 1E ) . As the lack of gH protein in β3 integrin-immunoprecipitate does not necessary mean the lack of interaction between these proteins , the reverse immunoprecipitation was performed . By immunoprecipitating the gH protein from the lysate of TB40/E-infected cells , we confirmed that gH of TB40/E interacts exclusively with β1 integrin ( Figure 1F ) . However , the immunoprecipitation of the gH protein from the lysate of AD169-infected monocytes suggested that this HCMV glycoprotein can interact with both β1 and β3 integrins ( Figure 1G ) . More importantly , we found that pUL130 present in the gH/gL/UL128-131 complex of TB40/E strain engages both β1 and β3 integrins on the surface of monocytes ( Figure 1D ) . As predicted pUL130 was not detected in immunoprecipitates from monocytes infected with AD169 strain ( Figure 1E ) . As a control for a non-specific binding , we performed immunoprecipitation analyses on lysates from infected monocytes using an IgG isotype matched antibody . We did not detect any signal in our immunoprecipitated samples; however the specific antibodies ( to gH and pUL130 ) recognized viral proteins in the input viral lysate ( Figure 1H ) . Additionally , an immunoprecipitation assay performed on mock-infected cells using an antibody recognizing HCMV gH did not result in a pulldown of the β1 and β3 cellular integrins ( Figure 1I ) . In contrast , antibodies recognizing the β1 and β3 integrins did detect the appropriate integrins in the input lysate ( Figure 1I ) . Western blot analysis also determined that there were equal amounts of cellular proteins in cell lysates used in our co-immunoprecipitation experiments ( Figure 1J ) . The data together suggest that different components of the gH/gL/UL128-131 complex interact on the surface of monocytes with the distinct integrins that we previously showed were important for enhanced motility of and efficient entry into target monocytes [27] . This ability of viruses expressing the UL128-131 complex to bind to the β1 integrin ( via gH ) and to both β1 and β3 integrins ( via the UL128-131 complex ) provides new data as to why HCMV requires both integrins for entry into monocytes [27] ( Figure 1K ) . The results also suggest why only a single β-integrin is likely required for HCMV entry into fibroblasts [28] , [55] . Our findings may provide an explanation of why β1 integrins are key regulators of monocyte function as described by Yurochko et al . [57] . Based on our new results , we speculate that the interaction of the UL128-131 with β1 and β3 integrins may allow for a unique , synchronous engagement and activation of both β1 and β3 integrins by HCMV , which through the creation of the appropriate type and level of integrin-mediated signaling allows for efficient viral entry and the early functional changes in infected monocytes to occur . Long-term passaging of HCMV in fibroblasts alters the virus's genetic composition and those changes are not only limited to the UL128-131 region of the genome [34] , [37] , [58]–[60] . Consequently , in order to determine if the gH/gL/UL128-131 complex is directly responsible for the initiation of integrin signaling , leading to efficient HCMV internalization into monocytes , we decided to utilize two well-characterized bacterial artificial chromosome ( BAC ) -based , AD169-derived clones: BADwt , which contains a frame shift mutation in UL131A , and BADrUL131 , which possesses a repaired UL131 locus and thus has a functional gH/gL/UL128-131 complex [31] . We confirmed that pUL130 was only detected in lysates of BADrUL131 virus , and not in lysates from BADwt ( Figure S1A ) . As a control , we used the AD169 strain that , as showed in Figure 1A , does not express the gH/gL/UL129-131 complex and the TB40/E strain that possesses the pentameric complex on its envelope ( Figure S1A ) . We next investigated the ability of these viruses to induce the integrin/Src-signaling pathway . By using western blot analyses , we determined that BADrUL131 was able to increase the level of phosphorylated Src above the levels seen in mock- and BADwt-infected monocytes ( Figure 2A ) . Figure 2A shows a representative western blot experiment with a densitometry analysis , demonstrating 1 . 6-fold increase of Src activation in BADrUl131-infected monocytes compared to that seen in mock-infected cells . A cumulative densitometry analysis of Src activation in infected monocytes that incorporates results from three repeats of the experiment showed significant changes in the level of phosphorylated Src in BADrUL131-infected cells , when compared to mock- and BADwt-infected cells ( Figure S1B ) . Furthermore , the initial activation of Src in BADrUL131-infected monocytes translated into the activation of downstream signaling , resulting in increased levels of phosphorylated paxillin ( 2 . 4-fold increase ) , Erk ( 2-fold increase ) and SAPK/JNK ( 3 . 6-fold increase ) , compared to those seen in mock-infected cells ( Figure 2A ) . We also noticed that there was a lower level of phosphorylated forms of Erk and SAPK/JNK in BADwt-infected cells , compared to mock-infected monocytes ( Figure 2A ) , suggesting that BADwt infection may have a slight inhibitory effect on the integrin/Src-signaling pathway in target monocytes . We did not observe any significant differences in levels of total Src , paxillin , or Erk . These results not only substantiate the importance of the gH/gL/UL128-131 complex in the activation of integrin-mediated signaling and validate our previous studies demonstrating the ability of HCMV to engage cellular integrins and to induce the integrin/Src/paxillin signaling pathway in infected monocytes [27] , but they further provide insight into how HCMV stimulates integrin receptors on target monocytes . The data presented so far strongly support a clear correlation between the presence of a functional HCMV gH/gL/UL128-131 complex on the viral envelope and the ability of the virus to trigger integrin/Src-signaling in target monocytes . Therefore , we next examined the ability of BADwt and BADrUL131 to enter target monocytes using HCMV entry assay [26] , [27] . Based on the level of internalized vDNA in infected monocytes , we found that BADrUL131 was more efficiently ( ∼3-fold ) internalized into monocytes , when compared to BADwt ( Figure 2B ) . The differences in entry seen between BADwt and BADrUL131 were not due to the ability of these viruses to differentially bind monocytes; we did not observe any significant changes in the binding properties of BADwt and BADrUL131 to monocytes ( Figure S1C ) . We also analyzed the ability of the endotheliotropic TB40/E vs . the non-endotheliotropic TB40/F strains to enter monocytes . TB40/F lost its ability to infect endothelial cells , likely due to a frameshift mutation in UL128 gene region [37] , [39] , [59] , as a result of its prolong passaging in fibroblasts . Our results showed that TB40/E was more efficiently ( ∼3 fold increase ) internalized into monocytes than TB40/F . The efficiency of TB40/E internalization into monocytes was similar to that observed following infection with BADrUL131 ( Figure 2B ) . As a control , infected cells were also kept at 4°C without a temperature shift to 37°C . Using this assay , we determined that basal levels of HCMV genomic DNA in cells kept at 4°C was comparable to the levels of BADwt and TB40/F internalization into monocytes at 37°C ( Figure S2A ) . We speculate that even though we saw very low level of viral internalization at 4°C , this entry does not result in a productive infection . Additionally , the proteinase K-treatment might not have removed all non-internalized viral particles from the surface of infected cells , which would also have an effect on the levels of viral genomic DNA in cells maintained at 4°C . To investigate the ability of BADwt vs . BADrUL131 to establish a productive infection , we performed fluorescence in situ hybridization ( FISH ) analysis on HCMV-infected monocytes to monitor the localization/presence of vDNA at 5 dpi . The signal from the fluorescence probe recognizing HCMV vDNA was only found in monocytes infected with BADrUL131 , not in monocytes infected with BADwt ( Figure 2C; see the inset pictures for a magnified view of the representative cells ) . A majority of the cells analyzed showed evidence of vDNA staining at 5 dpi ( ∼80% ) . As a control , we verified that BADwt and BADrUL131-infected fibroblasts were both positive for vDNA using this probe ( DNS ) . As FISH only allows for an examination of a small number of cells , we also wanted to examine the infected cell population as a whole , thus we additionally conducted semi-quantitative PCR and RT-PCR analyses looking at HCMV genomic DNA and HCMV IE mRNA expression at 5 dpi and 3 weeks pi , respectively . The results obtained from these experiments support and extend the FISH data; we saw amplification of the HCMV UL123 genomic sequence only from DNA isolated from BADrUL131-infected cells at 5 dpi ( Figure 2D ) . Similarly , HCMV IE mRNA was only found expressed at 3 weeks pi in monocytes/macrophages infected with BADrUL131 ( Figure 2E ) , suggesting that only monocyte-derived macrophages initially infected as monocytes with BADrUL131 were able to express vRNA and , thus were the only cells productively infected . Taken together , our data indicate that the presence of the gH/gL/UL128-131 complex on the HCMV envelope is important for the activation of virus-induced , integrin/Src-mediated signaling pathway in target monocytes and for the efficient viral internalization into these cells that ultimately results in productive viral infection . We have documented that the integrin/Src/paxillin signaling axis must be functionally activated for HCMV to enter blood monocytes [27] , and we now demonstrate that the gH/gL/UL128-131 complex is the key trigger for this activation of integrin-mediated signaling and for efficient internalization ( Figures 1 and 2 ) . However , because the signaling networks in HCMV-infected monocytes are complex in their nature and involve crosstalk between different receptors [18] , [22] , [25]–[27] , [61]–[63] , we wanted to clarify if the Src-mediated signaling pathway interacted molecularly with the EGFR-mediated pathway . Both pathways were shown to be important for efficient HCMV internalization into monocytes; however , both pathways were also found to have a distinct role in regulating the biology of HCMV-infected monocytes [26] , [27] . Our new results showed that pretreatment of monocytes with PP2 ( specific Src tyrosine kinase inhibitor ) and/or AG1478 ( specific EGFR tyrosine kinase inhibitor ) did not affect the efficiency of BADwt internalization ( Figure 3A ) . However , when monocytes were pretreated with PP2 or AG1478 prior to BADrUL131 infection , viral internalization was inhibited by approximately 56% and 25% , respectively , compared to DMSO-pretreated , BADrUL131-infected cells ( Figure 3A ) . The addition of both PP2 and AG1478 prior to BADrUL131 infection blocked viral internalization by more than 70% compared to DMSO-pretreated , BADrUL131-infected monocytes , however , the cumulative effect of both drugs was not significantly different from the effect of PP2 alone ( Figure 3A ) . To test for the importance of integrin engagement by the gH/gL/UL128-131 complex in the efficient HCMV entry into monocytes , we pretreated cells with function blocking antibodies to β1 or β3 integrins prior to infection with BADwt and BADrUL131 . As shown by the level of internalized vDNA , we determined that this pretreatment did not have any effect on BADwt entry into target monocytes; however the blocking of β1 or β3 integrins inhibited the ability of BADrUL131 to enter these cells by ∼75% or ∼40% , respectively , as determined by densitometry analysis ( Figure 3B ) . The effect of function blocking antibodies on BADrUL131 internalization correlated with their inhibitory impact on the activation of the integrin/Src/paxillin signaling pathway in monocytes infected with HCMV expressing the pentameric complex ( Figure S2B and [27] ) . We did not detect any effect of these blocking antibodies on the integrin/Src/paxillin signaling axis in cells infected with BADwt ( Figure S2C ) . Together , these data provide additional support for the idea that the functional activation of integrin/Src/paxillin signaling pathway , triggered by the HCMV gH/gL/UL128-131 complex , has a causative effect on efficient HCMV entry into target monocytes . It also suggests that even though BADrUL131 predominantly utilizes the integrin-mediated signaling for its internalization into monocytes , the EGFR-mediated signaling also plays at least a supporting role in this process . We reported earlier that HCMV infection leads to increased expression of paxillin in target monocytes via integrin/Src-signaling [27] . We also demonstrated that by knocking down paxillin expression , we were able to significantly decrease HCMV entry into monocytes [27] . To assess the importance of paxillin regulation in BADwt vs . BADrUL131 internalization into monocytes , we used siRNA to knock down paxillin expression as previously described [27] and as before we accomplished a paxillin knockdown efficiency of ∼80–90% ( Figure S2D ) . Using our entry assay , we found that the lack of paxillin expression did not influence BADwt entry into monocytes , while in contrast , BADrUL131 internalization into paxillin-deficient monocytes was inhibited ∼45% ( Figure 3C ) . To attempt to answer if the lack of paxillin activation might be responsible for the inability of BADwt to efficiently enter monocytes , a rescue experiment using α-thrombin , documented to increase phosphorylation of Src and paxillin [64] , [65] , was also performed . We found that treatment of monocytes with α-thrombin was able to increase the levels of the phosphorylated forms of paxillin and Src with the peak of this activation at 15 min . post the α-thrombin treatment ( Figure S2E and data not shown ) . Because our results demonstrated that the kinetics of HCMV- and α-thrombin-mediated activation were similar ( Figure S2E and [27] ) , we wanted to mimic the pace of paxillin phosphorylation triggered by the gH/gL/UL128-131 complex by exposing BADwt-infected monocytes to α-thrombin just before shifting temperature from 4°C to 37°C in our entry assay . We found that the efficiency of BADwt internalization into monocytes with a normal paxillin expression was significantly ( 2-fold ) enhanced by α-thrombin treatment ( Figure 3D ) . Because α-thrombin can stimulate other signal transduction pathways [66]–[69] , we wanted to ensure that α-thrombin-mediated effect on BADwt internalization was paxillin-dependent . Thus , we examined the effect of α-thrombin on BADwt internalization into monocytes deficient of paxillin expression . Monocytes were transduced with scrambled siRNA or siRNA specific for paxillin mRNA , as previously described [27] . The positive effect of α-thrombin on the entry of BADwt into control siRNA-transduced monocytes was diminished in monocytes lacking paxillin expression ( Figure 3D ) , thus suggesting that the inducing effect of α-thrombin on BADwt internalization was paxillin-dependent . In summary , these data support the critical role for the activated integrin/Src/paxillin-signaling pathway , induced through the interaction of integrins with the HCMV gH/gL/UL128-131 complex , in efficient HCMV internalization into monocytes . Fibroblasts are a very well studied in vitro model system of HCMV infection . Although we have previously shown that HCMV receptor/ligand engagement activates fibroblasts [25] , we have also documented that there are key differences between the nature and duration of the signaling during HCMV infection of monocytes vs . fibroblasts . For example , HCMV infection of monocytes results in the induction of cellular differentiation , long-term cellular survival , and PI ( 3 ) K-independent HCMV entry into monocytes [17] , [18] , [26] , [61] , [70] , which is not observed in fibroblasts . Additionally , HCMV clinical isolates were reported to infect fibroblasts less efficiently than highly passaged laboratory adapted strains with the opposite being true for the infection of endothelial cells [71] , [72] , suggesting to us that the presence of the gH/gL/UL128-131 complex differentially affects fibroblasts vs . monocytes . Thus , we next investigated if there were differences in receptor-mediated signaling in BADwt- vs . BADrUL131-infected fibroblasts . Using western blot analyses , we demonstrated that in fibroblasts , BADwt triggered an increase in the level of phosphorylated Src ( 1 . 9-fold increase ) as compared to mock-infected fibroblasts ( Figure 4A ) . This initial activation of Src induced downstream signaling , resulting in increased levels of activated paxillin ( 1 . 8-fold increase ) and Erk ( 1 . 6-fold increase ) as compared to mock-infected cells ( Figure 4A ) . In contrast , the levels of activated Src and paxillin were not changed in BADrUL131-infected fibroblasts , compared to mock-infected cells ( Figure 4A ) . Interestingly , the level of phosphorylated Erk in fibroblasts infected with BADrUL131 was lower than that seen in mock-infected cells ( Figure 4A ) , suggesting that BADrUL131 may even have an inhibitory influence on molecules downstream of integrin/Src/paxillin-signaling in fibroblasts . As showed by our entry assay , the differences in stimulating the receptor-mediated signaling in fibroblasts by BADwt vs . BADrUL131 did not translate into differences in the ability of these two viruses to be internalized ( Figure 4B ) , supporting previously published data [34] . Because of the importance of paxillin regulation for HCMV entry into monocytes , we assessed the role paxillin plays in BADwt vs . BADrUL131 internalization into fibroblasts . siRNA technology was used to knock down paxillin expression and , similarly to monocytes ( Figure S2D and [27] ) , we were able to decrease paxillin expression in siRNA-transfected fibroblasts by 80 to 90% ( Figure S2F ) . We found that the expression of paxillin did not influence either BADwt or BADrUL131 entry into fibroblasts ( Figure 4C ) , suggesting that although BADwt triggers paxillin phosphorylation during infection of fibroblasts ( Figure 4A ) , paxillin regulation is not required for efficient viral internalization into this cell type . To assure that the differences seen in the ability of BADwt and BADrUL131 to stimulate the integrin/Src-signaling in fibroblasts , as well as the similar abilities of these viruses to enter fibroblasts , was not attributed to differences in binding of BADwt vs . BADrUL131 to fibroblasts , we performed a binding assay . We did not observe significant changes in the binding of the BADwt and BADrUL131 to fibroblasts ( Figure S1D ) . Results presented earlier in this manuscript demonstrated that the functional regulation of paxillin in target monocytes is central for efficient internalization of HCMV virions possessing the gH/gL/UL128-131 complex . We have also recently documented that paxillin expression at the level of mRNA and protein is elevated upon HCMV infection of monocytes , and its regulation is critical for HCMV-mediated pathological motility of target monocytes [27] . Paxillin is a scaffolding and signal transduction protein that plays an important role in regulating the interaction between multiple proteins involved in cell motility and adhesion [73] . Increased paxillin ( Tyr118 ) phosphorylation has been shown to enhance actin polymerization and cytoskeleton rearrangement [74] , [75] . Thus , we hypothesized that paxillin is a central regulator of HCMV-mediated changes in infected monocytes and through its role in actin remodeling governs a “hyper” motility of and HCMV entry into monocytes . The role of paxillin-mediated actin rearrangement in cellular motility has been extensively studied [76]–[78] , while the importance of paxillin in rapid actin filament rearrangement during viral entry remains unknown . The modulation of the actin cytoskeleton is utilized by several viruses to enhance their infectivity ( i . e . Epstein Barr Virus , Vaccinia virus [79] ) and entry ( i . e . Human Immunodeficiency Virus , Adenovirus [80] , [81] ) , suggesting that HCMV may also require actin rearrangement as a part of its entry process . Therefore , we asked if HCMV modulates actin rearrangement in infected monocytes . G-actin is a monomeric form of actin , which can polymerize in an ATP-dependent manner to create conformationally changed F-actin-based filaments [82] . Based on western blot analysis , we found that at 15 min . pi , the ratio of F-actin to total actin fell by approximately 50% , suggesting a decrease in polymerized F-actin in HCMV-infected monocytes when compared to mock-infected cells ( Figure S3A ) . Our results also showed that there was a higher level of F-actin at 60 min . pi in HCMV-infected monocytes when compared to mock-infected cells ( ∼50% higher; Figure S3A ) , which corresponds to the increased motility of these cells at later times post infection [17] , [18] , [26] , [27] , [61] . As a control for our experiments , monocytes were also treated with jasplakinolide ( an inducer of actin polymerization ) and with latrunculin A ( an inhibitor of actin polymerization ) [83] , [84] . We found that at 15 min . post treatment , jasplakinolide increased the ratio of F-actin to total actin by approximately 40% , and latrunculin A decreased this ratio by 35% when compared to untreated monocytes ( Figure S3B ) . The presented data suggest that HCMV induces actin rearrangement very early post infection , correlating with the early time frame of viral internalization ( [85] and our unpublished results ) . Because we showed that HCMV was characterized by a lower efficiency of internalization in paxillin-deficient monocytes [27] , we next wanted to investigate if the deficiency in paxillin expression affected actin rearrangement . By comparing levels of F-actin to total actin in siRNA-treated monocytes , we demonstrated that cells transfected with paxillin siRNA for 48 h were characterized by a 40–50% decrease in the ratio of F-actin to total actin when compared to this ratio in mock- and control siRNA-treated monocytes ( Figure S3C ) . These results support the idea that paxillin , via its regulation of the actin cytoskeleton in monocytes , could be involved in viral entry and that its loss leads to the suppression of normal actin turnover and a decrease in actin polymerization [86]–[89] . To examine the direct role of actin rearrangement in the HCMV internalization process into monocytes , cells were pretreated with jasplakinolide or latrunculin A 1 h prior to infection with BADwt , BADrUL131 or TB40/E . From an examination of internalized vDNA , we found that latrunculin A significantly inhibited BADrUL131 ( ∼60% decrease ) and TB40/E ( ∼53% decrease ) entry into monocytes compared to DMSO-treated cells ( Figure 5A ) . When monocytes were pretreated with jasplakinolide , BADrUL131 internalization into monocytes was decreased by ∼70% and TB40/E entry was decreased by more than 80% ( Figure 5A ) . Both pharmacological compounds also had an impact on the internalization process of BADwt; however , the effects were not as substantial as that seen following infection with BADrUL131 and TB40/E . Latrunculin A decreased BADwt internalization efficiency by ∼45% , and jasplakinolide was able to decrease this process by 34%; however , this result was not statistically significant ( Figure 5A ) . The results presented so far suggest that paxillin , through its regulation of the actin cytoskeleton , affects HCMV internalization into monocytes . To address paxillin's possible direct influence of HCMV entry into monocytes via actin rearrangement , we examined if we could rescue the low entry efficiency of BADrUL131 into paxillin-deficient monocytes ( [27] and Figure 3C ) by inducing actin polymerization in these cells . Shelhass et al . [90] has demonstrated that human papillomavirus ( HPV ) localizes to long , tubular invaginations of plasma membrane in cells treated with cytochalasin D ( an inhibitor of actin cytoskeleton dynamics ) . These structures were shown to hold several viral particles and are thought to emerge due to the inability of endocytic vesicles to pinch off . We hypothesized that HCMV infection of paxillin-deficient monocytes may mimic aspects of HPV entry and thus we could induce the completion of the HCMV entry process into paxillin-deficient monocytes by inducing actin polymerization . Monocytes were treated with siRNA recognizing paxillin mRNA or control siRNA for 48 h and then infected at 4°C with BADrUL131 . After an hour-long-incubation at 37°C; cells were treated with an inducer of actin polymerization ( jasplakinolide ) for an additional hour . Again , we found that BADrUL131 was not able to efficiently enter paxillin-deficient monocytes ( Figure 5B ) . Importantly , jasplakinolide increased ( ∼4× ) the internalization of BADrUL131 in paxillin-deficient monocytes compared to that seen in monocytes treated only with paxillin siRNA ( Figure 5B ) . The level of BADrUL131 internalization in jasplakinolide-treated cells was higher than that seen in control siRNA-treated cells ( Figure 5B ) , which may suggest that there is a concentration of HCMV particles inside plasma membrane structures when paxillin expression is limited , that in turn fully enter when actin polymerization is induced . Noteworthy , we saw an inhibitory effect of jasplakinolide-treatment on BADrUL131 internalization into control siRNA-transfected monocytes ( Figure 5B ) , which may suggest that not all viral particles entered monocytes within the first hour at 37°C . Together , these data strongly support the notion that HCMV utilizes a paxillin-regulated actin rearrangement as a mechanism for its efficient entry into monocytes . It also indicates that there is an interrelationship between the presence of the gH/gL/UL128-131 complex on the HCMV envelope and the dependence of HCMV entry on actin regulation . Moreover , these studies provide new clues about the molecular mechanisms for how HCMV mediates enhanced motility of and efficient viral entry into target monocytes . The involvement of actin rearrangement in the internalization process of HCMV into monocytes indicates that the virus might use one of the endocytic pathways to enter these cells [91] . To address this possibility , we tested a panel of pharmacological inhibitors targeting different modes of endocytosis . Monocytes were pretreated with nystatin ( disrupts caveolar structure and function [92] ) , genistein ( disrupts caveolae-mediated endocytosis [93] ) , Rac1 inhibitor ( Rho GTPase inhibitor [94] ) and dynasore ( dynamin inhibitor [95] ) prior to BADrUL131 infection . We found that among these inhibitors , only dynasore significantly ( ∼35% ) blocked HCMV internalization into monocytes ( Figure 5C and Figure S3D ) . Additionally , we analyzed the ability of BADrUL131 to enter dynamin-deficient monocytes . Cells were transduced with siRNA recognizing dynamin mRNA or a control siRNA for 48 h . Even though , we only obtained ∼30% dynamin knockdown in monocytes , as determined by examining the level of dynamin mRNA ( Figure S3E ) , BADrUL131 entry into these monocytes , as showed by levels of internalized vDNA using a semiquantitative PCR analysis , was inhibited by 60% ( Figure 5D ) , which provided an even stronger support for the importance of dynamin regulation in efficient HCMV entry into monocytes . We found that dynasore did not have any effect on BADwt entry into monocytes ( Figure 5C ) , suggesting again that only virus expressing the gH/gL/UL128-131 complex is able to utilize a dynamin-dependent entry mode into monocytes , which in turn leads to productive infection of HCMV in monocyte-derived macrophages ( Figure 2C , 2D , and 2E ) . Additionally , we demonstrated that dynasore , jasplakinolide and latrunculin A were not able to inhibit HCMV entry into fibroblasts , as the efficiency of BADwt and BADrUL131 internalization was not altered by drug-pretreatment in these cells ( Figure 5E and 5F ) . Interestingly , we also observed that the role of integrin/Src/paxillin-signaling , the actin cytoskeleton and dynamin regulation in efficient HCMV entry into monocytes matched their role in HCMV internalization into epithelial cells , as seen by levels of internalized BADrUL131 and TB40/E DNA in epithelial cells pretreated with PP2 , AG1478 , dynasore , jasplakinolide , and latrunculin A ( Figure S3F ) . Taken together , our data indicate that HCMV , expressing the gH/gL/UL128-131 complex , uses a dynamin- and actin-dependent endocytic-like route to enter into target monocytes and that the internalization process of HCMV is strikingly different to that utilized by the virus in fibroblasts , however it closely resembles a type of HCMV entry utilized in epithelial cells .
HCMV has been demonstrated to be an activating stimulus in fibroblasts and endothelial cells [96] , [97] and , as our laboratory has shown , in monocytes [16]–[18] , [26] , [27] , [61] , [62] . We previously showed that HCMV infection of monocytes leads to a wide range of biological changes that shape the behavior of target monocytes and set them apart from model systems , highlighting HCMV's unique influence on the biology of infected monocytes following primary infection . The biological changes seen in HCMV-infected monocytes allow the virus to use the natural sentinel role of circulating monocytes to exit the blood stream and translocate to multiple host organ tissues , where monocytes , which are non-permissive for viral replication , undergo a distinct HCMV-driven differentiation into macrophages that support viral replication and production of progeny virions [16]–[18] , [63] , [70] . Our previous data strongly suggested that HCMV binding to target cells triggers specific biological changes via receptor/ligand-initiated processes [16]–[18] , [25] . HCMV has been shown to bind to several different cellular receptors , with utilization of these receptors by the virus being apparently cell type specific [22] , [26]–[29] , [45] , [55] , [98] , [99] . We have recently reported that EGFR and integrins were engaged by HCMV on the surface of monocytes resulting in the receptor-mediated signaling pathways found to be critical for efficient HCMV internalization into monocytes , virus-induced “hyper” motility of and the prolonged survival of these cells [22] , [26] , [27] . Our data indicated that HCMV overcomes its restricted replication in monocytes by inducing EGFR- and integrin-mediated signaling , allowing the virus to use monocytes as virus-carriers for its systemic spread . However , the specific viral trigger responsible for monocyte activation has not been identified . We proposed that the ability of HCMV to induce distinct signal transduction pathways resulting in functional changes in monocytes is determined by the nature of the viral glycoproteins expressed on the mature viral envelope and , specifically , by the presence of the HCMV gH/gL/UL128-131 pentameric complex . The complex is required for endothelial/epithelial cell infection , virus transfer to leukocytes and infection of monocytes [30] , [31] , [34] , [35] , [45] . Nevertheless , the mechanism allowing the gH/gL/UL128-131 complex to promote the viral entry process has not been revealed . It has been proposed that the UL128-131 complex dictates tropism simply due to the ability of gH/gL/UL128-131 complex to bind to different cell types [41] . We postulate that this region not only dictates binding , but it also governs the type and/or levels of receptor-mediated signaling . To investigate the role of the gH/gL/UL128-131 pentamer in the ability of HCMV to induce signal transduction pathways in target monocytes , we first used several HCMV strains that differed in the presence of the gH/gL/UL128-131 complex . We found that only viruses expressing the gH/gL/UL128-131 complex were able to activate integrin/Src-signaling and be efficiently internalized into target monocytes . Therefore , our data established a functional connection between the presence of the gH/gL/UL128-131 complex on the mature HCMV envelope , the receptor-mediated signal transduction and the ability of HCMV to be efficiently internalized into target monocytes . These results supported our previous report documenting the critical role of integrin/Src-signaling for efficient HCMV internalization into monocytes [27] . The current report argues for the first time that the gH/gL/UL128-131 complex plays a role as a specific ligand for activating monocytes and , consequently , allowing for viral entry into these cells . While looking at the molecular mechanism of the interaction between the gH/gL/UL128-131 complex and cellular integrins , we determined that the gH/gL/UL128-131 complex enables HCMV to simultaneously interact with two distinct heterodimeric integrins on the surface of monocytes . Specifically , gH only interacts with β1 integrins , which have been demonstrated to be central regulators of cellular immediate-early gene induction in monocytes [57] , and the UL128-131 trimer engages both β1 and β3 integrins . gH from the gH/gL/ ( gO ) complex was able to engage both β1 and β3 integrins on monocytes , however this engagement did not result in the induction of integrin-mediated signaling in infected monocytes . We speculate that the gH/gL/UL128-131 complex may provide a more effective interaction with cell surface receptors and therefore create a higher affinity binding , which in turn induces a higher level of cellular activation . Additionally , the UL128-131 complex may also interact with different regions of the cellular integrins , allowing for augmented integrin-mediated signaling seen in infected monocytes . We argue that HCMV uses the pentameric gH complex to execute this intrinsic β1 integrin-mediated stimulation of monocytes for its own advantage , while the UL128-131 complex alone allows for the additional required stimulus for integrin-mediated signaling in monocytes . Therefore , the whole gH/gL/UL128-131 complex allows for a synchronous engagement of β1 and β3 integrins by HCMV and the creation of an appropriate type and level of integrin-mediated signaling in target monocytes . The aforementioned results hint that a close spatial cooperation between different integrins and possibly other surface receptors ( i . e . EGFR [26] ) on monocytes is critical for specific HCMV-mediated changes in monocytes . Wang et al . demonstrated that lipid rafts were important for the HCMV-mediated interaction between integrins and EGFR and for orchestrating the receptor-initiated signalosome [28] . Our data also suggest that lipid rafts are important for receptor-mediated signaling in HCMV-infected monocytes ( unpublished data ) . In our studies , we used two well-characterized viruses expressed from a BAC system: BADwt ( an AD169 clone ) , containing a frameshift mutation in UL131A and thus is without a functional pentameric complex , and BADrUL131 , that possesses a repaired and functional gH/gL/UL128-131 complex [31] , to better understand the role of the pentameric complex in viral-mediated signaling in and entry into target monocytes . In monocytes , we found that only BADrUL131 was able to induce the integrin/Src activation and their downstream signaling partners , strongly suggesting that the gH/gL/UL128-131 complex is a specific ligand responsible for activating the integrin-signaling pathway . Moreover , we found that the ability of BADrUL131 to initiate monocyte activation translates into a more efficient process of viral internalization into these cells compared to that seen for BADwt . Like BADrUL131 , we also found that TB40/E strain was more efficiently internalized into monocytes , as compared to TB40/F . We are aware , however , that Ryckman et al . [71] reported a difference in the ability of HCMV lacking the UL128-to-UL150 genes ( HCMV TRΔ4 strain ) to be absorbed on the surface of epithelial cells and fibroblasts compared to wild type TR strain . These authors did not find a decreased capability of AD169 to bind these cells [71] , suggesting that possible proteins associated with the changed absorption rate of HCMV on monocytes , fibroblasts and epithelial cells are not encoded from the UL128-131 gene cluster . Importantly , by monitoring the cellular localization and expression of vDNA and vRNA , our studies also determined that only BADrUL131 , but not BADwt , productively infects monocytes . Taken together , our data strongly support the thesis that the presence of the gH/gL/UL128-131 complex on the HCMV envelope is critical for the activation of the virus-induced , integrin/Src-mediated signaling pathway in target monocytes and for efficient viral entry into these cells , as well as for productive infection . Additionally , our report supports and provides a mechanistic explanation to results presented in a recent report showing that the UL128 component of the gH/gL/UL128-131 complex was important in HCMV infectivity of monocytes [35] . To determine the importance of receptor-mediated signaling mediated by the gH/gL/UL128-131 complex in the increased internalization of BADrUL131 compared to BADwt , we tested the role the integrin/Src- and EGFR–mediated signaling pathways played in virus entry . We did not observe any influence of these signaling pathways on BADwt entry into monocytes . In contrast , BADrUL131 internalization was significantly inhibited in monocytes that had both the integrin and/or EGFR signal transduction pathways impeded . Additionally , our studies demonstrated that , unlike for BADwt , β1 and β3 integrins are engaged by BADrUL131 to stimulate the integrin/Src/paxillin-signaling and are critical for efficient BADrUL131 entry into monocytes , supporting our previous report [27] . Our experiments also demonstrated that paxillin regulation plays an important role in BADrUL131 internalization into monocytes , as only the internalization of BADrUL131 , and not BADwt , was hindered in paxillin-deficient monocytes . Interestingly , we were able to artificially increase the entry efficiency of BADwt , characterized by a low infectivity rate in monocytes , by inducing α-thrombin-mediated paxillin phosphorylation in these cells . This α-thrombin-mediated effect was abrogated in monocytes lacking paxillin expression , indicating that regulation based on the integrin/Src/paxillin signaling axis is vital for efficient HCMV entry into monocytes . Taken together , these results strengthen our previously stated hypothesis of the important role the viral pentameric complex plays in the induction of the receptor-mediated signalosome required for efficient HCMV entry into target monocytes . Long-term passaging of HCMV clinical isolates introduces changes into the viral genome [30] , [39] , [100]–[104] . However , it has not been evident why there is a strong selection against the UL128-131 . We believe that our present studies shed light on this matter , as the effect of BADwt and BADrUL131 on the activation of receptor-mediated signaling pathways in fibroblasts was opposite to that seen in infected monocytes . In fibroblasts , it was BADwt , not BADrUL131 , which triggered Src phosphorylation and activation of the downstream signaling cascade . Interestingly , we noticed an inhibitory effect of BADrUL131 on Erk activation , which previously was found to be required for efficient HCMV infection [105]–[107] . Additionally , our data indicate that paxillin regulation does not play an important role in HCMV entry into fibroblasts and the different potential of BADwt vs . BADrUL131 to induce the integrin/Src/paxillin-signaling axis in these cells did not influence their ability for internalization , suggesting that the expression of the UL128-131 complex does not provide any advantage to the virus . Thus , this locus might be more preferentially mutated when HCMV clinical isolates are passaged in fibroblasts . We noticed that HCMV induces relatively low activation of signaling molecules in both monocytes and fibroblasts ( ∼2× higher than in mock-infected cells ) , when analyzed by western blot analysis . We argue that this effect is significant , as we estimate , based on the biology of the virus-cell interactions and our experimental design , that only between 10–20% of cell surface integrins are engaged by HCMV [108] , which in turn translates to only ∼10–20% of the total number of Src molecules being activated in an infected cell . Additionally , we observed that higher number of viral particles used for infecting monocytes translated into greater activation of the integrin-mediated signaling in these cells . Taken together , our and others' results suggest that , even though HCMV is capable of activating integrin/Src-signaling in both fibroblasts and monocytes , there are significant differences in the utilization of receptor-mediated signaling networks between these two cell types [26]–[28] , [99] . Biologically , these differences likely correspond to the nature of the cell and the role the cell plays during the infection process . There is a growing list of evidence demonstrating that herpesviruses can utilize different mechanisms of internalization into different cell types . Depending on the cell type , herpes simplex virus ( HSV ) was shown to enter cells using both pH-dependent and pH-independent fusion with endosomes , as well as direct fusion with the plasma membrane [109]–[111] . Similarly , Miller and Hutt-Fletcher demonstrated that Epstein-Barr virus ( EBV ) penetrates normal B cells by pH-independent endosomal fusion; however , direct fusion with the plasma membrane is utilized in EBV infection of epithelial cells and transformed B cells [112] . Direct fusion with the plasma membrane was described as a mode of HCMV entry into fibroblasts [113] , and endocytosis was postulated to be involved in the HCMV internalization process into epithelial and endothelial cells [41] , [85] . Actin has been implicated as a key player in the process of endocytosis [42]–[44] . We have recently documented that the expression of paxillin , a protein regulating actin cytoskeleton dynamics and governing endocytosis [114] , is elevated upon HCMV infection of monocytes and is critical for efficient HCMV entry into this cell type [27] . The findings in our current report indicate that this regulatory protein might be used by HCMV in target monocytes as a convergence point that links integrin/Src-signaling to the regulation of the actin cytoskeleton that is needed in the HCMV entry process . The role of paxillin-mediated actin rearrangement in cellular motility has been studied extensively [76]–[78] , while the importance of paxillin in rapid actin filament rearrangement during viral entry has remained unknown . Here , we documented that HCMV caused a rapid and transient actin depolymerization in infected monocytes . Moreover , we found that there was a decreased level of polymerized actin in paxillin-deficient monocytes compared to mock- or control siRNA-treated cells , which supports earlier reports showing that increased paxillin phosphorylation enhanced actin polymerization and cytoskeleton rearrangement [74] , [75] . Thus , we suggest the importance of actin regulation in HCMV internalization and implicate endocytosis as a possible route of entry into monocytes . We demonstrated that the disruption of the actin cytoskeleton prevents the efficient entry of BADrUL131 , TB40/E , indicating that there is an interrelationship between the presence of the gH/gL/UL128-131 complex on the HCMV envelope and the dependence of HCMV entry into monocytes on actin regulation . A disruption of actin regulation did not have any effect on BADwt and BADrUL131 entry into fibroblasts . Thus , our data again provide evidence that the effect of the UL128-131 complex on HCMV entry is cell type specific . Interestingly , we found that we could rescue the inability of BADrUL131 to enter paxillin-deficient monocytes by inducing actin polymerization after viral binding to the cell , suggesting that , similarly to studies on HPV entry [90] , by inhibiting actin polymerization in monocytes HCMV may localize to tubular invaginations of plasma membrane due to the inability of endocytic vesicles to pinch off . However , this theory needs further investigation . Together , our data support the notion that a tight regulation of the actin cytoskeleton plays an essential role in the ability of HCMV to efficiently enter monocytes . The exploitation of actin rearrangement by HCMV points out the possible role of endocytosis in the viral entry process [91] . Using several pharmacological inhibitors targeting different modes of endocytosis , we found dynasore , a dynamin inhibitor , to be a potent agent impeding this process . Moreover , a 60% inhibition of HCMV entry was observed in infected monocytes that were deficient in dynamin expression , which provides an even stronger support for an importance of dynamin regulation in efficient HCMV entry into monocytes . As dynasore lacked an inhibitory effect on HCMV entry into fibroblasts , and with no apparent effect of this drug on BADwt entry into monocytes , we argue that only virus expressing the gH/gL/UL128-131 complex is able to utilize the endocytic entry into monocytes . Consequently , our data indicate that HCMV , expressing the gH/gL/UL128-131 complex , uses a dynamin- and actin-dependent endocytic route of entry into target monocytes , which as our current studies showed is similar to the characteristics of HCMV entry seen in epithelial cells . A similar strategy of viral entry is employed by human adenovirus 2 , in which the penton complex interacts with integrins promoting dynamin- and actin-based viral endocytosis [81] , [115]–[118] . Thus , HCMV joins the growing list of viruses ( i . e . EBV [79] , Vaccinia virus [79] , HIV [80] , Adenovirus [81] ) that utilize the actin cytoskeleton to enhance their infectivity . It has been suggested that , within a number of endocytic pathways , macropinocytosis is the most likely route of HCMV entry into in vivo relevant cell types [119] . As HCMV measures 200–300 nm in diameter [120] , small vesicles created in clathrin- , caveolea- or micropinocytosis-based uptakes could not accommodate virions of such size; this possibility is also supported by our results . Additionally , EBV and HSV in certain cell types have also been considered to utilize macropinocytosis [110] , [111] , [121] . Mercer and Helenius compiled the experimental criteria allowing for a characterization of macropinocytosis as a mode of viral entry [91] . Our findings showed that HCMV entry into monocytes is actin rearrangement- and dynamin-dependent , supporting macropinocytosis as the mechanism of entry ( the current report ) . However , we also know that the HCMV entry process into monocytes is not regulated by PI ( 3 ) K [26] , an important regulator of macropinocytosis [91] , suggesting that HCMV may use a non-classical macropinocytosis route for its internalization into target monocytes . Thus , these data again underscore the specificity of the biological processes in HCMV-infected monocytes regulated via distinct signaling events . As Figure 6 depicts , our overall data showed that HCMV equipped with the pentameric gH/gL/UL128-131 complex is able to engage integrins on the surface of monocytes and , through the induction of the integrin/Src/paxillin signaling pathway , to regulate actin cytoskeleton , which consequently leads to efficient viral internalization and productive infection in target monocytes . These immediate early biological changes in target monocytes resulting in the appropriately executed entry process further enable HCMV , through prolonged receptor-mediated regulation and virus-enhanced cellular motility [27] , to utilize the natural sentinel role of these cells . This mode of action allows HCMV to avoid the host immune response and effectively spread within the host . Scivano et al . recently demonstrated that HCMV-infected fibroblasts released a mixture of endothelial cell-tropic ( high level of the gH/gL/UL128-131 complex ) and non endothelial cell-tropic ( low level of the gH/gL/UL128-131 complex ) virions and the spread of this virus was supernatant-driven [122] . In contrast , endothelial cells released virions with low levels of the pentameric complex into the supernatant , and virions with a high level of the gH/gL/UL128-131 complex were cell-associated . Therefore , the spread of the endothelial cell-originated virus was predominantly focal . Additionally , we previously showed that HCMV promotes expression of adhesion molecules in endothelial cells , promoting the recruitment of naïve monocytes and augments monocyte transendothelial migration by increasing the permeability of endothelium [123] . Moreover , we found that the virus was transferred to monocytes while translocating through the endothelium [123] . Taken together , the expression of the gH/gL/UL128-131 complex on the viral envelope and the requirement for the appropriate level/type of integrin/Src/paxillin signaling , induced by the pentameric complex in infected cells , might be the evolutionary mechanism that assures the predominance of close proximity HCMV infection minimizing the recognition of virus by the immune system and at the same time allowing for efficient viral spread within the host . In summary , our findings provide , for the first time , an explanation of why the gH/gL/UL128-131 complex is critical for viral internalization into a clinically relevant cell type , and connects it with the importance of receptor-mediated cellular changes for a successful HCMV infection . Our report also indicates that the inability of the gH/gL/UL128-131 complex to stimulate the integrin/Src/paxillin-mediated signaling pathway in fibroblasts may explain the conundrum of why there is a strong negative pressure against the UL128-131 region during the passaging of clinical isolates in fibroblasts . Therefore , to our knowledge , our studies are the first to demonstrate a possible molecular mechanism for why the gH/gL/UL128-131 complex dictates HCMV tropism .
We explicitly state that the Louisiana State University Health Sciences Center-Shreveport ( LSUHSC-S ) Institutional Review Board ( IRB ) approved our study ( the approved study number is H99-064 ) and that all HIPAA and LSUHSC-S IRB guidelines were followed for the use of human subjects in our study . In addition , the National Institutes of Health has approved our study using human subjects as part of our funded application ( AI050677 ) . We also explicitly state that informed written consent was provided by study participants ( using an approved consent form , from the approved protocol H99-064 ) ; that is , an approved consent form was signed by the participants and collected for all subjects utilized in our study . No vulnerable populations were utilized in our study ( minors , pregnant women , prisoners , etc . ) ; thus only the participants themselves signed the consent approved forms and no legal guardians were required to provide consent to participate in the study . As per HIPAA and LSUHSC-S IRB guidelines , we follow all protocols on the welfare of human subjects and on the privacy of all collected information . The LSUHSC-S IRB is under the guidance of the Human Research Protection Program ( HRPP ) and is headed by the chancellor of the University , Dr . Robert Barish , M . D . ( the website for our HRPP is as follows [http://www . lsuhscshreveport . edu/HRPP/HRPPHome . aspx] ) . The board is as follows: Institutional Official ( Robert A . Barish , MD , MBA ) ; HRPP ( Robert E . Walter , MD , MPH; Aliese Seawright , MS , C . I . P; Kathleen Bloomingdale , BS; and Janice Soderstrom , RN , BSN ) ; QA/QI ( Sherry Mosura , RN , BSN , CCRC; Yvette Viverette , RN , CCRC ) ; and , IRB ( Christina Copeland , MPH; Shweta Khedlekar , MS; Ann Johnson; Crystal McGee ) . AD169-derived bacterial artificial chromosome ( BAC ) viruses BADwt and BAD rUL131 ( containing a repaired UL131A mutation , such that the gH/gL/UL128-131 complex is identical to that of HCMV TR strain ) , were previously constructed and characterized [31] . In addition to these BAC-based viruses , several additional HCMV strains were used: a green fluorescent protein ( GFP ) -labeled TB40/E-UL32 ( TB40/E ) [39]; a high fibroblast-passaged TB40 ( TB40/F ) that has lost the ability to infect endothelial cells [39]; our Towne/E ( with the UL128-131 complex intact; Towne p . 40 ) ; a intermediate fibroblast-passaged Towne ( Towne p . 51 ) ; and , a high fibroblast-passaged Towne ( without a functional UL128-131 complex; Towne p . 57 ) [123] . The TB40/E and TB40/F strains [39] are related ( TB40/F is a high passaged TB40/E ) as is our Towne/E and Towne/F strains [39] . TB40/E and Towne/E retain endothelial and epithelial cell , as well as monocyte tropism , while TB40/F and Towne/F have lost the ability to infect these clinically relevant cells types and are restricted to fibroblast infection . As shown by the data in Figure 1 , those viruses retaining endothelial and epithelial cell , and monocyte tropism possess an intact UL128—131 complex , while those restricted to fibroblast infection have lost this complex . It is currently unclear if additional mutations have been incorporated into the more highly passaged viruses . The viruses were cultured in human embryonic lung ( HEL ) fibroblasts with 4% heat-inactivated fetal bovine serum ( FBS ) and purified on a sucrose gradient and resuspended in RPMI 1640 ( Cellgro , Mediatech Inc . , Herndon , VA ) . Monocytes were infected with purified virus at the multiplicity of infection ( M . O . I . ) of 0 . 1 for entry assays or M . O . I . of 5 for the various functional assays , unless otherwise stated . Mock infection was carried out by adding an equivalent volume of RPMI 1640 ( Cellgro ) to monocytes . Double-density gradient centrifugation was used to purify human peripheral blood monocytes [57] . Whole blood was collected from donors by venipuncture . Mononuclear cells were then collected by centrifugation through a Ficoll Histopaque 1077 ( Sigma-Aldrich , Inc . , St . Louis , MO ) gradient . Next , the collected cells were washed in 1× Phospho-Buffered Saline ( PBS; Cellgro ) to remove platelets . Monocytes were then isolated by centrifugation through a Percoll ( Pharmacia Biotech , Inc . , Piscataway , NJ ) gradient and suspended in RPMI 1640 ( Cellgro ) supplemented with 1% human serum ( Sigma-Aldrich Inc . ) . After isolation , monocytes were cultured under non-adherent conditions ( unless specified otherwise ) in RPMI 1640 ( Cellgro ) supplemented with 1% human serum ( Sigma-Aldrich Inc . ) at 37°C with 5% CO2 overnight , prior to any treatment , if not stated otherwise . Primary human embryonic lung ( HEL ) fibroblasts ( passage 13–20 ) were cultured in Eagle's minimal essential media ( MEM , Cellgro ) supplemented with 10% heat-inactivated fetal bovine serum ( FBS , Gemini , Woodland , CA ) , penicillin ( 100 IU/ml; Cellgro ) , and streptomycin ( 100 µg/ml; Cellgro ) and Plasmocin ( 25 µg/ml; InvivoGen , San Diego , CA ) . Primary human mammary epithelial cells ( HMEC ) were culture in EBM-2 Basal Medium ( Cambrex Inc . , East Rutherford , NJ ) . The following standard treatment groups were employed in our study: dimethyl sulfoxide ( DMSO; Sigma-Aldrich Inc . ) as a solvent control; 1 µM PP2 ( Src tyrosine kinase inhibitor; EMB Biosciences , Inc . ; La Jolla , CA ) ; 1 µM AG1478 ( EGFR tyrosine kinase inhibitor; EMB Biosciences , Inc . ) ; 0 . 5 µM jasplakinolide ( inducer of actin polymerization; Enzo Life Sciences International , Inc . ; Plymouth Meeting , PA ) ; 2 . 5 µM latrunculin A ( inhibitor of actin polymerization; Enzo Life Sciences International , Inc . ) ; 50 µg/ml nystatin ( disrupts caveolar structure and function; Enzo Life Sciences International , Inc . ) ; 200 µM genistein ( disrupts caveolae-mediated endocytosis; Enzo Life Sciences International , Inc . ) , 100 µM Rac1 inhibitor ( Rho GTPase inhibitor; Enzo Life Sciences International , Inc . ) and 50 µM dynasore ( dynamin inhibitor; Enzo Life Sciences International , Inc . ) were added 1 h prior to HCMV infection at 37°C with 5% CO2 . In one of the viral entry assays , 0 . 5 µM jasplakinolide was added after the shift in temperature to 37°C . In one of the viral entry assays , 5 U/ml of α-thrombin ( Enzyme Research Laboratories , Inc . , South Bend , IN ) was added prior to the shift in temperature to 37°C . Cells were also treated with 5 µg/ml of function-blocking anti-β1 integrin ( Millipore , Bedford , MA ) and/or anti-β3 integrin ( Millipore ) antibodies for 1 h at 4°C prior to HCMV infection , as well as with an IgG isotype control ( Santa Cruz Biotechnology , Inc . ) Monocytes or fibroblasts were pretreated with pharmacological agents and then infected . To examine the kinetics of receptor-mediated signaling pathways , cells were harvested using lysis buffer [50 mM Tris-HCl at pH 7 . 5 ( Fisher Scientific , Fair Lawn , NJ ) , 5 mM ethylenediaminetetraacetic acid ( EDTA; BioRad Laboratories , Hercules , CA ) , 100 mM sodium chloride ( Fisher Scientific ) , 1% Triton X-100 ( Fisher Scientific ) , 0 . 1% sodium dodecyl sulfate ( SDS; MP Biomedicals , Inc . , Solon , OH ) , and 10% glycerol ( MP Biomedicals , Inc . ) ] at the time points indicated . Samples were mixed with Laemmli's SDS-Sample Buffer ( Boston BioProducts , Boston , MA ) containing 2-mercaptoethanol ( Fisher Scientific ) . Equal protein amounts of the different samples were separated by continuous polyacrylamide gel electrophoresis ( SDS-PAGE ) and transferred to ImmunoBlot polyvinylidene difluoride ( PVDF ) membranes ( BioRad Laboratories ) . Western blot analyses were performed using primary antibodies recognizing the phosphorylated and non-phosphorylated forms of Src [phospho-Src ( Tyr416 ) antibody ( Cell Signaling Technology , Inc . , Danvers , MA ) and pan-Src ( Santa Cruz Biotechnology , Inc . , Santa Cruz , CA ) ] , of paxillin [phospho-paxillin ( Tyr18 ) antibody and pan-paxillin antibody ( Cell Signaling Technology , Inc . ) ] , of SAPK/JNK [phospho-SAPK/JNK ( Thr183/Tyr185 ) antibody and pan-SAPK/JNK antibody ( Cell Signaling Technology , Inc . ) ] , of Erk [phospho-Erk ( 204 ) antibody and pan-Erk1 antibody ( Santa Cruz Biotechnology , Inc . ) ] , of p70 S6K [phospho-p70 S6K ( Thr389 ) and pan-p70 S6K antibody ( Cell Signaling Technology , Inc . ) ] , as well as antibodies recognizing cellular F-actin ( Thermo Fisher Scientific; Rockford , IL ) , and recognizing HCMV proteins: pUL130 ( gift from Dr . Thomas Shenk ) and pp65 ( Virusys Corporation; Taneytown , MD ) . Probing for actin ( Santa Cruz Biotechnology , Inc . ) was used as a loading control . Donkey anti-rabbit ( GE Healthcare Life Sciences , Piscataway , NJ ) , donkey anti-mouse ( GE Healthcare Biosciences ) and donkey anti-goat ( Santa Cruz Biotechnology , Inc . ) conjugated with horseradish peroxidase ( HRP ) were used as secondary antibodies . Western blots were developed using ECL Plus Western Blotting Detection Reagents ( GE Healthcare Life Sciences ) . Monocytes were infected with HCMV at 4°C for 1 h . Subsequently , 2 mM of DTSSP [3 , 3′-dithiobis ( sulfosuccinimidylpropionate ) ] ( Thermo Fisher Scientific; Rockford , IL ) was added at 4°C for additional 2 h . Cells were spun down and lysed . Antibodies recognizing gH ( Virostat Inc . , Portland , ME ) , pUL130 , β1 ( Millipore ) , β3 integrins ( Millipore ) or an IgG isotype control ( Santa Cruz Biochenology , Inc . ) were added to lysate overnight at 4°C and then Protein A/G PLUS Agarose was added for 4 h at 4°C . Protein A/G PLUS Agarose beads with bound protein complexes were spun down , washed with a lysis buffer and resuspended in Laemmli Sample Buffer ( BioRad Laboratories ) containing β-mercaptoethanol ( Fisher Scientific ) . Proteins were separated on SDS-PAGE gels , and transferred to ImmunoBlot PVDF membranes . Antibodies recognizing the HCMV gH ( Virostat Inc . ) , pUL130 , β1 or β3 integrins ( Santa Cruz Biotechnology , Inc . ) were used in the western blot assays . Monocytes or fibroblasts were pretreated and infected . Total cellular DNA was harvested with E . Z . N . A . Tissue DNA Kit ( Omega Bio-Tek , Inc . ) . The quantitative PCR amplification and detection were performed as previously described [27] . Semiquantitative PCR was also used to assess the entry efficiency of different HCMV strains and the efficiency of dynamin knockdown . Product amplification was carried out using MyCycler thermocycler ( BioRad Laboratories ) , with the following PCR mix: 1× iTaq buffer ( BioRad Laboratories ) containing 1 . 25 U of iTaq DNA polymerase ( Invitrogen Corp . ) and a 50 µM concentration of each deoxynucleotide ( Invitrogen Corp . ) . Primers specific for HCMV IE1-72 ( sense , 5′-AGTGACCGAGGATTGCAACG-3′; antisense , 5′-CCTTGATTCTATGCCGCACC-3′ ) , GAPDH ( sense , 5′-GAAGGTGAAGGTGGAGT-3′; antisense , 5′-GAAGATGGTGATGGGATTTC-3′ ) , 18S rRNA ( sense , 5′-CGAGCCGCCTGGATACC-3_; antisense , 5′-CAGTTCCGAAAACCAACAAAATAG-3′ ) and dynamin ( Santa Cruz Biotechnolgy , Inc . ) were used to amplify target sequences . All primers , exempt primers for dynamin , were obtained from Integrated DNA Technologies , Inc . ( Coralville , IA ) . Monocytes or fibroblasts were resuspended in Human Monocyte Nucleofector Solution or Basic Nucleofector Solution - Primary Fibroblasts , respectively ( Lonza Group Ltd , Basel , Switzerland ) containing 300 nM of paxillin siRNA ( Dharmacon , Inc . , Lafayette , CO; sequence: 5′-GUGUGGAGCCUUCUUUGGUUU-3′ ) , 300 nM of dynamin siRNA ( Santa Cruz Biotechnology , Inc . ; sequence: 5′-CCAUCAUGCACCUCAUGAUTT ) , 300 nM control siRNA ( Santa Cruz Biotechnology , Inc . ) or RNase-free water ( Invitrogen Corp . ) and then transfected using an AMAXA Nucleofector ( Lonza Group Ltd ) . siRNA-transfected monocytes or fibroblasts were mixed with pre-equilibrated Human Monocyte Nucleofector Medium [supplemented with 10% human serum ( Sigma-Aldrich Inc . ) ] or Basic Nucleofector Medium – Primary Fibroblasts [supplemented with 4% fetal bovine serum ( Sigma-Aldrich Inc . ) ] , respectively ( Lonza Group Ltd ) and incubated for 48 h at 37°C in 5% CO2 . Performed as previously described [26] , [27] . Briefly , monocytes , fibroblasts , or epithelial cells were treated and then HCMV infected ( M . O . I . of 0 . 1 ) for 1 h at 4°C , washed with 1× PBS ( Mediatech , Inc . ) and temperature shifted to 37°C for 1 h . Cells were washed and treated with 2 mg/ml solution of Proteinase K ( Promega ) for 1 h at 4°C . Monocytes or fibroblasts were then harvested , total DNA was isolated and quantitative real-time PCR or semiquantitative PCR were performed using primers complementary to genomic HCMV DNA ( the UL123/IE1-72 Exon 1 region ) and cellular 18S rRNA . Results were plotted as a mean ±SEM . Student's T-tests were performed and p value of <0 . 05 was used for the measurement of statistical significance between samples . To monitor the accuracy of the assay , representative samples of infected monocytes were kept at 4°C without a temperature shift and processed as described above . Isolated monocytes were infected with BADwt or BADrUL131 ( MOI = 1 ) at 37°C for 1 hour then plated on fibronectin-coated cell culture dishes in RPMI 1640 supplemented with 10% human serum . Cells were incubated at 37°C in 5% CO2 for 5 days and 3 weeks post infection and media was changed every 5 days . At 5 days post infection , total DNA was extracted . At 3 weeks post infection , total RNA was extracted . Semiquantitative PCR and RT-PCR were performed using primers complementary to the viral UL123/IE1-72 Exon 1 region and cellular 18S rRNA . Alexa Fluor 488-labelled DNA probe was synthesized according to a protocol in FISH Tag DNA Green Kit ( Invitrogen ) using BAC DNA encoding the whole HCMV genome . Monocytes were HCMV infected ( M . O . I . of 5 ) and kept at 37°C in 5% CO2 in RPMI 1640 supplemented with 10% human serum for 5 days . Cells were washed twice with 1× PBS and then fixed for 15 min . with 4% paraformaldehyde ( Fisher Scientific ) . Cells were cytospun onto slides , dried and fixed with methanol ( VWR , International , Radnor , PA ) :acetic acid ( VWR International ) ( 3∶1 ) for 15 min at room temperature ( RT ) . Then cells were dehydrated at RT by subsequent washes with 70% , 85% , and 100% ethanol ( Alcohol and Chemicals Co . , Shelbyville , KY ) . Next , DNA inside cells was denatured by incubating cells with 70% formamide ( Fisher Scientific ) in 2× SSC ( Boston BioProducts ) at 70°C for 2 min . The dehydratation step followed at −20°C with subsequent washes with 70% , 80% , and 95% ethanol . Denatured DNA probe was then applied to cell spots and hybridization continued overnight at 37°C . After this time , cells were washed with 50% formamide in 2× SSC at 37°C , washed in 1× PBS and the blocking solution [1× PBS/5% normal goat serum ( Santa Cruz Biotechnology , Inc . ) /1∶80 of Fc Blocking Reagent ( Miltenyi Biotec GmbH , Bergisch Gladbach , Germany ) /0 . 3% Triton X-100 ( Fisher Scientific ) ] was applied for 1 h at RT . Next , cells were incubated overnight with 20 µg/ml of anti-Alexa 488 rabbit IgG ( Invitrogen ) resuspended in the blocking solution . After this time , cells were washed with 1× PBS , the blocking solution was applied for 1 h . Cells then were incubated for 1 h with goat anti-rabbit Alexa 647 IgG ( Invitrogen ) resuspended in the blocking buffer . Next , cells were washed and mounted using SlowFade Gold with DAPI antifade reagent ( Invitrogen ) . Images were taken using Leica TCS SP5 confocal microscope . | We previously demonstrated that HCMV , by engaging cellular receptors , changes the biology of blood monocytes , allowing for efficient viral entry into these cells and their use as virus-carriers in HCMV systemic spread . However , it was unclear how HCMV induces receptor-mediated signaling in infected cells . Here we report that HCMV by expressing a specific complex of five glycoproteins , present on HCMV clinical isolates , engages cellular integrin receptors and subsequently triggers integrin-mediated signaling leading to efficient viral entry into monocytes and productive infection of monocyte-derived macrophages . We also demonstrate that the HCMV pentameric complex has an inhibitory effect on integrin-mediated signaling in fibroblasts , an in vitro model system of HCMV infection , suggesting that the presence of the pentameric complex is not advantageous for HCMV infection of fibroblasts . Together , our results argue that HCMV uses distinct mechanisms to enter monocytes and fibroblasts . In support , our findings indicate that HCMV utilizes an endocytic-like route of entry into monocytes that is in contrast to viral fusion at the cell surface seen in fibroblasts . Our studies provide a molecular explanation for a previously observed critical role of the HCMV pentameric complex during infection of clinically relevant cell types , which in the future may lead to the development of better targets for antiviral therapy . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"biology"
] | 2013 | The HCMV gH/gL/UL128-131 Complex Triggers the Specific Cellular Activation Required for Efficient Viral Internalization into Target Monocytes |
Allotetraploid cotton species are a vital source of spinnable fiber for textiles . The polyploid nature of the cotton genome raises many evolutionary questions as to the relationships between duplicated genomes . We describe the evolution of the cotton genome ( SNPs and structural variants ) with the greatly improved resolution of 34 deeply re-sequenced genomes . We also explore the evolution of homoeologous regions in the AT- and DT-genomes and especially the phenomenon of conversion between genomes . We did not find any compelling evidence for homoeologous conversion between genomes . These findings are very different from other recent reports of frequent conversion events between genomes . We also identified several distinct regions of the genome that have been introgressed between G . hirsutum and G . barbadense , which presumably resulted from breeding efforts targeting associated beneficial alleles . Finally , the genotypic data resulting from this study provides access to a wealth of diversity sorely needed in the narrow germplasm of cotton cultivars .
Domesticated cotton has a polyploid genome consisting of an AT- and DT-genome ( the “T” subscript indicates tetraploid nucleus ) . Approximately 1 mya , a polyploidization event gave rise to six described AD allotetraploid species with genome sizes ~2400 Mbp , mostly native to Central and South America [1–4] . The AT-genome ( 1700 Mbp ) is ~2-fold larger than the DT-genome ( 900 Mbp ) and there is approximately a 2-fold greater genetic distance between the related diploid G . raimondii ( D5 ) and DT than between the related diploids G . herbaceum ( A1 ) or G . arboreum ( A2 ) and AT . There are two major clades among the six tetraploid species , one containing G . hirsutum ( AD1 ) and the other containing G . barbadense ( AD2 ) . Both of these species were independently domesticated and produce long spinnable fiber . The remaining tetraploid species ( AD3 –AD6 ) AD1 is the source of the vast majority ( ~90% ) of worldwide cotton production [5] . AD2 accounts for another ~5%; its longer fibers are valued for high quality textiles . Attempts to produce stable AD1 x AD2 hybrids have resulted in fertile and productive F1 hybrids , but development of hybrid seed is generally cost-prohibitive . In addition , hybrid breakdown , hybrid sterility , and selective elimination of genes make genomic resources difficult to develop . As such , introgression of genetic material from AD2 into AD1 ( or vice versa ) is of particular interest . While introgression between species increases their respective genetic diversity , conversion events between sub-genomes of a polyploid would reduce diversity within a genome . Homoeolog conversion—also called gene conversion , non-reciprocal homoeologous recombination , or homoeologous gene conversion—is a phenomenon in which an allele from one genome of a tetraploid overwrites its homoeolog in the other genome . For example , a DT-genome allele overwrites its AT-genome homoeolog , resulting in 4 copies of the DT-genome allele and 0 of the AT-genome allele , instead of 2 of each as would normally be expected . Homoeologous conversion has been identified in various tetraploid groups , including Brassica [6 , 7] and Gossypium [8 , 9] . Homoeologous conversion may be caused by non-reciprocal homoeologous recombination or other sources of double-strand break repair , although the specific mechanisms or causes for such events is still uncertain . It has been hypothesized that homoeologous recombination is a major force in the evolution of desirable traits in allopolyploid crops [10] , suggesting that it may be the reason that fiber traits in cotton have been selected on the DT-genome . The majority of genetic diversity among allopolyploid cotton species has been attributed to homoeologous conversion [11] . Identification of homoeologous conversion events using short read data from cotton or other allopolyploid genera requires specialized software . We have identified and implemented two different strategies to categorize mapped reads from tetraploid cotton to their genome of origin: PolyCat [12] and PolyDog [13] . Both programs are freely available as part of BamBam [14] at https://sourceforge . net/projects/bambam/ and were used as part of this study . PolyCat uses SNP-tolerant mapping of GSNAP [15] with an index of known homoeo-SNPs ( SNPs that differentiate the A- and D-genomes ) instead of its traditional use to index SNPs in the human genome sequence . Consequently , the reads are aligned to a single diploid sequence ( representing a relative of one of the parent genomes ) with minimized mapping bias between genomes . The end result of these strategies are sets of reads that belong to the AT- or DT-genomes ( in addition to reads that do not overlap a homoeo-SNP ) . Reads separated by genome provide a rich dataset for genome analyses within and between sub-genomes . The results of these analyses provide insight into genetic diversity , evolution , and specific traits of specific plant species . Previous re-sequencing efforts of other domesticated plant genomes such as corn , tomato , and cotton diploids have investigated mutations , selection , and linkage disequilibrium [16–18] . In this study , we apply Illumina technology to re-sequence and compare the genomes of 34 cotton tetraploids from 6 species at average coverage 23x per accession , whereas previous cotton tetraploid resequencing efforts have used only minimal coverage . We we examine the comparative evolution and genetic diversity of the polyploid cotton species and genomes by mapping reads to the diploid A- and D-genome reference sequences of G . arboreum [19] and G . raimondii [20] , as well as to the recently published drafts of the cotton tetraploid genomes [21 , 22] . Mapping to the diploid sequences for this report is tenable because 1 ) the AT- and DT-genomes do not have common loci positions and 2 ) >25% of the draft tetraploid sequences remain formally unanchored to either AT- or DT-genomes . Much of our study included comparisons between A and D ( or AT vs . DT ) , and the comparisons are only possible in regions present in both A and D genomes , making the draft tetraploid sequences less informative . To improve results based on diploid sequences , we account for the differences between the respective diploid and tetraploid genomes by adjusting the diploid reference sequences to the genotypes observed in the tetraploid species .
PolyDog read mapping and categorization uses a complete representation of the tetraploid genome by mapping each set of reads to both diploid reference genomes of A2 and D5 . Approximately 60% of reads from tetraploids mapped to unique loci on the D5 reference , while 70% mapped to unique loci on the A2 reference ( Fig 1 ) . The larger mapping percentage for the A2 reference is likely because the AT-genome is larger than the DT-genome , so more reads drawn randomly from the tetraploid should be A-like than D-like . The difference is only 10% because much of the extra A-genome sequence is either repetitive ( preventing unique mapping by short reads ) or simply absent from the reference sequence . More reads were categorized by both PolyCat and PolyDog to the AT-genome than to the DT-genome . This is likely due to 1 ) the larger size of the A-genome and 2 ) the greater genetic distance between D5 and DT , which slightly decreases the effectiveness and accuracy of read categorization . When using the A2 reference instead of the D5 reference , the frequency of categorization was lower because less homoeo-SNPs have been defined in the A2 reference SNP index . In addition , a greater fraction of the A2 reference is non-homoeologous sequence , resulting in more reads that map to the reference but will not be able to be categorized because they only map to A-genome unique sequence . More reads overall were categorized by PolyDog than by PolyCat because PolyDog is able to categorize these reads mapped to non-homoeologous regions [18] . Categorization error rates were measured by mapping diploid reads to each diploid genome ( S1 Table ) . The end result of read mapping and categorization was a read alignment ( BAM ) file for each genome ( AT and DT ) in each tetraploid accession . We also mapped reads to the tetraploid TM-1 reference sequence [20] . The numbers of mapped and categorized reads were less than those obtained with PolyDog using the diploid reference sequences . In addition , a significant percentage of the tetraploid sequence was unanchored to either an AT- or DT-genome . Unanchored scaffolds could be due to either partial assembly or mis-assembly . Thus , further analyses did not use the tetraploid sequence as a genome reference ( S1 Table ) . Eventually , additional improvement of the reference tetraploid sequences may provide better rates of read mapping than PolyDog , but PolyDog is currently the most thorough method of mapping polyploid reads in cotton . We analyzed evolutionary relationships by examining SNPs among the PolyDog-categorized reads . Within read alignments , we identified SNPs between genomes ( termed “homoeo-SNPs” ) and between accessions ( “allele-SNP” ) . Homoeo-SNPs were first identified between the diploids A2 and D5 and then between the PolyCat-categorized AT- and DT-genomes of AD1 , AD2 , AD3 , AD4 , and AD5 ( i . e . between sub-genomes ) . Between 19 . 2 and 28 . 5 million homoeo-SNPs ( 35 . 6 million total unique loci ) were found when using the D5 reference ( Table 1 ) . There were 11 . 2 million homoeo-SNPs positioned on the D5 reference sequence that were shared within all tetraploid species ( S1 Fig ) . Of these homoeo-SNPs , 9 . 4 million homoeo-SNPs were shared within all tetraploid species and they were also found between the diploid genomes . About 12–15% of homoeo-SNPs were in annotated genes . There were 1 , 358 genes with no homoeo-SNPs identified in the tetraploid sequences aligned to the D5 reference sequences . These SNPs are available on CottonGen as D13 . snp4 . x , where x = 0 for homoeo-SNPs found in the diploids , x = 1 for AD1 , x = 2 for AD2 , etc . We identified allele-SNPs within sub-genomes , between accessions of each species , using PolyDog-categorized reads . After filtering SNPs ( <10% minor allele frequency ) , there were 15 , 864 , 224 and 10 , 437 , 663 allele-SNPs in the AT- and DT-genomes , respectively . In both AD1 and AD2 , the number of AT-genome allele-SNPs was about 1 . 5x the number of DT-genome allele-SNPs ( Table 2 ) . Although breeding strategies typically ignore the genome size difference between AT and DT , the average diversity ( allele-SNPs per bp ) in the DT-genome was nearly 2x greater than the average diversity in the AT-genome after normalizing by genome size . Most of this SNP diversity was intergenic . Within gene annotations , less allele-SNPs were found in the AT-genome ( 947 , 157 allele-SNPs ) than in the DT-genome ( 1 , 638 , 565 allele-SNPs; S2 Fig ) . There were 1 , 173 genes that had 0 allele-SNPs in the AT-genome while their respective homoeologs had 5 or more allele-SNPs in the DT-genome . There were 1 , 835 genes that had 0 allele-SNPs in the DT-genome while their respective homoeologs had 5 or more allele-SNPs in the AT-genome ( S3 Fig ) . Copy number variants ( CNVs ) indicate regions of historic duplication and/or deletion , and there are various strategies used to identify them [23 , 24] . CNVs were detected in the ‘continuous-coverage’ of PolyDog-categorized BAM alignment files by CNVKit [24] . Deletions in the AT-genome were the longest and most common type of copy number variant , with ~69 blocks and 19 Mbp per accession ( Fig 2; S2 Table ) . Deletions in the DT-genome were much less frequent , with ~31 blocks and less than 5 Mbp per accession . Duplications were considerably less frequent than deletions , with less than 10 blocks and 1 Mbp per accession . In the DT genome , a similar number of duplications were found in AD1 and AD2 , but AT-genome duplications were more common in AD1 than in AD2 . No pattern in frequency of duplications or deletions appeared to distinguish wild and domesticated lines . In comparisons between species , AD4 had few duplications and deletions , and had a particularly low number of DT-genome duplications . Certain combinations of overlapping CNVs were also used to detect homoeologous conversion events ( see below ) . Deletions were much more conserved than duplications , although this is likely related to the larger number of deletions detected because shared deletions more likely to occur by chance ( S3 Table ) . The limited number of gene deletions suggest that the sub-genomes within the polyploid have not diploidized , and that there very small differences in the amount of genome fractionation ( i . e . gene loss ) between sub-genomes . Consequently , we use the term ‘sub-genome’ sparingly ( when needed for reader clarity ) because one genome is not nested inside another genome , and evidence that the AT- or DT-genomes ( i . e . sub-genomes ) are less than a complete genome is very limited ( S3 Table ) . Duplications in the AT-genome were more conserved than duplications in the DT-genome , but duplications differed greatly from accession to accession , even among the closely related AD1 cultivars . Generally , conservation rates of CNVs were higher in cultivars than in wild accessions and could have been the result of a recent shared ancestry . A new homoeologous conversion event would result in a long series of consecutive conversion-SNPs and overlapping duplications/deletions between homoeologs . Given the 5–10 million-year history of nuclear co-residency , the conversion events would be somewhat fractionated by historical recombination or by mutation accumulation . Two approaches were used to investigate genome conversion events in cotton: SNPs and overlapping CNVs . The SNP-based method would detect older homoeologous conversion events that have subsequently been obfuscated over time . The CNV method would detect recent conversion events . Events between the two temporal extremes should be faintly detected by both methods , though the date of polyploidization provides a hard time limit to how ‘ancient’ conversion events may actually be . In the first approach , gene conversion was detected by a parsimony-based method of SNPs , similar to that employed by other studies [8 , 9 , 11 , 21 , 25] . Reads were categorized to the AT- and DT-genomes with PolyCat , in order to allow intergenomic comparison at a nucleotide level . Genotypes were called using InterSnp with a minimum allele coverage of 5 reads . Polymorphic loci were selected where 75% of individuals had an alternate allele . These were tested for a genotype pattern indicative of homoeologous conversion in G . hirsutum and G . barbadense by comparing the tetraploid genotypes to the diploids ( as a proxy ancestor genotype ) . However , the diploids A2 and D5 do not precisely represent the true progenitors of the AT- and DT-genomes [5] . Mutations that have occurred in the extant diploid after their divergence from the progenitors of the polyploid will result in false positive events of simple conversion detection because both tetraploid genomes will match the diploid that did not have the mutation . For example , the equivalence of A2 = AT = DT ≠ D5 could be due to a mutation in the D5 lineage ( a D5 autapomorphism ) , rather than to a homoeologous conversion . To correct for diploid autapomorphies , we use the AD4 as an outgroup for AD1 and AD2 intra-genome comparisons [4 , 26] . If a putative homoeologous conversion was detected in AD4 as well as in AD1 and/or AD2 , then it was due to 1 ) a conversion event immediately after ( or coincidental ) with polyploidization or to 2 ) an autapomorphic mutation unique to one of the diploid lines [8 , 9 , 12] . Using the D5 reference sequence , 1 , 322 , 948 putative A-dominant events were found in AD1 and could be compared to AD4 . Of those , only 52 , 680 ( 4 . 0% ) were putative homoeologous conversion events after compared to the AD4 sequence . The remaining 1 , 270 , 268 were false positives ( autapomorphies in the D5 diploid ) or possibly occurred immediately after polyploidization . Similar numbers were observed for AD1 and AD2 ( S4 Table ) . A greater percentage of D-dominant conversion events were found: 65 , 276 ( 6 . 7% ) out of 979 , 045 . We repeated this analysis using the A2 genome reference . This change of reference sequence resulted in detection of a similar number of events , but more A-dominant than D-dominant conversions . This suggests that the choice of reference sequence may be a source of false positive events . Similar ratios of ‘true’ ( AD4-considered ) and false ( AD4-ignored ) conversion events were observed in AD1 and AD2 , and a little less than half of the likely homoeologous conversion loci were shared by AD1 and AD2 suggesting events prior to the division between the AD1 and AD2 clades . The number of conversion events can also be examined by considering consecutive , putative conversion-SNPs because not every pair of ‘ancient’ conversion-SNPs from a single event would not have been interrupted by recombination or by mutations . Very few consecutive loci in the genome supported homoeologous conversion and most were two-consecutive SNPs and not a larger series of consecutive SNP loci ( Table 3 ) . As with the conversion-SNPs discussed in the previous analysis , many more regions of consecutive homoeologous conversion SNPs were detected as dominant for the same genome as the reference . When the A2-reference sequence was used , fewer consecutive SNPs representing fewer regions were found , but they overlapped more genes . Thus , we found that nearly all of the SNP-based evidence for genome conversion to be indistinguishable from coincidental mutational noise within AD4 and other polyploid genomes , and from error inherent to our SNP-detection methods ( e . g . choice of genome reference etc . ) . A second approach was used to investigate conversion events across regions much larger than the size of a sequence read . In this case , read categorization should mis-categorize reads within converted regions resulting in a duplication of one loci ( i . e . ~2x coverage ) and a corresponding deletion ( i . e . no coverage ) at its homoeologous locus . In other words , overlapping CNVs ( duplications and deletions ) can be detected between bam files of AT- and DT-genome categorized reads . Very few putative homoeologous conversions of this type were detected ( Table 4 ) . As mentioned above , more deletions than duplications were found in all of the genomes analyzed and rarely did a deletion in one genome entirely ‘overlap’ a duplication at its homoeologous locus . One large possible conversion event was detected on Chromosome 12 , containing nearly all of the genes that are located in regions with evidence of conversion ( full or partially converted , S5 Table ) . This event was also detected in several accessions; however , various additional facts suggest that it was not a true conversion event ( although it may have been a true duplication and true deletion ) : 1 ) the accessions exhibiting this possible conversion are not monophyletic . They include some accessions of AD1 and AD2 , but not the other members of those species . 2 ) The duplication associated with this possible conversion event is ubiquitous among tetraploid lines , while the deletion associated with the possible conversion occurred in only a subset of the individuals with the duplication . 3 ) The duplication/deletion events ( deletion events in particular ) do not have the same start and stop sites . For these reasons , we suggest that this possible conversion event—the only major event suggested by our sequencing data—was likely not a true conversion event because a complex series of introgression and selection that would be needed to occur to find it in two separate species of cotton . Ascribing the overlap of real duplications and of real deletions to homoeologous conversion event ( s ) invokes a complicated interpretation to data that may be only coincidental detection of CNVs . There are six described tetraploid speices of cotton [27] . While AD1 and AD2 have been domesticated , the remaining tetraploid species ( AD3 –AD6 ) have not been domesticated because they do not produce spinnable fiber . Another unnamed island endemic of the Northern Line Islands is under consideration as a seventh tetraploid species ( Wendel , personal communication; AD7 ) . G . ekmanianum ( AD6 ) belongs to the AD1 clade and has only recently been described as a distinct species separate from AD1 [3] . G . darwinii ( AD5 ) belongs to the AD2 clade . G . mustelinum ( AD4 ) diverged from the other tetraploids prior to the divergence between the AD1 and AD2 clades , making it a useful outgroup for analyses of the cotton tetraploids . The position of G . tomentosum ( AD3 ) from Hawaii is either part of the AD1 clade or an outgroup to the split between AD1 and AD2 . The AT- and DT-genome SNP phylogenies positioned species consistent with previous observations [5 , 27] . The A-genome donor to the tetraploid lines was similar to extant , diploid G . herbaceum ( A1 ) and G . arboreum ( A2 ) , while the closest extant diploid relative of the D-genome donor is likely G . raimondii ( D5 ) [8] . The large number of SNPs between the A- and D-genomes ( between diploid and within tetraploid genomes ) result in separate monophyletic branches . Thus , separate phylogenetic analyses were performed for the AT-and DT-genomes . The tetraploids primarily split into two clades , one containing AD1 and the other containing AD2 . AD4 is basal to this split . AD5 is closely related to AD2 , while AD6 and AD7 are close to AD1 . AD3 is in the AD1 clade , but diverged shortly after the AD1 vs AD2 split , making it a more distant relative of AD1 than are AD6 and AD7 ( Figs 3 and 4 ) . In separate consensus bootstrap trees for the nuclear genomes , nearly all splits have 99–100% bootstrap support and only 2 splits ( both within the AD1 cultivars ) have less than 90% support ( 80% and 82% ) . The cultivated varieties in AD1 clustered together with wild AD1 accessions nearby ( Fig 3 ) , and the same pattern was observed with AD2 cultivars and wild accessions ( Fig 4 ) . Notably , PI-528167 ( although previously classified as an accession of AD2 ) clustered with the wild AD1 accessions . The two AD7 accessions formed a clade external to the wild AD1 , and AD6 was external to AD7 . We identified regions of introgression of AD2 alleles into AD1 cultivars by identifying SNPs between the PolyDog-categorized wild AD1 lines ( TX-0231 , TX-2094 , and PI-528167 ) and AD2 lines ( excluding PI-528167 because it is not actually AD2 ) . The wild AD1 lines were used to represent AD1 to avoid circularity in SNP examinations of introgression since wild accessions should have negligible amounts of introgression . Consequently , these SNPs provided a method to distinguish alleles that were truly introgressed instead of historical alleles that were ‘unimproved’ in one or more cultivars . Reads from AD1 cultivars with bases matching the wild AD1 consensus allele were assigned to the “AD1-like” category . AD1 reads from cultivars matching the consensus AD2 nucleotide indicated a locus of putative introgression . There were 3 , 558 , 401 and 1 , 913 , 744 diagnostic SNPs between the AD1 wild lines and the AD2 cultivars on the AT- and DT-genomes , respectively . Using a novel application of PolyCat where these SNPs of introgression were used as a ‘categorizing’ index ( as opposed to the standard use of PolyCat that uses homoeo-SNPs as the index ) , reads from each AD1 cultivar were categorized as either wild AD1-like or AD2-like . Regions with at least 10x coverage of AD2-like reads were identified with Eflen ( part of BamBam ) [14] . Genes in these introgressed regions were identified with BEDTools [28] . On average , each AD1 accession had 6 . 8 Mbp ( containing 1 , 605 genes ) of introgression on the AT-genome ( Fig 5 ) and 3 . 8 Mbp ( containing 1 , 934 genes ) of introgression on the DT-genome ( Table 5; Fig 6 ) . We performed a similar analysis to look for regions of introgression of AD1 alleles into AD2 cultivars . Between the AD1 cultivars and the wild AD2 lines ( all AD2 except Deltapine-340 , Phytogen-76 , and Giza-7 ) , we identified 5 , 217 , 270 and 2 , 803 , 879 diagnostic SNPs on the AT- and DT-genomes , respectively . As above , only wild AD2 lines were used to define “AD2-like” , so as to avoid circularity . We then used PolyCat to categorize reads from Deltapine-340 , Phytogen-76 , and Giza-7 as AD1-like or AD2-like . On average , each AD2 cultivar had 18 . 4 Mbp ( containing 1 , 731 genes ) of introgression on the AT-genome and 5 . 0 Mbp ( containing 1 , 679 genes ) of introgression on the DT-genome . Interestingly , Giza-7—an obsolete cultivar from the 18th century—had far fewer genes with evidence of introgression than the other cultivars . There was a large difference in the amount of DNA and the number of introgressed genes on the AT-genome , suggesting that breeding efforts have not equally focused on both genomes . In addition , the AT-genome has more noise ( false positive introgression at isolated loci ) in the introgression signal than the DT-genome , suggesting that one or more of our ‘wild’ AD2 accessions had some degree of introgression into their AT-genomes . The PolyCat analysis of introgression easily and robustly identifies areas of putative introgression , but it does not have a formal statistical test or quantitation of introgression . To validate PolyCat’s results , we also tested for introgression into AD1 and AD2 cultivars ( as opposed to wild lines ) according to the Patterson D-statistic , which uses three-population trees to measure admixture between genomes as a whole [29] . We performed the test for introgression of Phytogen-76 ( AD2 cultivar ) into 4 AD1 cultivars ( Maxxa , TM-1 , Coker-312 , and Tamcot-sphinx ) against TX-2094 ( wild AD1 ) and of Maxxa ( AD1 cultivar ) into 2 AD2 cultivars ( DeltaPine-340 and Phytogen-76 ) against PI-528243 ( wild AD2 ) . We found strong evidence of cross-species introgression into each cultivar ( S6 Table ) . Further , we again calculated the D-statistic , but only for those PolyCat-predicted regions of introgression . If introgressed regions were correctly identified by PolyCat above , then the D-statistic for those regions alone will be higher than when the D-statistic is calculated for the entire genome . Within the identified regions of introgression , the D-statistic was very high ( average 0 . 90 ) in each line , validating the PolyCat approach to identify regions of introgression .
In diploid organisms , gene conversion is considered a by-product of recombination where one allele is reconstructed using the second allele as a template [39] . In polyploids , a conversion event that uses homoeologous loci as a template can also result in conversion between ‘sub-genomes’ [8 , 9 , 11 , 21 , 25] . To distinguish between the traditional definitions of genetic conversion , we refer to the sequence-based events found between genomes ( a . k . a . sub-genomes ) sharing a nucleus as homoeologous conversions . Homoeologous conversion events were likely caused by historical non-reciprocal homoeologous recombination and it results in a region of a chromosome that is converted to the genotype of its homoeolog . Assuming this region was larger than the size of an sequence read ( 100 bp reads were used in this study ) , reads originating in the converted area would be incorrectly categorized as belonging to the homoeologous genome . For example , if the DT-genome overwrites a section of the AT-genome chromosome , then reads from that region were categorized as DT-genome , even though they originated from an AT-genome chromosome . Different methods can be used to search for two different types of homoeologous conversion: small , interspersed regions of SNP patterns ( SNP method ) , and large blocks of homoeologous conversion ( CNV method ) . In the SNP method , a consecutive pattern of shared nucleotides between diploid and tetraploid genotypes along the chromosome suggests homoeologous conversion . The SNP method is discreetly limited by read length , though we required for consecutive SNP occurrences ( independent of read length ) for homoeologous conversion to be considered . The vast majority of such pattern occurrences—both in our analysis and in that done by Guo et al . [11]—were positioned before the divergence of AD4 from the other polyploid species . A pre-AD4 homoeologous conversion is indistinguishable ( based on extant genotype pattern ) from an autapomorphic mutation occurring in one of the diploids . However , the length of time between the polyploidization event and AD4 divergence ( 0 to 0 . 5 million years ) was much shorter than the length of time where such an autapomorphy could occur in one of the diploids ( 1 to 2 million years ) . It is therefore likely that the majority of these putative homoeologous conversion events were actually autapomorphic mutations in the diploids . Examining putative homoeologous conversion events via SNP patterns , we observed 5% ( or less ) likely homoeologous conversions ( as opposed to likely autapomorphic mutations ) . This value is consistent with EST work predating the use of the reference sequences , which also suggested the possibility of autapomorphic SNPs yielding false positives for homoeologous conversion detection [8] . We found that homoeologous conversion detection was biased to favor dominant conversion events for the genome corresponding to the reference sequence used in the analysis . This suggests that many detected homoeologous conversions by SNPs may be due to artifacts of analysis and of imperfect data . Because of different genetic distances ( A2 is closer to AT and D5 is to DT ) and completeness of reference sequences [19 , 20] , false positive read mappings may have resulted in an overestimate of D-dominant homoeologous conversion events , as detected by both Guo et al . [11] and the current study . In the CNV method , large blocks of homoeologous conversion manifest as duplication in one genome and deletion in the homoeologous region of the other genome . These events can be detected using the CNV method ( duplication and deletion at homoeologous loci ) , although this detection suffers from increasing noise as the size of the sliding window is reduced , particularly under 1 kb . Overlapping duplication/deletion events have been detected in Brassica in whole genome sequencing data and their coverage patterns were attributed to non-reciprocal homoeologous recombination events [30] . These events detected by sequencing are reminiscent of chromosome rearrangements first observed by RFLP patterns in B . napus [6] . They are also likely recent events between the genomes because these large blocks of conversion have not been dissected by subsequent homologous recombination . In cotton , we did not detect clear support of any large blocks of homoeologous conversion . In addition , non-reciprocal homoeologous recombination has not been detected in cotton using genetic mapping technologies ( RFLPs , SSRs , or SNPs ) as it has in Brassica [6 , 7] . Perhaps , the block ( s ) on Chromosome 12 could be due to conversion events , but three pieces of independent evidence of conversion do not support it . While conversion events may occur frequently in other species , the size disparity between the AT- and DT-genomes may partly explain the lack of homoeologous conversion in cotton . Our results also cast light on the phylogenetic relationships among tetraploid species , including a newly characterized species , G . ekmanianum ( AD6 ) , and a possible new species of the Wake Island Atolls ( AD7 ) [2 , 3] . Previous work had constructed cotton phylogenies based on select genes [31] . However , we use an unprecedented breadth and depth of data in cotton with SNPs from across the nuclear genome , resulting in over 48 million allele-SNPs . Other studies have disputed the species status of AD6 and suggested that it is merely wild AD1 [32] . However , our results show that AD6 and AD7 are both external to the wild AD1 accessions TX-0231 and TX-2094 and they are distant from the AD1 cultivars . AD6 and AD7 also form distinct clades that cannot be considered as one monophyletic species . We conclude that the species status for AD6 ( G . ekmanianum ) and proposed AD7 are supported by whole genome sequence data . PI-528167 , although labeled an AD2 line , is clearly not AD2 , as it consistently clusters within the AD1 clade . Genotypic ( SSR ) and phenotypic data also suggest that PI-528167 is a wild AD1 rather than AD2 , corroborating this result ( Richard Percy , personal communication ) . These allele-SNPs form the beginning of a Cotton HapMap , similar to the database of SNPs constructed for the maize HapMap [16] . Our homoeo-SNP indices augment this database , resulting in a database of over 70 million SNPs among cotton species , though homoeo-SNPs are loci that researchers will want to avoid using for SNP-assays . The SNP data is organized according to their status as homoeo-SNPs between genome groups and allele-SNPs within genome groups . These SNPs are available for visualization and download on CottonGen [33] ( http://www . cottongen . org/data/download ) . Artificial selection associated with domestication causes a genetic bottleneck in all domesticated plant species . This bottleneck results in cultivars having less genetic diversity compared to wild lines , as seen in WGS data of recent studies of soybean [34 , 35] , tomato [17] , pepper [36] , bean [37] , rice [38] , and maize [39] . This phenomenon was observed in the AD1—and to a lesser degree AD2—cultivars , as manifested in the tight clustering of cultivars within the SNP-based phylogenetic trees . Small amounts of genetic diversity impose limits on the genetic potential of cotton breeding , since limited genetic diversity remained after domestication . Based on the WGS data produced in this study , significant genetic diversity exists in wild accessions of both G . hirsutum and G . barbadense . Some of the wild accessions sequenced here could be used for sources of additional genetic diversity in breeding programs . An effort to sequence all of the genetic diversity within cultivated and wild cotton accessions would provide a comprehensive perspective to inform genetic improvement of cotton . Domestication increased the conservation of copy number variants ( duplications and deletions ) among cultivars as opposed to wild cotton lines . This is likely be a reflection of selection during domestication , and perhaps our small degree of sampling . Our study shows that AT-genome duplications were more ( ~2x ) conserved than DT-genome duplications in AD1 cultivars , although not in AD2 . While many fiber QTL are found in the AT-genome as well as the DT-genome [40] , selection during domestication also appears to have favored AT-genome duplications . Also , AT-genome deletions were more conserved than DT-genome deletions in AD2 but not in AD1 . Since our sampling of AD2 accessions were mostly wild , it’s unlikely that this conservation was caused by artificial selection for those deletions . Rather , these deletions likely occurred in the ancestral AD2 line that gave rise to the modern species , possibly contributing to the speciation and fiber quality that distinguish AD2 from other tetraploid cotton species . Both of these findings ( greater numbers of duplications in AT of AD1 and greater numbers of deletions in DT of AD2 ) support the existence of independent domestication events for these two species . Evidence of past attempts to introduce desirable traits from AD1 into AD2 , or vice versa , was detected in the introgression detected in AD1 and AD2 cultivars ( Figs 5 and 6 ) . Some regions—including large regions of AT-Chr8 ( Fig 5 ) —exhibited evidence of introgression in all AD1 cultivars , suggesting a relatively early event , while other , larger regions—e . g . , DT-Chr09 ( Fig 6 ) —evidenced introgression in a smaller number of cultivars , suggesting more recent introgression in the pedigrees of these lines . Breeders have long attempted to transfer genes for disease resistance , fiber quality , and other traits between AD1 and AD2 , and we are now able to see genomic evidence of those efforts [5] . We also observed that an obsolete cultivar ( Giza-7 ) had fewer genes commonly introgressed compared to other cultivars and a greater level of noise ( i . e . fewer matched bases between wild AD1 and Giza-7 than other cultivars ) suggesting less selection for agronomic improvement . In addition to introducing specific , targeted traits , new combinations of introgression may provide an additional source of diversity for the extremely narrow germplasm of cotton cultivars . Resequencing the tetraploid genome of cotton provided insights into domestication , introgression , and homoeologous conversion in both G . hirsutum and G . barbadense . Our whole genome sequencing data supports the previously described independent domestication of these two polyploid species . The large genome-wide collection of SNPs between and within genomes provided an unprecedented examination of the single-nucleotide genetic diversity within the cotton genome , but a comprehensive assessment is not entirely complete . Additional re-sequencing of wild and domesticated cotton accessions will identify rare alleles , provide sufficient power for robust estimates of linkage disequilibrium ( LD ) , and further identify regions of unique sequence evolution and domestication . Here , our limited sampling of both tetraploid species prohibited effective investigation of LD and selective sweeps . Nevertheless , this resequencing data provided important insights into the polyploid nature of the tetraploid cotton genome . Polyploidy has been a key aspect of cotton evolution and necessitates special computational consideration to properly use short read sequence data because of the close sequence similarity of homoeologs . In light of the large amount of genome sequence data , we found rare evidence for limited homoeologous exchanges , though no conclusive homoeologous exchanges were identified . In general , the sequences of cotton homoeologous loci have not significantly changed after polyploidization , though some exceptions can be found in individual gene pairs . Further research is needed to identify any association between these exceptions and the phenotype of modern cotton .
In total , over 18 billion 100bp paired-end Illumina reads were generated by Huntsman Cancer Institute , BGI , University of California-Davis , and Mississippi State University across 33 accessions: 13 G . hirsutum , 15 G . barbadense , and 1 each of G . tomentosum , G . mustelinum , G . darwinii , G . ekmanianum , and 2 accessions from the Wake Island Atolls . Illumina sequence data for the diploids—3 G . herbaceum , 4 G . arboreum , and 4 G . raimondii—and one additional G . hirsutum were obtained from SRA . For Gossypiodes kirkii—an outgroup of the Gossypium genus—40 million 36 bp single-end Illumina reads were obtained from NCGR . Reads were trimmed with Sickle ( https://github . com/najoshi/sickle ) using a PHRED quality threshold of 20 . Sequences used and generated in the effort are available in the Sequence Read Archive ( S1 Table ) . Identification of homoeologous conversion events using short read data from cotton or other allopolyploid genera requires specialized software . We have identified and implemented two different strategies to categorize mapped reads from tetraploid cotton to their genome of origin: PolyCat [12] and PolyDog [13] . Both programs are freely available as part of BamBam [14] at https://sourceforge . net/projects/bambam/ and were used as part of this study . PolyCat uses GSNAP’s SNP-tolerant mapping with an index of known homoeo-SNPs ( SNPs that differentiate the A and D genomes ) instead of its traditional use to index SNPs in the human genome sequence . Consequently , the reads are aligned to a single diploid sequence ( representing a relative of one of the parent genomes ) with minimized mapping bias between genomes [15] . PolyCat then categorizes each tetraploid read to a genome ( AT or DT ) based on whether it matches the AT- or DT-genome bases at previously known homoeo-SNP loci [12] . PolyDog maps the same set of polyploid reads to two different diploid reference sequences ( e . g . one mapping analysis to an A-genome diploid reference and another mapping analysis to a D-genome diploid reference ) . Then PolyDog compares the quality of each read’s mapping to the different genome references and assigns the read to the genome that had a better mapping [18] . These two different approaches provide separate results that are used to address different , and sometimes complementary , biological questions . The major difference between results produced by PolyCat and PolyDog is that PolyCat only categorizes reads that map over known or putatively identified homoeo-SNPs . Consequently , it only categorizes reads from regions that are present in both genomes . If a read originates in a region specific to the AT-genome ( i . e . , no DT-genome homoeolog exists ) , then that read cannot be formally SNP-categorized as originating in the AT- or DT-genome . On the other hand , PolyDog can categorize reads virtually anywhere in the genome . In practice , this means that PolyDog categorizes more reads and produces a smoother coverage profile over more of the genome , while PolyCat produces islands of homoeologous coverage separated by regions that are either identical between genomes or specific to one genome or another [18] . PolyCat has a lower error rate than PolyDog and is preferred for situations in which the presence of genome-specific regions causes additional biases in the mapping results . PolyCat-categorized reads are all mapped to a single reference , allowing straightforward comparisons between AT and DT reads in regions of homoeology , particularly in areas of sequence conservation ( e . g . genes ) . PolyDog-categorized reads are mapped to two different references , making it difficult to perform direct homoeologous comparisons at a single nucleotide resolution . The primary alternative to read categorization methods is mapping reads to a ‘full’ reference sequence representing both genomes of tetraploid cotton , whether that sequence is a concatenation of two diploid genome sequences [20] or a de novo assembly of a tetraploid cotton [20 , 21] . This mapping approach is comparable to PolyDog , as it maps reads anywhere in the genome rather than only to homoeologous regions . As shown previously and in this study , the PolyDog method accurately maps ( and categorizes ) more reads to the two diploid references than traditional read mapping to the ‘full’ reference sequence method [18] . We primarily use PolyDog-categorized reads in this study , employing PolyCat only where it is necessary either to reduce the area in question to homoeologous regions or to directly compare homoeologs at a specific nucleotide position . All reads were aligned to both the D5 and A2 reference genomes with GSNAP using the options “-n 1 –Q” to require unique best mappings [15 , 20] . An index of homoeo-SNPs inferred from diploid whole-genome resequencing was used for GSNAP SNP-tolerant mapping ( “-v” option ) [18] . Reads were then categorized as originating in the AT- or DT-genome by PolyCat , using a diploid-based homoeo-SNP index . Briefly , the homoeo-SNP index was constructed by mapping reads from both diploids species to the ‘other’ genome reference ( e . g . A-genome reads to D-genome reference ) . SNPs between genomes were then identified and compiled into a SNP-index for GSNAP . The original diploid reads were then re-mapped ( e . g . A-genome reads to D-genome reference with SNP-tolerant mapping ) . In this second iteration , more reads were mapped because this time , reads were not penalized by mismatching SNPs during mapping . In addition , new SNPs between genomes were identified because now more reads were mapping to the reference . These new SNPs were added to this putative homoeo-SNP index . The process was repeated until no new putative homoeo-SNPs were found between diploids . Then reads from the tetraploid were mapped using the diploid SNP-index . Mapped reads overlapping putative homoeo-SNPs confirmed SNPs as homoeo-SNPs ( or not ) . The tetraploid reads were then categorized to the AT or DT-genome based on nucleotide matches at SNP loci . If the tetraploid base matched the A2-genome base , then read was categorized at AT . Some new homoeo-SNPs were discovered that were specific for the tetraploid genome A2 and D5 are not the actual genome ancestors of tetraploid cotton . These new tetraploid-specific homoeo-SNPs where also added to the SNP-index . Like the diploid reads , tetraploid reads were iteratively re-mapped to the diploid reference to identify additional homoeo-SNPs until no new homoeo-SNPs were found . This iterative process was repeated for each species so that each species has its own SNP-index . InterSnp ( part of BamBam ) was used to call SNPs between individuals with a minimum allele coverage of 5 reads per individual , and SNPs that consistently ( 75% of observed genotypes ) manifested in one genome of a species—and were consistently ( 75% ) absent in the other genome of that species—were called as homoeo-SNPs [26] . Only 1 accession each of AD3 , AD4 , and AD5 were available ( and these species have sufficiently narrow germplasm that one accession is a fair sampling of the species ) , so a 100% threshold was used , rather than 75% . Five tetraploid-based homoeo-SNP indices were then generated for each genome , one each for AD1 , AD2 , AD3 , AD4 , and AD5 , named D13 . snp4 . 1 through D13 . snp4 . 5 ( or A13 . snp2 . 1 through A13 . snp2 . 5 ) , respectively . We also made modified reference sequences for each genome of each tetraploid species by replacing the ancestral nucleotide with that indicated by the homoeo-SNP index . The newly identified species AD6 and AD7 are very closely related to AD1 ( as shown below ) , so mappings to AD6 and AD7 use the AD1-based homoeo-SNP indices and modified reference sequences . To estimate the number of SNPs between homoeologs , best-hits of reciprocal BLAST were used to establish a list of homoeologs AT-DT pairs [43] . Indel-induced mapping errors were corrected using GATK [44] . First , RealignerTargetCreator was run on a group of 20 AT-genome BAM files and on 20 DT-genome BAM files ( representing all tetraploid species ) . Second , IndelRealigner was used on each individual BAM file to adjust read alignments around the indels identified in the first step: 3 , 692 , 540 loci in the A2 reference and 2 , 195 , 978 loci in the D5 reference . SNPs and short indels were called—once for all AT-genome BAM files and once for all DT-genome BAM files—between the PolyDog-categorized genomes using InterSnp with a minimum coverage per allele of 5 reads and minimum frequency of 30% [14] . A neighbor-joining tree was constructed for each genome , bootstrapping 1000 sub-samples without replacement with 5% of SNPs in each sub-sample . Trees were generated by creating a distance matrix based on genotypes at all SNP loci , then running neighbor ( from PHYLIP ) with random sample ordering to build the actual tree [45] . The 1000 trees from the bootstraps were combined with consense ( from PHYLIP ) to make a single consensus tree . Trees were visualized in Geneious [29] . Small homoeologous conversions were analyzed by using PolyCat to categorize mapped reads from each tetraploid because PolyCat categorization allows for inter-genomic analysis at a nucleotide level [12] . Then SNPs were called with InterSnp across all species and genomes [26] . Consensus genotypes were called for each species at sites that had coverage from at least 75% of individuals ( 10/13 for AD1 and 11/14 for AD2 ) , and genotype patterns suggestive of homoeologous conversion in AD1 or AD2 were identified ( e . g . , A2 , AT , and DT have a C while D5 has a T ) . Copy number variants ( CNVs ) were called in the PolyDog-categorized AT- and DT-genomes of each sample , relative to their respective diploid relatives , using CNVKit [30] . Reads from 3 diploid A2 lines and 4 diploid D5 lines were mapped and categorized in the same manner as the reads from the tetraploids , providing reference coverage profiles for the A- and D-genomes , which serve to normalize for biases in sequence coverage that are shared between diploid and tetraploid members of a common genome . The coverage of each tetraploid genome was compared to the reference coverage profile of its diploid relative . The gene annotations for each reference sequence were provided as targets , and accessible regions of the genome were identified for filtering by a CNVKit utility script genome2access . py . Segments identified by CNVKit as having a log base 2 copy number of at least 1 . 0 were considered duplications in the tetraploid genome , and segments identified with a log base 2 copy number of -1 . 0 or less were considered deletions . | The polyploid genome of domesticated cotton contains two different copies of most genes because its genome is duplicated . Instead of 13 chromosomes like its wild relatives , domesticated cotton has 26 chromosomes ( 13 AT chromosomes and 13 DT chromosomes ) . Differences in the gene copies may hold keys to the genetic improvement of cotton . In fact , it has been thought that the two copies in the cotton genome interact in an unexpected way called gene conversion . In regular diploid genomes , gene conversion occurs when the maternal copy is used to ‘fix’ or ‘overwrite’ the paternal copy ( or vice versa ) during cell division . In cotton , this mechanism of conversion has been used to explain small DNA changes between genomes over evolutionary time . However , we do not see any evidence for conversion events in our new , large sequencing datasets . Our datasets and methods are much more robust than previous reports . Correction of the idea that “extensive homoeologous gene conversion is common in cotton” is important because 1 ) the cotton genome is used as a model for plant genome research and 2 ) efforts to induce homoeologous gene conversion in cotton breeding would be unsuccessful . In addition , this report discovers a large set of single-base changes ( SNPs ) among cotton varieties and makes them available to the research community for public use . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"biotechnology",
"genome",
"evolution",
"plant",
"products",
"computational",
"biology",
"departures",
"from",
"diploidy",
"sequence",
"assembly",
"tools",
"plant",
"science",
"tetraploidy",
"fiber",
"crops",
"phylogenetic",
"analysis",
"genome",
"analysis",
"crops",
"plant",
"genomics",
"molecular",
"biology",
"techniques",
"research",
"and",
"analysis",
"methods",
"crop",
"science",
"introgression",
"plant",
"genetics",
"molecular",
"evolution",
"polyploidy",
"cotton",
"molecular",
"biology",
"agriculture",
"molecular",
"biology",
"assays",
"and",
"analysis",
"techniques",
"agronomy",
"genetics",
"biology",
"and",
"life",
"sciences",
"genomics",
"evolutionary",
"biology",
"plant",
"biotechnology",
"evolutionary",
"processes"
] | 2016 | DNA Sequence Evolution and Rare Homoeologous Conversion in Tetraploid Cotton |
Myocardium damage during Chagas' disease results from the immunological imbalance between pro- and production of anti-inflammatory cytokines and has been explained based on the Th1–Th2 dichotomy and regulatory T cell activity . Recently , we demonstrated that IL-17 produced during experimental T . cruzi infection regulates Th1 cells differentiation and parasite induced myocarditis . Here , we investigated the role of IL-17 and regulatory T cell during human Chagas' disease . First , we observed CD4+IL-17+ T cells in culture of peripheral blood mononuclear cells ( PBMC ) from Chagas' disease patients and we evaluated Th1 , Th2 , Th17 cytokine profile production in the PBMC cells from Chagas' disease patients ( cardiomyopathy-free , and with mild , moderate or severe cardiomyopathy ) cultured with T . cruzi antigen . Cultures of PBMC from patients with moderate and severe cardiomyopathy produced high levels of TNF-α , IFN-γ and low levels of IL-10 , when compared to mild cardiomyopathy or cardiomyopathy-free patients . Flow cytometry analysis showed higher CD4+IL-17+ cells in PBMC cultured from patients without or with mild cardiomyopathy , in comparison to patients with moderate or severe cardiomyopathy . We then analyzed the presence and function of regulatory T cells in all patients . All groups of Chagas' disease patients presented the same frequency of CD4+CD25+ regulatory T cells . However , CD4+CD25+ T cells from patients with mild cardiomyopathy or cardiomyopathy-free showed higher suppressive activity than those with moderate and severe cardiomyopathy . IFN-γ levels during chronic Chagas' disease are inversely correlated to the LVEF ( P = 0 . 007 , r = −0 . 614 ) , while regulatory T cell activity is directly correlated with LVEF ( P = 0 . 022 , r = 0 . 500 ) . These results indicate that reduced production of the cytokines IL-10 and IL-17 in association with high levels of IFN-γ and TNF-α is correlated with the severity of the Chagas' disease cardiomyopathy , and the immunological imbalance observed may be causally related with deficient suppressor activity of regulatory T cells that controls myocardial inflammation .
At the present time , about 7 . 7 million people are infected and 28 million are at risk of being infected with Trypanosoma cruzi in Central and South America [1]–[3] . This hemoflagellate protozoan is the etiological agent of Chagas' disease . Most of the infected individuals remain asymptomatic during chronic infection ( 60–70% ) , characterizing the indeterminate form of the disease . Conversely , 30–40% of chronically infected patients progress to cardiac and/or digestive pathologic involvement [4] , [5] , and prognostic markers for heart disease progression are required . A balanced immune response during T . cruzi infection is critical to control the parasite burden in heart and digestive tissues [6] , [7] . Production of pro-inflammatory cytokines is required for activation of the effector T lymphocytes responses and is associated with the pathogenesis of Chagas' disease cardiomyopathy ( CC ) , while regulatory cytokines ( mainly IL-10 ) are related to protection [8] , [9] . Peripheral blood mononuclear cells ( PBMC ) from patients with CC produce more IFN-γ , TNF-α and IL-6 , and less IL-4 and IL-10 , compared to individuals with the indeterminate form of the disease [1] , [3] , [7] , [10]–[14] . However , other studies failed to demonstrate any correlation between production of Th1 and Th2 cytokines profile and the clinical stages of Chagas' disease [15] , being that further investigations to elucidate such mechanisms are necessary , one aim of this work . Regulatory T cells ( Treg ) are an important source of regulatory cytokines and are involved in the control of the local inflammatory response and in avoiding extensive tissue destruction . However , their presence in the site of infections is frequently regarded as an inducer of parasite persistence [16] . Treg are able to migrate to the site of cardiac inflammation triggered by T . cruzi , and to suppress the effector function of CD4 and CD8 T cells during infectious processes [17] . They suppress the proliferation of effector T cells ( CD4+CD25− ) when co-cultured , and can also inhibit the activation of auto-reactive T cells through the expression of co-inhibitory molecules ( CTLA-4 ) and the production of suppressive cytokines ( IL-10 , TGF-β , IL-35 ) [12] , [18] , [19] . Recent studies suggest that indeterminate Chagas' disease patients have higher frequency of CD4+CD25high T cells in comparison to cardiac and non-infected individuals in their peripheral blood [20] , [21] . Consequently , the measurement of CD4+CD25high T cells suppressive activity in patients with indeterminate and cardiac forms of disease could be an important tool to evaluate a regulatory mechanism that prevents cardiac damage , which was another aim of this work . Treg do not seem to play a major role in regulating the effector responses of CD8 T cells in the myocardium during the acute and chronic experimental T . cruzi infection , since the blockade of CD25 did not change the inflammatory response or parasite burden in mice [13] , [22] , [23] . However , the treatment with anti-GITR resulted in increased mortality , TNF-α production , and myocarditis with enhanced migration of CD4 , CD8 , and CCR5 leukocytes to the heart in the T . cruzi infected mice [13] . If Treg could be involved in the control of immune response and cardiac disease progression in Chagas' disease patients is other aim of this work . An additional lineage of effector CD4+ T helper lymphocytes , with potential regulatory properties , produces IL-17A that acts in several cells types leading the production of GM-CSF , IL-1 , IL-6 , and TNF-α , activation of NOS2 , metalloproteinases and chemokines , resulting in leukocytes recruitment [24]–[27] . Treatment of T . cruzi infected mice with anti-IL-17A mAb lead to increased myocarditis , premature mortality , and decreased parasite load in the heart , suggesting that IL-17 controls the host resistance . Also , IL-17 regulates Th1 cells differentiation , cytokine and chemokine production and the influx of inflammatory cells to the heart tissue [28] . IL-17A−/− mice infected with T . cruzi had a lower survival rate , multiple organ failure , and sustained parasitemia compared with wild-type mice , indicating that IL-17A is crucial to leukocyte activation that are critical for parasite killing [29] . Although it is not very clear , it seems to be a relationship between Tregs and Th17 cells . Differentiation of Th17 in the presence of Treg leads to increased specific cytokine release , what could be due the consumption of IL-2 [30] , [31] . Similarly , Treg cell depletion results in a reduced frequency of IL-17 producers through modulation of IL-2 [32] . In addition , Treg can also be converted into a variety of T effector cells , including Th17 cells [33] . The purpose of the present study was to analyze the potential participation of IL-17 and Treg in the development of different clinical manifestations of human chronic Chagas' heart disease . Our hypothesis was that patients with chronic Chagas' disease undergoing cardiomyopathy produce increased levels of IL-17 and have a reduced frequency or suppressive activity of Treg compared with those patients with the indeterminate form of the disease . We provide novel information about immunological mechanisms involved in the human T . cruzi infection that could be used for the development of chemotherapies , as well as for the evaluation of prognostic markers of disease .
The inclusion of the 39 subjects ( 10 controls ) in our investigation had the prior approval of an institutional ethics committee ( Hospital das Clínicas de Ribeirão Preto – USP , São Paulo , Protocol number 2285/2007; Brazil ) . Signed informed consent was obtained from all participants . All patients ( n = 29 ) had at least two positive serology tests for Chagas' disease , as determined by ELISA , immunofluorescence or hemagglutination techniques . All patients underwent a detailed clinical evaluation , 12-lead rest electrocardiogram ( EKG ) , chest X-ray and a 2D-echocardiogram . Twenty one patients had not received etiologic treatment and 8 had received full treatment with benznidazole ( 5 mg/kg/day ) for roughly 60 days . According to their clinical and laboratory characteristics ( Table 1 ) , the chagasic patients were divided in 3 groups: Group 1 ( n = 10 ) : Patients not treated with benznidazole and not showing signs of or only having mild cardiomyopathy , Group 2 ( n = 11 ) : Patients not treated with benznidazole but with moderate/severe cardiomyopathy , Group 3 ( n = 8 ) : Patients previously treated with benznidazole ( cardiomyopathy-free or mild cardiomyopathy patients ) . Healthy Individuals from the same endemic areas were included in this study as controls , composing the Group 4 ( n = 10 ) . All of them presented negative serologic tests for Chagas' disease and were matched by age and gender with the Chagas' disease patients . Protein lysate of T . cruzi ( Y strain ) obtained from LLMCK2 fibroblast cell line was used as the source of antigens . Briefly , the parasites were harvested , washed and submitted to 6 freeze/thaw cycles in liquid nitrogen and 37°C . The lysate was centrifuged at 12 , 000 g , the supernatant collected and the protein concentration determined . Peripheral blood was harvested with heparin ( 50 U/mL ) from healthy individuals and Chagas' disease patients . PBMC were isolated using Ficoll-Hypaque ( Pharmacia Biotech ) density gradient centrifugation , washed , counted , and used for CD4+CD25+ T cell isolation or cultured with specific antigen . PBMC ( 5×106 cells/mL ) were cultured for 48 h with T . cruzi antigen ( 10 µg/mL ) and phytohaemagglutinin ( PHA ) ( 1 µg/mL ) ( Sigma-Aldrich , St . Louis ) in 48 wells plates ( final volume of 0 . 5 mL ) and labeled with specific antibodies for phenotypic analysis in flow cytometer and determination of cytokine production in the supernatant of PBMC . As the concentration of IL-17 peaked at 48 h culture , we choose this time point for supernatant collection and cytokine assay . The cultured PBMC were washed in cold phosphate buffered saline ( PBS ) and samples of 5×105 cells/tube incubated for 30 min at 4°C with PBS-5% rabbit normal serum to block unspecific bidding , followed by the addition of 0 . 5 µg of phycoerythrin ( PE ) , allophycocyanin ( APC ) or fluorescein isothiocyanate ( FITC ) -labeled antibodies anti-CD3 , anti-CD4 , anti-CD25 , anti-GITR , anti-CTLA-4 and anti-CD103 ( all from BD-Pharmingen ) for additional 30 minutes at 4°C in the dark . To detect the intracellular expression of Foxp3 the cells were fixed with cytofix/cytoperm solution ( BD Biosciences ) for 15 min at room temperature ( RT ) , washed and stained with anti-Foxp3 or anti-IL17 peridinin chlorophyll protein ( PERCP ) -labeled , for 30 min at 4°C in the dark . Subsequently , the cells were washed twice and suspended in 100 µL of PBS-1% formaldehyde . In the assays involving intracellular detection of IL-17 , the cells were incubated for additional 6 h in the presence of GolgiStop solution , according manufacturer's recommendations ( BD Biosciences ) and then treated as described above . Data acquisition was performed using a FACSCanto II ( BD ) and the multivariate data analysis performed with the FlowJo software ( Treestar , USA ) , after collecting 50 , 000 events/sample . Distinct gating strategies were used to analyze the regulatory T cell and IL-17-producing CD4 T cell . Characterization of Treg started with gating the lymphocytes on FSC versus SSC dot plot . The T-lymphocyte subpopulations were further selected on FL1 ? anti-CD4 versus FL2 ? anti-CD25 dot plots . The percentage of cells expressing CTLA-4 , CD-103 , GITR and Foxp3 were analyzed in CD4 T cells , considering three different gates , according to the level of expression ( or not ) of CD25 . The percentage of cells expressing intracellular IL-17 was analyzed within the gate of CD3+CD4+ population . Cytokine production was assayed in supernatant culture of PBMC stimulated or not with T . cruzi antigen . ELISA sets were IL-10 , IL-17 , IFN-γ and TNF-α ( R&D , Minneapolis , MN ) , and procedures were undertaken according to manufacturers' instructions . Optical densities were measured at 450 ηm . Results are expressed as picograms per milliliter . To verify the regulatory function of CD4+CD25+ T cells isolated from PBMC of moderate/severe cardiomyopathy or free/mild cardiomyopathy patients , they were cultured with PBMC ( 2×105/well ) from normal donors , at ratio 1∶5 and 1∶10 , in 96-well U-bottom plates , in presence of PHA ( 1 µg/mL ) , at 37°C and 5% CO2 . CFSE ( Molecular Probes ) was added at a final concentration of 1 . 25 µM . The solution was well mixed and incubated at RT for 5 min . An equal volume of serum was used to quench the reaction , after which , the cells were washed with PBS with 5% serum . On day 3 of culture , lymphocytes were collected , washed twice and suspended in 100 µL of PBS-1% formaldehyde . Data acquisition was performed using a FACSCanto II and the multivariate data analysis was performed in the FlowJo software . The data expressed as percentage of inhibition were calculated based on the PHA-induced proliferation of allogeneic T cells cultured without CD4+CD25+ T cells . Statistical analysis was performed using Mann-Whitney or Kruskal–Wallis tests , performed for the comparison of two or three variables between groups ( INSTAT Software; GraphPad ) . The association between IFN-γ levels , regulatory T cell activity and left ventricular ejection fraction were tested by using the Spearman correlation ( INSTAT Software; GraphPad ) . All values were considered significantly different at P<0 . 05 .
We first aimed to study the ability of cells from patients with different forms of the disease to produce IL-17 , IL-10 , IFN-γ and TNF-α after T . cruzi antigen stimuli . Similar levels of IL-17 were observed in all groups ( Figure 1A ) . In contrast , cells from free/mild cardiomyopathy patients produced higher amounts of IL-10 than cells from moderate/severe cardiomyopathy patients group ( Figure 1B ) . In addition , the response to T . cruzi antigen regarding the production of TNF-α and IFN-γ was higher in patients with moderate/severe cardiomyopathy ( Figure 1C , 1D ) . The production of IL-17 by CD4+ T cells in PBMC from patients belonging to each experimental group , after being cultured with T . cruzi antigen obtained from trypomastigotes forms was also assessed using flow cytometry analysis . CD3+CD4+IL-17+ T cells from free/mild cardiomyopathy patients ( 3 . 73% ) displayed increased frequency when compared to healthy individuals ( 0 . 99% ) . Conversely , moderate/severe cardiomyopathy patients ( 1 . 23% ) , Bz-treated patients ( 1 . 84% ) and healthy individuals had similar frequency of these cells ( representative dot plots are shown in Figure 2A ) . No significant differences were found in the intensities of IL-17 expression ( MIF ) in CD3+CD4+ T cells among the groups of Chagas' disease patients . When we analyzed the data obtained with the patients of all groups , we found that the percentage of CD4+T cells expressing IL-17 were expressively increased in the cardiomyopathy-free/mild group of patients ( 1 . 74±0 . 92 ) compared with all the other groups . The mean of the percentage of CD4+T cells expressing IL-17 in moderate/severe cardiomyopathy patients , Bz-treated patients and healthy individuals were 0 . 99±0 . 75 , 0 . 90±0 . 58 and 0 . 67±0 . 57 , respectively ( Figure 2B ) . These findings were confirmed on confocal examination of PBMC . To characterize Treg population , CD4 versus CD25 dot plots were done and CD25+ lymphocytes classified in low and high or CD25− T cells ( as in Figure 3A ) . No significant differences in the frequencies of CD4+CD25high , CD4+CD25low and CD4+CD25− T cells were found among patients presenting different clinical forms of the disease as well as in controls ( P = 0 . 118 comparing healthy vs . free/mild cardiomyopathy; P = 0 . 893 , healthy vs . moderate/severe; P = 0 . 438 , healthy vs . treated; P = 0 . 109 , free/mild vs . moderate/severe cardiomyopathy; P = 0 . 247 , free/mild vs . treated; P = 0 . 494 , moderate/severe cardiomyopathy vs . treated ) ( Figure 3B , C and D ) . These results suggest that assessing the percentage of CD4+CD25+ could not be a reliable immunological approach to predict the different clinical forms of Chagas' disease . We next determined the frequency of cell that co-express CD103 , GITR , CTLA-4 , and Foxp3 on CD4+ T cell expressing high , low or absence of CD25 . Free/mild cardiomyopathy patients presented higher frequency of CD4+CD25high T cells expressing Foxp3 ( P = 0 . 033 ) and CTLA-4 ( P = 0 . 042 ) than moderate/severe cardiomyopathy patients ( Figure 3E ) . High percentage of CD4+CD25+Low T cells expressing Foxp3 ( P = 0 . 016 ) and CTLA-4 ( P = 0 . 046 ) were also observed in free/mild cardiomyopathy patients compared with moderate severe cardiomyopathy patients ( Figure 3F ) . Moreover , severe/moderate cardiomyopathy patients showed lower frequency of CD4+CD25− T cells expressing CTLA-4 ( P = 0 . 035 ) than free/mild cardiomyopathy patients , and Bz treated Chagas' disease patients ( Figure 3G ) . The mean intensity of fluorescence ( MIF ) of CTLA-4 , CD103 , GITR and Foxp3 was similar in all groups studied . Interestingly , the expression of CTLA-4 , but not CD103 , GITR and Foxp3 , in CD4+CD25high T cells was decreased in moderate/severe cardiomyopathy compared with free/mild cardiomyopathy patients and health individuals ( Figure 4A ) . These data show that CTLA-4 expression and frequency of CTL-4+ T cells correlates with less severe cardiac disease . Moreover , it may indicate that treatment with benznidazol , with the consequent parasite elimination , may have important implications in the cardiac disease progression . We next aimed to study if the reduced frequency of CD4+CD25+ T cell expressing CTLA-4 and Foxp3 that we found in severe cardiomyopathy patients correlated with deficient regulatory activities . CD4+CD25+ T cells from healthy individuals , free/mild cardiomyopathy patients and severe cardiomyopathy patients were sorted , and suppressive activity was evaluated in vitro through co-culture assay with allogeneic T cells stimulated with PHA . The purity of CD4+CD25+ T cells isolated from free/mild cardiomyopathy patients and severe cardiomyopathy patients were about 99% . Interestingly , the inhibitory activity of CD4+CD25+ T cells from healthy individuals ( 62 . 95±5 . 37 ) ( P = 0 . 0159 ) and free/mild cardiomyopathy patients ( 57 . 40±9 . 18 ) ( P = 0 . 0189 ) were significantly higher than that observed with CD4+CD25+ T cells from moderate/severe cardiomyopathy patients ( 33 . 76±4 . 67 ) , when cultured at a ratio of 1∶5 Treg∶allogeneic T cell ( Figure 4B ) . Of note , no differences were observed among the groups when the ratio of Treg∶effector was 1∶10 , possible due to a dilution effect in suppressive activity of these cells ( Figure 4C ) . The impairment in suppressive activity observed in CD4+CD25+ T cells from patients suffering from severe cardiomyopathy correlates with the observation of reduced amounts of CD4+CD25+ T cells expressing CTLA-4 and Foxp3 in this group of patients . We next correlated LVEF with the levels of IFN-γ in the sera of all Chagas' disease patients and Treg suppressive activity obtained after allogeneic cultures ( as described ) . Our results showed that IFN-γ levels during chronic Chagas' disease are inversely correlated to the LVEF ( P = 0 . 040 , r = −0 . 594 ) ( Figure 5A ) . Accordingly , the levels of regulatory T cell activity are directly correlated with LVEF ( P = 0 . 022 , r = 0 . 500 ) ( Figure 5B ) . We thus hypothesized that patients with chronic Chagas' disease undergoing cardiomyopathy produce increased levels of IL-17 and have a reduced frequency or suppressive activity of Treg compared with those patients with the indeterminate form of the disease . To our surprise , however , we found a positive correlation between frequency of CD4+IL-17+ T cell and CD4+CD25+HighFoxp3+ ( P = 0 . 042 , r = 0 . 418 ) ( Figure 5C ) . In addition , no significant correlation was observed between TNF-α ( P = 0 . 159 , r = 0 . 133 ) , IL-10 ( P = 0 . 265 , r = 0 . 066 ) production and LVEF .
In this investigation we first evaluated the production of IFN-γ , TNF-α , IL-10 and IL-17 in PBMC obtained from groups of Chagas' disease patients and in a group of benznidazol-treated individuals . The cultures of PBMC from patients with moderate/severe cardiomyopathy produced higher IFN-γ and TNF-α , and lower IL-10 levels than those observed in PBMC culture from free/mild cardiomyopathy patients , which is in accordance with previous reports by other researchers [1] , [7] , [9] . An imbalance in the production of cytokines IFN-γ and IL-10 was also observed in the present study , assaying these cytokines in the sera from chronic cardiac Chagas' disease patients: This imbalance has been implicated in the pathogenesis of Chagas heart disease [1] , [7] , [34] . Production of more IFN-γ and less IL-10 in cardiac patients supposedly results in efficient control of parasites replication but with more lesions to myocardium [5] . In addition , the analysis of IFN-γ production by ELISPOT of CD8 T cells from Chagas' disease patients showed that the frequency of IFN-γ producing-CD8 T cells is very low among those patients suffering the most severe form of the disease , and among individuals living in areas of active transmission of the disease , indicating that severe Chagas' cardiomyopathy could be related with the frequency of IFN-γ – producing T cells [31] , [35] . On the other hand , one study comparing the levels of mRNA expression for the cytokines IL-5 , IL-10 , IL-13 and IFN-γ in PBMC from healthy individuals , and patients with cardiomyopathy or indeterminate forms of Chagas disease , found no differences among these groups [11] . Hence , there is not a consensus regarding the exact participation of classic Th1 cytokine profiles in the mechanisms that lead to the cardiac lesions during Chagas' disease . It is therefore possible that other cytokine and cellular profiles participate in the immunological imbalance observed during Chagas' disease . One candidate is IL-17 which has effectively been involved in the control of parasites and in the induction of myocarditis in T . cruzi experimental infection [24] . In the present study PBMC from free/mild cardiomyopathy patients exhibited a higher expression of IL-17 in CD4+ T cells than that observed in PBMC from patients with severe/moderate cardiomyopathy and in cells from healthy individuals . Likewise , in the experimental model the inhibition of IL-17 resulted in enhanced production of IFN-γ and increased cardiac inflammation [24] . Moreover , impaired activation of immune-related cells that are critical for the killing of T . cruzi is observed in the absence of IL-17A gene [25] . Our data confirmed that PBMC from the group of moderate/severe cardiomyopathy patients produce more IFN-γ and TNF-α and less IL-10 than the cells obtained from the other groups . Cells from the same group of patients expressed more IL-17 when cultured with parasite antigens . In the same way , the infection with the trypanosomatid Leishmania donovani , the etiological agent of Kala Azar ( KA ) , stimulates the differentiation into Th17 cells in PBMC obtained from healthy donors , leading to IL-17 and IFN-γ production [36] . As a result , IL-17 should be important in the control of cardiac inflammation by playing a negative feedback role on the production of IFN-γ and chemokines during T . cruzi infection in humans and mice , modulating the cardiac immune-mediated lesions found in Chagas' disease patients . Here we showed that the production of IL-17 is increased in patients without or with mild cardiac manifestations of the disease , which together with the results showing efficient suppressive activity of Treg in the same group of patients , suggest that IL-17 may be involved in the control of the immune response and , therefore , in the modulation of cardiac disease progression . As pointed before , IL-17 is also crucial for the control of parasite growth and host survival [24] , [25] . These data are in agreement with that from a study on a cohort of subjects during a severe outbreak of the infection by the trypanosomatid L . donovani , in which the analysis of Th1 , Th2 , and Th17 cytokine responses by cultured PBMCs from revealed that IL-17 is associated with protection against severe KA [32] . The frequency of CD4+CD25+ regulatory T cells among patients with different clinical forms of Chagas disease was also examined in the present study . Surprisingly , all groups of patients showed a similar frequency of CD4+CD25+ T cell and CD4+CD25high T cells . This is not in agreement with a previous report showing lower values of CD4+CD25high T cells among school children with the indeterminate form of Chagas disease than that values observed in healthy children [17] . However , the same authors reported later in a study that patients with the indeterminate form of Chagas' disease exhibited a higher frequency of CD4+CD25high T cell expressing Foxp3 and IL-10 as compared to those individuals with cardiomyopathy [37] . The last study was confirmed by a recent report showing that asymptomatic patients had increased amounts of Treg than those with cardiomyopathy [38] . Thus , it is possible to assume that a low frequency of regulatory T cell during early stages of Chagas' heart disease might be associated with the development of more serious chronic manifestations of Chagas' heart disease . As pointed out before , our study does not confirm these data probably due to our very well characterized groups of patients . All patients underwent a detailed clinical evaluation , 12-lead rest electrocardiogram ( EKG ) , chest X-ray and a 2D-echocardiogram . However , in the experimental model of Chagas' disease the inhibition of Treg function with anti-GITR markedly increased the parasitemia , myocarditis and mortality compared with control mice [39] . As we did not detect differences in the percentages of Treg between the groups of patients in the present study , we investigated the suppressive activity of these cells . First , we assayed the expression of surface markers ( CD103 , CTLA-4 and GITR ) as well as the transcriptional factor Foxp3 . A higher expression of CTLA-4 and Foxp3 in the CD4+CD25high T cells from free/mild cardiomyopathy patients was observed , when compared to moderate/severe cardiomyopathy patients . Moreover , the analysis of CD4+CD25Low cells population demonstrated that free/mild cardiomyopathy patients and Bz treated patients have a higher occurrence of CTLA-4 than moderate/severe cardiomyopathy patients . The high amount of CD4+CD25+ T cells expressing CTLA-4 and Foxp3 ( that activate the regulatory T cell machinery ) in CD4+ CD25+ T cells of free/mild cardiomyopathy and Bz treated patients may be related to high suppressor activity of these cells . Furthermore , the greater number of CD4+CD25− T cells of free/mild cardiomyopathy patients and Bz treated patients expressing CTLA-4 than cells from moderate/severe cardiomyopathy patients , probably contributes to the modulation of immune response in the heart . A higher incidence of T cells expressing CTLA-4 among CD4+CD25− T cells from free/mild cardiomyopathy patients when compared to moderate/severe cardiomyopathy patients , also suggest a better negative control of the immune response , since CTLA-4 expression in CD25− T cells is known to suppress the immune response [40] . The suppressive activity of Treg in PBMC from all groups of patients herein described was examined based in their capacity to suppress T cell proliferation . As we suspected , CD4+CD25+ T cells from Chagas' disease patients with severe cardiomyopathy presented reduced capacity to suppress T cell proliferation when compared to free/mild cardiomyopathy patients and healthy individuals . This phenomenon may be correlated by low frequency of CTLA-4 in the CD4+CD25− T cells from cardiac patients . Nevertheless , it was previously reported that cardiomyopathy patients exhibit a higher percentage of CD4+CD25high T cells expressing CTLA-4 [33] . The mechanism leading to reduced expression of Foxp3 and CTLA-4 and consequently , deficient suppressive activity of CD4+CD25+ T cells from patients with cardiomyopathy has not been elucidated . It is possible that a defective control of the immune response by Treg/Th17 may contribute to the pathogenesis of Chagas' heart disease , in a similar way as patients with other inflammatory and autoimmune diseases such as multiple sclerosis , systemic lupus erythematous , type 1 diabetes , psoriasis and rheumatoid arthritis have compromised functional activity of Treg [41] , [42] . We also analyzed the levels of cytokines produced by PBMC from patients after in vitro stimulation with T . cruzi antigens , and we showed that IFN-γ production during chronic Chagas' disease is inversely correlated to LVEF , while normal regulatory T cell activity directly correlates with it . In addition , TNF-α production levels were lower in free/mild cardiomyopathy patients than in patients with moderate/severe cardiomyopathy . This finding is in agreement with a previous study reporting that patients with significant left ventricular ( LV ) dysfunction ( LV ejection fraction ≤50% ) showed higher levels of TNF-α , compared to Chagas' disease patients without LV dysfunction [43] . Moreover , studies in patients with dilated cardiomyopathy reported a significant increase of TNF-α among these individuals when compared with healthy controls , suggesting that the elevation of TNF-α could be an immune pathogenic mechanism in the progression to cardiomyopathy . Here we showed that the production of TNF-α ( and not IFN-γ ) tends to be lower among benznidazole-treated individuals . Although further research are required to explore the mechanisms by which benznidazole can induce these differential effects on cytokines production , these findings has been experimentally addressed before , and coincide with our current results . For example , it was shown that IFN-γ mediates the protective effect of benznidazole against T . cruzi infection [44] , and slightly inhibits the synthesis of TNF-α in murine cells [45] . The levels of this cytokine may also constitute an important marker of ventricular dysfunction in chronic chagasic cardiomyopathy [46] , [47] . One import result found in the present study was a positive correlation between IL-17 and Foxp3 expression in PBMC among Chagas' disease patients . Therefore , more IL-17 and Foxp3 expression is preferentially found in free/mild cardiomyopathy patients . Thus , the expanded Treg are better able to control the inflammatory response in presence of Th17 . This data are in agreement with the previous demonstration that Th17 are preferentially differentiated in the presence of Treg [28] due the consumption of IL-2 by Treg [26] , [27] Chronic autoimmune inflammation originates when this process is deregulated , and then therapeutic intervention becomes necessary to restore that balance between Th17 and Treg . It is clear that genetic characteristics of both the host and the parasite are important in determining the outcome of the infection . Our data suggest that genetic aspects of the immune response involved in the functions of Treg , IL-17 , and some related genes may deserve further investigation and may shed light on the comprehension of the immune pathogenesis of Chagas' disease . In summary , our results show that CD4+CD25+ Treg from patients with severe cardiomyopathy display a deficient suppressive activity , leading to uncontrolled production of pro-inflammatory cytokines ( TNF-α and IFN-γ ) from leukocytes . Moreover , patients with less aggressive forms of the disease ( cardiomyopathy free or mild cardiomyopathy individuals ) produce higher levels of the cytokines IL-10 and IL-17 . Reduced CD4+CD25+ regulatory T cell function and low levels of IL-17 also correlated with more advanced cardiomyopathy . We think that these findings may be helpful in the design of immunotherapeutic approaches for eventual primary , secondary and tertiary prevention of chronic Chagas' cardiomyopathy . | Dilated cardiomyopathy is one of the clinical forms of Chagas' disease ( CD ) after the infection caused by the parasite Trypanosoma cruzi . Even though strategies adopted in most Latin-American countries in the last decades towards vector control have been effective in reducing the incidence of CD , active transmission is maintained in some regions , and secondary prevention approaches are still required for the infected patients , mostly because the specific anti-parasitic medications are toxic and perhaps of limited efficacy in chronically infected individuals . Moreover , there are no markers to predict the risk of developing dilated cardiomyopathy in asymptomatic , chronically infected patients , although the failure in the mechanisms that control the immune response can be involved in the development of Chagas' heart disease . In this study we show that preserved activity of regulatory T cells and the production of the cytokine IL-17 are connected with a more benign evolution of the disease , which brings a new understanding on the mechanisms associated with progression of CD . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"cardiomyopathies",
"immune",
"cells",
"cytokines",
"immunity",
"to",
"infections",
"immunology",
"parasitic",
"diseases",
"cardiovascular",
"neglected",
"tropical",
"diseases",
"immunoregulation",
"infectious",
"diseases",
"t",
"cells",
"immune",
"response",
"immune",
"system",
"chagas",
"disease",
"clinical",
"immunology",
"immunity"
] | 2012 | Deficient Regulatory T Cell Activity and Low Frequency of IL-17-Producing T Cells Correlate with the Extent of Cardiomyopathy in Human Chagas' Disease |
DNA replication is fundamental for life , yet a detailed understanding of bacterial DNA replication is limited outside the organisms Escherichia coli and Bacillus subtilis . Many bacteria , including mycobacteria , encode no identified homologs of helicase loaders or regulators of the initiator protein DnaA , despite these factors being essential for DNA replication in E . coli and B . subtilis . In this study we discover that a previously uncharacterized protein , Rv0004 , from the human pathogen Mycobacterium tuberculosis is essential for bacterial viability and that depletion of Rv0004 leads to a block in cell cycle progression . Using a combination of genetic and biochemical approaches , we found that Rv0004 has a role in DNA replication , interacts with DNA and the replicative helicase DnaB , and affects DnaB-DnaA complex formation . We also identify a conserved domain in Rv0004 that is predicted to structurally resemble the N-terminal protein-protein interaction domain of DnaA . Mutation of a single conserved tryptophan within Rv0004’s DnaA N-terminal-like domain leads to phenotypes similar to those observed upon Rv0004 depletion and can affect the association of Rv0004 with DnaB . In addition , using live cell imaging during depletion of Rv0004 , we have uncovered a previously unappreciated role for DNA replication in coordinating mycobacterial cell division and cell size . Together , our data support that Rv0004 encodes a homolog of the recently identified DciA family of proteins found in most bacteria that lack the DnaC-DnaI helicase loaders in E . coli and B . subtilis . Therefore , the mechanisms of Rv0004 elucidated here likely apply to other DciA homologs and reveal insight into the diversity of bacterial strategies in even the most conserved biological processes .
The ability to maintain , replicate , and express genetic information encoded in DNA is critical to all domains of life . DNA replication studies in Escherichia coli and Bacillus subtilis have elucidated the mechanisms of bacterial DNA replication initiation , elongation , and termination , but the applicability of many of these findings to other bacteria is less clear . Briefly , initiation begins when DnaA , the initiator protein , binds to specific sites located at the origin of replication ( oriC ) and oligomerizes , forming a nucleoprotein complex that results in the melting of the adjacent DNA [1] . Next , helicase loaders and accessory primosomal proteins , with the help of DnaA , load the replicative helicase onto melted DNA [2 , 3] . The replicative helicase then binds the primase , which lays down short RNA primers [2] . Clamp loader complexes load DNA Polymerase III ( Pol III ) onto primed DNA , allowing replication elongation to begin [1] . Elongation proceeds bi-directionally from oriC until it reaches termination sites bound by terminator proteins [4] . Although the general stages of DNA replication are likely conserved in all bacteria , many steps have not been studied outside of the model organisms E . coli and B . subtilis . In particular , DNA replication is not well understood in mycobacteria , including the human pathogen Mycobacterium tuberculosis ( Mtb ) . While high-fidelity DNA replication and repair is critical to maintain chromosomal integrity , mutations generated by error-prone DNA replication can enhance Mtb virulence and lead to antibiotic resistance [5] . The study of DNA replication and repair in mycobacteria is particularly relevant given that all acquired drug resistance in Mtb arises through chromosomally encoded mutations [6 , 7] . Mycobacteria encode homologs of some , but not all , DNA replication proteins and it is not clear how most mycobacterial homologs function relative to their E . coli counterparts . For example , Rock et al . recently showed that the DNA Pol III ε exonuclease , which is essential for replication fidelity in E . coli , is dispensable in Mtb [8] . There are also a number of processes essential for efficient DNA replication in E . coli and B . subtilis for which homologs have not been identified in mycobacteria , including regulators of DnaA activity ( Hda in E . coli , YabA in B . subtilis [9] ) , proteins that load the replicative helicase ( DnaC in E . coli , DnaI in B . subtilis [2 , 10] ) , and replication terminator proteins ( Tus in E . coli and RTP in B . subtilis [11] ) . The proteins required for these processes in E . coli and B . subtilis are functionally analogous but are not conserved in sequence . Therefore , functionally similar mycobacterial proteins could exist and remain unidentified due to sequence divergence . In this study we have discovered that Rv0004 in Mtb and MSMEG_0004 in Mycobacterium smegmatis are essential for DNA replication even though they are absent from E . coli and B . subtilis , the organisms traditionally used to study bacterial DNA replication . Rv0004 had never before been studied but is predicted to contain a domain of unknown function 721 ( DUF721 , PF05258 ) . In a recent publication , Brezellec et al . used bioinformatics to identify DUF721-containing proteins in 23 out of 26 bacterial phyla and named the members of this protein family “DciA , ” for DnaC DnaI antecedent , based on the finding that these proteins preceded the DnaC-DnaI helicase loading systems of E . coli and B . subtilis [12] . While Brezellec et al . illustrate how widely distributed DciA homologs are , which underscores the importance of work on DciA proteins to the field of bacterial DNA replication , the experimental data is limited to showing that Pseudomonas aeruginosa DciA is important for DNA replication in P . aeruginosa and associates with DnaB in a bacterial two-hybrid assay [12] . Therefore , a molecular and biochemical analysis of DciA has yet to be performed and is necessary to define the basis of DciA’s interaction with DnaB outside of the bacterial two-hybrid system , to determine how DciA’s association with DnaB relates to its role in DNA replication , and to elucidate other activities for DciA in the cell . In this manuscript we perform the first mechanistic studies on a DciA homolog and show that Mtb Rv0004 ( DciAMtb ) directly binds DNA and the replicative helicase DnaB to regulate the interaction of DnaB with the initiator protein DnaA . We discover that the DUF721 in DciA proteins comprises a protein-protein interaction domain that is predicted to structurally resemble the N-terminus of DnaA . We provide data to support the importance of this domain by showing that the mutation of a single conserved tryptophan within DciAMtb’s DnaA N-terminal-like domain leads to defects in DciAMtb’s cellular activity and can affect the association of DciAMtb with DnaB . In addition , using live cell imaging during depletion of dciAMtb we have uncovered a previously unappreciated role for DNA replication in the coordination of mycobacterial cell division and cell size . Together , these studies elucidate a mechanism by which DciA proteins affect DNA replication initiation , identify a function for a conserved protein domain , and provide insight into the influence of DNA replication on cell cycle in mycobacteria .
In previous work we probed the transcriptional responses of M . smegmatis , a non-pathogenic model organism for Mtb , to double-stranded DNA ( dsDNA ) breaks [13] . We identified MSMEG_0004 as being upregulated in response to dsDNA breaks that occurred during logarithmic ( log ) but not stationary phase . By measuring the expression of MSMEG_0004 during log versus stationary growth phase in the absence of induced DNA damage , we found that MSMEG_0004 is also more highly expressed in log phase in the absence of stress ( Fig 1A ) . We observed a similar result for the Mtb homolog Rv0004 ( Fig 1B ) . Together , these data suggest that expression of 0004 genes is important while the bacteria are actively growing and dividing . MSMEG_0004 homologs , which are not present in eukaryotes , E . coli , or B . subtilis , encode hypothetical proteins that contain a domain of unknown function ( DUF721; PFAM05258 , Fig 1C ) and are predicted to be nucleic acid-binding proteins ( COG5512 family members ) [14] . Brezellec et al . recently identified DUF721 as being widely conserved in proteins they termed DciA [12] . Based on the presence of DUF721 in Rv0004 and MSMEG_0004 , we will refer to Rv0004 as dciAMtb and MSMEG_0004 as dciAMsm . dciAMsm is located in an operon next to oriC that also contains dnaN and recF , which encode the DNA Pol III beta clamp and a DNA repair protein , respectively ( Fig 1D ) [15 , 16] . This operon is located between dnaA and the gyrB-gyrA operon , which encode the replication initiator protein and bacterial gyrase , respectively . This genome structure is conserved between M . smegmatis and Mtb except for MSMEG_0002 , a gene that is predicted to encode a 6-phosphogluconate dehydrogenase and is encoded at a separate genomic location in Mtb ( Fig 1D ) [17] . The orientation of the DNA replication and repair genes dnaA , dnaN , recF , gyrB , gyrA near oriC is a conserved feature in many bacteria [18] . Like dciAMsm , dnaA , dnaN , and recF were all more highly expressed in log phase versus stationary phase , consistent with roles in DNA replication ( S1 Fig ) . The genomic location and the increased expression of dciAMtb and dciAMsm during log phase support a link between the mycobacterial DciA proteins and DNA replication or repair . In order to study the roles for dciAMsm and dciAMtb in mycobacteria , we took a reverse genetic approach . Attempts to delete dciAMsm from M . smegmatis and dciAMtb from Mtb were unsuccessful , suggesting these genes are essential for viability . To study the dciA genes , we constructed a merodiploid strain in which the dciAMtb gene was integrated at the M . smegmatis attB site under the control of a promoter that contains tet operator sites and is linked to a kanamycin resistance cassette ( Fig 2A , S1 and S2 Tables ) . We then deleted dciAMsm from its endogenous locus ( Fig 2B ) , resulting in the strain ΔdciAMsm attB::tetdciAMtb ( Fig 2A ) . We used the ΔdciAMsm attB::tetdciAMtb strain to study dciAMtb in the context of M . smegmatis . Unfortunately , we were unable to engineer similar strains in Mtb , likely due to the difficulty of manipulating the genome near oriC . To confirm that dciA is essential in M . smegmatis , we used a gene-switching technique [19] to replace the kanamycin resistance cassette-linked dciAMtb allele at the attB site in the ΔdciAMsm attB::tetdciAMtb strain with a zeocin-resistant plasmid that was either empty or expressed dciAMtb . We were able to switch in the zeocin-resistant plasmid expressing dciAMtb but we were unable to recover a dciA gene null mutant ( Fig 2C ) , further supporting that dciAMsm is essential for viability . To study the loss of dciA expression in M . smegmatis , the ΔdciAMsm attB::tetdciAMtb strain was transformed with an episomal plasmid expressing a Tet-ON repressor ( TetR ) [20] ( S1 and S2 Tables ) . In this M . smegmatis ΔdciAMsm attB::tetdciAMtb +pTetR strain , referred to as Tet-DciA , dciAMtb transcript is only expressed in the presence of anhydrotetracycline ( ATc ) . We monitored dciAMtb depletion by diluting cultures of Tet-DciA grown in the presence of ATc ( +ATc ) into liquid media lacking ATc ( -ATc ) and collecting RNA at 3 , 8 , 16 , 24 , 28 , and 32 hours after growth -ATc . After 16 hours in depleting conditions , dciAMtb transcript levels were 96% lower than at 3 hours ( Fig 2D and 2E ) . To characterize the requirement of dciAMtb expression for growth , we monitored the growth of the Tet-DciA strain during dciAMtb transcript depletion . Over the first 24 hours , there was no significant difference in the growth of Tet-DciA in depleting ( -ATc ) or replete ( +ATc ) conditions ( Fig 2F ) . However , after diluting the stationary phase culture back to early log phase at 24 hours , depleted cells grew more slowly than controls ( Fig 2F ) . We calculated the doubling times of Tet-DciA cultured ± ATc and found that after 24 hours , M . smegmatis depleted of dciAMtb grew significantly more slowly than the replete controls ( 5 . 8 hour versus 3 . 7 hour doubling time ) ( Fig 2G ) . To determine whether the growth defect of dciAMtb-depleted cells was due to an inability to recover from stationary phase versus DciAMtb playing a critical role in growth during log phase , we performed a continual log liquid growth curve where cultures were diluted to early log phase every 8 hours to ensure that they did not enter stationary phase . Similar to the standard growth curve experiments , M . smegmatis depleted of dciAMtb did not show a significant growth defect until after 24 hours of growth–ATc ( Fig 2H ) . We calculated the doubling times of Tet-DciA cultured ± ATc and found that after 24 hours of continual log growth , M . smegmatis depleted of dciAMtb grew significantly more slowly than the replete controls ( 4 . 7 hour versus 3 . 4 hour doubling time ) ( Fig 2I ) . After 40 hours of growth in depleting conditions , suppressors of the Tet-ON system are selected for and the levels of dciAMtb transcript are no longer controlled by ATc . Together , these data demonstrate that DciA is important for M . smegmatis growth . We have shown that dciA is essential for growth in culture and is involved in responding to DNA damage in M . smegmatis . However , a role in DNA damage responses alone cannot explain the essentiality of dciA , since genes key for mycobacterial DNA repair pathways are not essential in vitro [21–24] . The location of mycobacterial dciA genes in an operon with and adjacent to genes involved in DNA replication raises the question of whether DciA plays a role in this essential process in mycobacteria . In addition , P . aeruginosa DciA ( DciAPa ) has been implicated in DNA replication [12] . To investigate a role for mycobacterial DciA in DNA replication and the bacterial cell cycle , we monitored the cellular morphology of Tet-DciA cultured ± ATc in the continual log growth curve ( Fig 2H ) . By 24 hours of depletion , Tet-DciA cultured without ATc were on average 12 . 9% ( p <0 . 01 ) longer than Tet-DciA cultured in dciAMtb-replete conditions ( S2 Fig ) . After 36 hours , Tet-DciA cultured without ATc averaged over 60% longer ( p <0 . 0001 ) than Tet-DciA grown in dciAMtb-replete conditions ( Fig 3A , 3B and 3D ) . The elongated cellular morphology observed during dciAMtb depletion indicates a block in cell cycle progression . To characterize where in the cell cycle dciAMtb-depleted M . smegmatis is blocked , we analyzed nucleoid morphology and septum formation . To observe nucleoid morphology , Tet-DciA was grown ± ATc , DNA was stained with DAPI , and cells were visualized by fluorescent microscopy ( Fig 3A ) . The nucleoid in dciAMtb-replete Tet-DciA ( +ATc ) appears as several distinct puncta throughout the length of the cell ( Fig 3A , top row , second panel ) , as has been reported previously for M . smegmatis [25] . Following dciAMtb depletion , the DAPI-stained nucleoid still appeared as several distinct puncta , but was not distributed throughout the length of the cell . Instead , there were areas free of DNA staining found at the poles ( Fig 3A ) . We quantified the abnormal nucleoid morphology using “% DNA occupation , ” which represents the nucleoid length as a percentage of the total cell length . Tet-DciA depleted for dciAMtb exhibit significantly lower % DNA occupation starting at 16 hours of depletion ( Fig 3C and 3D; S2 Fig ) . The observation that depletion of dciAMtb leads to significantly lower % DNA occupation at 16 hours even though cell lengths are not significantly longer at this time point ( S2 Fig ) indicates that abnormal nucleoid morphology is the earlier phenotype , and that a DNA-related function causes the cell cycle block and slowed growth ( Fig 2I ) . We also observed the presence of 9% anucleate cells by DAPI staining upon dciAMtb depletion ( Fig 3A ) . Anucleate cells result when DNA replication and cellular division are uncoupled . The presence of anucleate cells also indicates that cell division is still able to occur . To confirm that septum formation was intact , we used transmission electron microscopy ( TEM ) and found that cells depleted for dciAMtb were still able to form normal septa ( Fig 3E ) . To visualize cell division of dciAMtb-depleted M . smegmatis in real time , we performed live-cell imaging [26] of Tet-DciA grown ± ATc in a constant-flow microfluidic device ( S1 and S2 Movies ) . Using FM4-64 membrane stain , we observed that dciAMtb-depleted cells can form septa and undergo cell division ( S2 Movie ) . We also detected increased cell death in dciAMtb-depleted cells compared to controls , where dead cells stop elongating and take up more FM4-64 dye , thus exhibiting a rapid increase in fluorescence ( S2 Movie ) . 9 . 4% of dciAMtb-depleted cells displayed these cell death characteristics . In support of our fixed fluorescent microscopy , Tet-DciA cells grown -ATc were on average 56% longer at birth and division than Tet-DciA grown +ATc ( S3 Fig , S1 and S2 Movies , p <0 . 0001 ) . However , despite the increased average length of dciAMtb-depleted cells , we also observed division of unusually small cells ( S3E Fig ) , many of which died shortly after division . In general , we observed greater variability in birth length in dciAMtb depleted versus replete cells , with coefficients of variation of 41 . 7% and 18 . 2% , respectively ( S3C Fig ) . The increased heterogeneity among cell birth lengths during dciAMtb depletion indicates a disruption in the coordination of septum formation with cell growth , leading to the dysregulation of cell size . Thus , time-lapse imaging demonstrates that in addition to dciAMtb-depleted cells being elongated on average , they are also characterized by increased variation in cell size and frequency of death . Together , our data show that cell division is intact but chromosome replication or segregation is blocked during dciAMtb depletion . In support of this conclusion , the abnormal nucleoid morphologies observed in M . smegmatis depleted of dciAMtb phenocopy those of M . smegmatis depleted of DnaA , the chromosomal replication initiator protein [25] ( Fig 4A–4D , S4 Fig ) , but not M . smegmatis depleted of FtsZ , the protein that comprises the Z-ring precursor of the septum [27] ( Fig 4E–4H , S5 Fig ) . The data so far support a model that DciA homologs are required for either DNA replication or chromosome segregation . To determine if dciAMtb-depleted M . smegmatis is defective in DNA replication , we directly measured rates of DNA synthesis using a nucleotide incorporation assay [25 , 28] . Specifically , we determined the rates of [5 , 6-3H]-thymidine incorporation into DNA by Tet-DciA cells grown ± ATc in continual log phase . We found that the rate of DNA synthesis in M . smegmatis was significantly lower at 24 hours ( Fig 5A and 5C ) and 36 hours ( Fig 5B and 5C ) in Tet-DciA grown -ATc relative to Tet-DciA grown +ATc , proving that the rate of DNA replication itself decreases upon dciAMtb depletion . To further confirm this defect in DNA replication , we stained Tet-DciA grown ± ATc with DAPI and measured DNA content using flow cytometry . dciAMtb-depleted cells had lower DNA content per cell based on the mean fluorescence intensity ( MFI ) of DAPI staining relative to controls ( Fig 5D and S6A Fig ) . Together , these data demonstrate that DciAMtb is involved in DNA replication , but do not differentiate between which step ( s ) of DNA replication DciAMtb acts . To determine how DciA proteins function in DNA replication , we investigated macromolecular interaction partners of DciAMtb . DciAMtb has a calculated isoelectric point ( pI ) around 12 , indicating that the protein is positively-charged at neutral pH . Other proteins in mycobacteria with high isoelectric points include histone-like proteins ( H-NS , HupB ) and integration host factor ( IHF ) , which are all known to bind DNA [29–32] . DciAMtb , as a member of COG5512 , is also predicted to be a nucleic-acid binding protein [14] . Binding to nucleic acid could be relevant to the role for DciA in DNA replication given the numerous protein-nucleic acid complexes that form during this process . We tested the DNA binding activity of purified DciAMtb protein ( S7A Fig ) in electromobility shift assays ( EMSAs ) with 32P-radiolabeled DNA . Due to its role in DNA replication ( Figs 3 and 5 ) , we first tested whether DciAMtb was able to bind oriC DNA in vitro . DciAMtb was able to bind and shift a 553 basepair ( bp ) dsDNA fragment containing Mtb oriC DNA [33] ( Fig 6A ) . Given its high isoelectric point and the negative charge of DNA , we hypothesized that DciAMtb would be able to bind any DNA sequence and not just oriC . Indeed , we found that DciAMtb was also able to shift a 333 bp dsDNA sequence from a site in the genome distantly located from oriC ( S7B Fig ) . DciAMtb’s DNA binding activity is not limited to dsDNA sequences as DciAMtb was also able to bind and shift a 72 nucleotide single-stranded DNA ( ssDNA ) oligo ( S7C Fig ) . These data indicate that DciAMtb can bind diverse double and single stranded DNA molecules in vitro , including oriC . This sequence independent DNA-binding activity could relate to the role of DciAMtb in DNA replication . We next identified the mycobacterial proteins that associate with DciAMtb by performing co-immunoprecipitation mass spectrometry ( co-IP/MS ) . We engineered a strain of M . smegmatis that encodes an HA-tagged version of dciAMtb ( HA-DciAMtb ) as its only dciA allele ( S1 Table ) . We generated cell lysate from this strain and immunoprecipitated HA-DciAMtb along with associated protein complexes using an anti-HA antibody conjugated agarose ( Sigma ) . After eluting with HA peptide , eluates were separated by SDS-PAGE , silver-stained , and bands specific to the HA-DciAMtb lane ( S7D Fig ) were isolated and analyzed by MS . We observed a similar banding pattern when we performed these experiments with DNAse-treated cell lysates ( S7E Fig ) . The most abundant band on the silver-stained gel contained ClpX ( S7D Fig ) , a component of the essential ClpXP protease . The association of DciAMtb with Clp protease may explain our inability to detect native DciA proteins by western blot . In addition to components of the Clp protease , we also found that DciAMtb associates with a number of proteins involved in DNA replication and repair ( Fig 6B , S4 Table ) . Since we have shown that DciAMtb is involved in DNA replication ( Fig 5 ) , we sought to confirm the association of DciAMtb with the co-immunoprecipitated DNA replication proteins . These proteins included gyrase , DnaX ( τ clamp-loader subunit of DNA Pol III ) , DnaB ( replicative helicase ) , and DnaA ( replication initiator protein ) . To prioritize our studies , we identified DNA replication proteins that are conserved in E . coli and B . subtilis but do not yet have known homologs in mycobacteria , namely DnaA regulators and DnaB helicase loaders ( Table 1 ) . Since DnaA and DnaB were both found to associate with DciAMtb through co-IP/MS , we first tested whether DciAMtb directly interacts with these proteins . We performed pull-down experiments similar to the co-IP approach described earlier , but using purified recombinant HA-DciAMtb , DnaB , and DnaA . We immobilized HA-DciAMtb onto anti-HA agarose and added DnaA or DnaB . Analysis of the protein complexes eluted with HA peptide showed that HA-DciAMtb pulls down both DnaA ( Fig 6C lane 4 ) and DnaB ( Fig 6C lane 6 ) , but not a negative control protein , RelMtb1-394 ( S7F Fig ) . Since HA-DciAMtb , DnaA , and DnaB can all bind DNA , we tested whether the interactions between these proteins were dependent on nucleic acid by performing the same pull-down experiments using recombinant proteins purified from Benzonase-treated lysates to degrade nucleic acids . When proteins purified from Benzonase-treated lysates were used , HA-DciAMtb failed to pull down DnaA ( Fig 6C lane 5 ) but retained its interaction with DnaB ( Fig 6C lane 7 ) . Therefore , DciAMtb interacts directly with DnaB but depends on nucleic acid to associate with DnaA . We confirmed that DciAMtb and DnaB directly interact by performing the reciprocal pull-down with purified FLAG-tagged DnaB ( DnaB-FLAG ) immobilized as bait and DciAMtb as prey ( Fig 6D ) . We performed Bio-Layer Interferometry ( BLI ) to quantify the affinity of the interaction between DnaB and DciAMtb . The association and dissociation of varying concentrations of DciAMtb to biotinylated DnaB loaded onto streptavidin-coated biosensor pins ( ForteBio ) were measured ( Fig 6E ) and an affinity constant ( KD ) of 210 . 9 ± 8 . 19 nM was calculated ( fit R2 = . 9018 ) ( S7G Fig ) . These data demonstrate that DnaB and DciAMtb interact in a dose-dependent manner . The calculated affinity constant in the nanomolar range suggests that the interaction could occur under physiological conditions , although cellular concentrations of DnaB and DciAMtb need to be quantified to confirm this . In E . coli , the interaction between DnaA and DnaB is required for efficient loading of DnaB [51 , 52] . The result that DciAMtb directly binds DnaB and can indirectly associate with DnaA led us to probe whether DciAMtb affects DnaB-DnaA complex formation . Mtb DnaA has been shown to interact with the N-terminus of DnaB ( residues 1–206 ) [36] , and we confirmed that the full length Mtb DnaB-FLAG and DnaA-HA proteins directly interact in our system ( Fig 6F lane 1 and Fig 6G lane 6 ) . To determine the effect of DciAMtb on the DnaA-DnaB interaction , we performed pull-downs with DnaA-HA as bait and DnaB-FLAG as prey in the presence of varying amounts of DciAMtb . All pull-downs were performed using proteins purified from Benzonase-treated lysates to exclude contributions from nucleic acid interactions . As increasing amounts of DciAMtb were added , DnaA-HA pulled down more DnaB-FLAG ( Fig 6F and 6H ) . These results indicate that DciAMtb facilitates DnaA-DnaB complex formation . In contrast , when DnaB-FLAG was immobilized as bait and DnaA-HA was added as prey , increasing the amount of DciAMtb present did not change the amount of DnaB-FLAG that associated with DnaA-HA ( Fig 6G and 6H ) . Therefore , DciAMtb can affect DnaB-DnaA complex formation , but this is dependent on which protein is immobilized . One possible explanation for this observation is that DnaB , which functions as a hexamer [34] , is unable to properly hexamerize while immobilized as bait , affecting DciAMtb activity . The ability of DciAMtb to positively affect the DnaB-DnaA interaction suggests that DciAMtb promotes rather than inhibits DNA replication , which is consistent with the observations that dciAMtb depletion leads to decreased DNA synthesis and DNA content ( Fig 5 ) . The ability of DciAMtb to affect the association of DnaB with DnaA suggests that DciA functions during DNA replication initiation . To test whether DciA is enriched at oriC where initiation occurs , we performed chromatin immunoprecipitation quantitative PCR ( ChIP-qPCR ) experiments . These experiments were performed with log-phase cultures of M . smegmatis strains that express HA-tagged DciA ( HA-DciA ) , untagged DciA ( no tag ) , and HA-tagged CarD ( HA-CarD ) , a mycobacterial transcription factor that associates with RNA Polymerase and is known to localize to every promoter throughout the M . smegmatis genome [53] . Protein-nucleic acid complexes were immunoprecipitated from each culture using anti-HA resin and co-immunoprecipitated DNA was probed for sequences specific for oriC ( S8A Fig ) , the rplN promoter , and internal to the sigA gene using qPCR . Enrichment of sequences within a given sample was determined relative to the no tag control . As expected , only DNA fragments containing promoters ( oriC and rplN promoter ) were specifically and significantly enriched for following immunoprecipitation with HA-CarD ( S8B Fig ) . The only DNA fragments that were significantly enriched for following immunoprecipitation of HA-DciAMtb were those containing the oriC ( Fig 6I ) . As a control , no sequences were enriched in input samples before immunoprecipitation ( Fig 6I and S8C Fig ) . The specific enrichment of DciAMtb at oriC and not at other areas of the chromosome indicates that DciAMtb is involved in DNA replication initiation , which is consistent with its role in affecting the interaction between DnaB and the DnaA initiator protein ( Fig 6F–6H ) . To investigate how DciAMtb facilitates the association between DnaA and DnaB , we used structural prediction tools to gain further insight into the protein architecture . Both Phyre2 [54] and I-TASSER [55] predicted that a region in the C-terminus of DciAMtb within DUF721 ( ~92–142 aa ) is structurally similar to the N-terminal domain ( NTD ) of DnaA in B . subtilis ( DnaABs , PDB: 4TPSD ) and E . coli ( DnaAEc , PDB: 2E0GA ) ( Fig 7A and 7B ) . We named this region of DciAMtb the DnaA NTD-Like ( DANL ) domain . The DnaA NTD is responsible for many protein-protein interactions important for DNA replication , including the interaction of DnaAEc with DnaB , DiaA , and other DnaAEc monomers , the interaction of DnaABs with SirA , and the interaction between H . pylori DnaA and HobA [51 , 52 , 56–58] . Phenylalanine 46 ( F46 ) in DnaAEc , which is equivalent to F49 in DnaABs , is specifically important for DnaAEc to load DnaB [57] . Though shifted by one residue , our structural alignment shows that a tryptophan at position 113 in DciAMtb ( W113 ) is the closest aromatic amino acid to F49 in DnaABs ( Fig 7A , 7C and 7D ) . Consurf alignment [59] reveals that the region around W113 is one of two highly conserved regions of the DciAMtb protein that are both located in DciAMtb’s C-terminus ( Fig 7E ) . The position of W113 in the predicted structural model and the conservation of this region across DciA homologs suggest that W113 may be important for DciAMtb activity . To test if W113 in the DciAMtb DANL domain is important for DciAMtb function , we engineered M . smegmatis to expresses a version of DciAMtb where the W113 is mutated to an alanine ( W113A ) as its only allele of dciA . The W113A mutation leads to a growth defect ( Fig 8A and 8B ) , elongated cellular and abnormal nucleoid morphologies ( Fig 8C–8F ) , and decreased DNA content ( Fig 8G; S6B Fig ) . The observation that the mutation of a single residue can cause similar phenotypes to those observed during dciAMtb depletion confirms that the dciAMtb-depletion phenotypes were not due solely to the depletion of an essential protein . These experiments also show that the W113 residue , which is located within the region of DciAMtb that is predicted to be structurally similar to the protein-protein interaction domain of DnaA , is important for DciAMtb activity . In order to determine how W113 is contributing to DciAMtb function , we purified DciAMtbW113A protein and tested its ability to perform the functions we have assigned to DciAMtb . DciAMtbW113A was able to bind and shift DNA ( S9A Fig ) at similar concentrations to DciAMtb wild-type protein ( Fig 6A ) . DciAMtbW113A was also able to bind DnaB in the presence ( S9B Fig ) and absence ( S9C Fig ) of nucleic acid , as well as affect DnaA-DnaB complex formation similarly to wild-type DciAMtb ( S9D–S9F Fig ) . The DNA and DnaB binding activities of DciAMtbW113A indicate that the DciAMtbW113A protein is functional , structurally intact , and not grossly misfolded . Given the location of W113 in the predicted DANL protein-protein interaction domain , we were surprised that the W113A mutation did not affect the interaction between DciAMtb and DnaB . To test whether the DANL domain itself was involved in the interaction with DnaB , we purified the HA-tagged DciAMtb N-terminus ( HA-DciAMtbΔ89–167 ) and HA-tagged DciAMtb C-terminus ( HA-DciAMtbΔ1–88 , containing the DANL domain ) protein truncations , and tested their ability to bind DnaB . We found that both the N-terminus ( HA-DciAMtbΔ89–167 ) and the C-terminus ( HA-DciAMtbΔ1–88 ) of DciAMtb can individually bind DnaB-FLAG ( Fig 8H ) . While we predicted the DANL-domain within the C-terminus of DciAMtb would bind DnaB based on its similarity to the DnaA NTD , we did not anticipate that the DciAMtb N-terminus would also interact with DnaB . This is particularly interesting since there are no known or predicted structures for the N-terminus of DciAMtb . We were unable to isolate viable M . smegmatis dciAMtbΔ89–167 or dciAMtbΔ1–88 mutants using the gene-switching approach described earlier , indicating that both halves of the protein contribute to DciAMtb’s essential cellular function . Therefore , the ability of both the N- and C-terminus of DciA to associate with DnaB is unlikely due to redundant roles for these two regions of the protein . We tested whether the DnaB-binding activity provided by the N-terminus of DciAMtb was precluding our ability to assess the contribution of W113 to the interaction with DnaB . To investigate this , we purified the HA-tagged DciAMtbW113A C-terminus ( HA-DciAMtbW113AΔ1–88 ) and tested its ability to bind DnaB-FLAG . We found that HA-DciAMtbW113AΔ1–88 associated with more DnaB-FLAG than the wild-type HA-DciAMtbΔ1–88 ( Fig 8I ) . Therefore , the W113A mutation in DciAMtb affects the association of the C-terminal DANL domain with DnaB and it is possible that the W113A M . smegmatis mutant strain could have defects due to altered DnaB binding by DciAMtbW113A .
DciA proteins were recently discovered in an evolutionary and phylogenetic analysis that defined them by the presence of the DUF721/PF05258 domain [12] . Two classes of dciA genes were identified , those located in the dnaN-recF operon and those located elsewhere . Using P . aeruginosa dciA ( dciAPa ) as an example of the second class of dciA genes , Brezellec et al . showed that DciAPa can associate with DnaB in a bacterial two-hybrid assay and that DNA replication is blocked during DciAPa depletion in P . aeruginosa [12] . However , the molecular details of how DciA associates with DnaB , the mechanisms by which DciA facilitates replication , whether DciA performs other activities in the cell , and whether the findings for DciAPa hold true for the other class of DciA proteins remained unknown . In this study we address these gaps in knowledge beginning with our discovery that DciAMtb , a member of the first class of DciA proteins , is an essential component of the DNA replication machinery that directly interacts with the DnaB helicase . Through detailed mechanistic work , we then expand on the work in P . aeruginosa to show that DciA localizes to the oriC , directly binds DNA , and affects the association of DnaB with DnaA , likely contributing to a role during DNA replication initiation . We also assign a role to DUF721 as a DnaA-NTD-like ( DANL ) protein-protein interaction domain and identify a tryptophan within the DANL domain that is critical for DciAMtb function in vivo and can affect DnaB binding in vitro . We find that both the N-terminus and C-terminus of DciAMtb are able to directly bind DnaB , indicating a complicated association between DciA and the replication machinery that will be the focus of future studies . The DANL domain of DciAMtb adds to the examples of DNA replication proteins across bacteria that share structural similarity despite divergent sequences ( e . g . HobA and DiaA [58] as well as winged-helix domains of bacterial , mammalian , and archaeal DNA replication proteins [60 , 61] ) . It also remains possible that the DciA DANL domain interacts with additional proteins beyond DnaB , which will be explored in future studies . To study the role of DciA in mycobacteria , we used a Tet-DciA strain that depletes dciAMtb transcripts following the removal of ATc from the media . dciAMtb transcripts were depleted by 16 hours following the removal of ATc , at which point we observed abnormal nucleoid morphology . The abnormal nucleoid morphology is the earliest phenotype observed during dciAMtb depletion and is followed by elongated cell lengths at 24 hours and then slower growth between 24 and 36 hours of dciAMtb depletion ( S2 Fig ) . These data suggest that the defect in DNA replication that leads to abnormal nucleoid morphology results in the cell cycle block and a subsequent growth defect . Live-cell imaging revealed that both shorter and longer dciAMtb-depleted cells underwent division , leading to increased heterogeneity among cell birth lengths and indicating a dysregulation of cell size and division ( S1 and S2 Movies , S3 Fig ) . It is unknown what mechanisms control cell size in mycobacteria but our data implicates DciA in affecting the coordination of cell division and cell size . Since this is the first time a mycobacterial DNA replication mutant has been analyzed by live-cell imaging , future studies will determine if this dysregulation of cell size control is a general feature of mycobacterial DNA replication mutants or if it is DciA-specific . These studies will shed light onto the mechanisms that control mycobacterial cell size and cell cycle . Helicase loaders have not been identified in the majority of bacterial phyla , making it tempting to assign this role to DciA . However , DciA homologs do not encode ATPase domains and , thus , do not fit the definition of a helicase loader [2] . Their small size , association with the replicative helicase , and importance in DNA replication are more reminiscent of the primosomal proteins B . subtilis DnaD and E . coli DnaT [18 , 62] . However , DnaD is present in the cell at very high levels [18] , and neither DnaD nor DnaT has a high isoelectric point or a structurally predicted DANL domain . Therefore , DciA is unique from any protein that has been shown to facilitate DnaB association with replication complexes . There are two possible explanations for the lack of identified helicase loaders in mycobacteria . DnaB could be loaded by DciAMtb through a novel mechanism that is independent of any external ATPase activity and thereby replaces the need for a canonical helicase loader . Alternatively , DciAMtb could function in concert with an unknown ATPase in order to load the helicase . Our co-IP experiments identified a few known and hypothetical ATPase and ATP-binding domain containing proteins ( S4 Table ) . Future studies will investigate the interactions of DciAMtb with these proteins and their roles in replication , as well as whether DciAMtb affects activities of DnaB outside of its interaction with DnaA ( Fig 6F and 6H ) . Our studies also revealed that DciAMtb can interact with single and double-stranded DNA in vitro . DciA homologs have high isoelectric points that likely mediate DNA-binding ( Fig 6A ) . Integration host factor ( IHF ) also has a high isoelectric point and has been shown in E . coli to bind and bend oriC DNA [63] , possibly to bring DnaA-boxes closer together and promote replication initiation [64] . The effect of IHF on oriC in mycobacteria has not yet been studied , but Lsr2 , which has a high isoelectric point and is the mycobacterial functional analog of H-NS , is able to bind to open reading frames proximal to oriC [50] . ChIP-qPCR experiments showed that despite the ability of DciAMtb to bind DNA without sequence specificity , it is enriched at the oriC . This suggests that the specific localization of DciAMtb to oriC relies on its association with DnaB and the DNA replication initiation complex . We originally identified DciAMsm as being upregulated in response to DNA damage [13] . Although we do not know the exact role for DciA in the response to DNA damage , one can imagine a number of possibilities . First , DNA damage and DNA replication are inextricably linked , as DNA damage can disrupt the movement of the replisome and the repair of DNA damage involves proteins that are also involved in DNA replication . Second , DciAMtb could have a role in replication restart , which involves the loading of the replicative helicase onto non-oriC DNA and is dependent on proteins that bind DNA in a structure-dependent but sequence-independent manner in other bacteria [65] . Replication restart has not yet been studied in mycobacteria ( Table 1 ) . Lastly , DciAMtb could have a direct role in the repair of DNA damage independent of its role in DNA replication by interacting with DNA damage repair helicases . DciAMtb co-immunoprecipitated with a number of helicases involved in DNA repair ( Fig 6B ) , and future studies determining whether DciAMtb directly interacts with these helicases will help to elucidate the connection between DciA and DNA damage responses . Although we can detect immunoprecipitated HA-DciAMtb by MS , we have been unable to detect DciAMsm , DciAMtb , or HA-DciAMtb from cell lysate by western blot using a polyclonal antibody that we raised against DciAMtb or an anti-HA antibody . Since ClpX , as well as ClpP1 and ClpC1 , co-immunoprecipitated with HA-DciAMtb ( S4 Table ) , one explanation for the low level of DciAMtb protein in the cell is that DciAMtb is a target of Clp protease , which is essential in mycobacteria [66] . ClpX and ClpC1 are adaptors for ClpP , which consists of ClpP1 and ClpP2 in mycobacteria . Raju et al . used clpP1P2 depletion strains in M . smegmatis and Mtb to identify proteins that were present at higher levels during clpP1P2 depletion , suggesting that they are targets of Clp [66] . DciAMtb protein levels were 2 . 53 fold higher ( P value of . 00285 ) during clpP1P2 depletion in Mtb , further supporting that DciAMtb is targeted by Clp . Interestingly , DnaB was also found to be significantly enriched during clpP1P2 depletion in M . smegmatis and Mtb . The ClpP protease is known to regulate cell cycle progression proteins in Caulobacter crescentus , where , like mycobacteria , ClpP is essential [67] . Therefore , it is possible that regulation of DNA replication proteins comprises one of the essential roles for Clp in mycobacteria . Together , this study has elucidated functions of DciAMtb that are likely applicable to other DciA homologs . dciA genes are not found in E . coli and B . subtilis , the organisms that have traditionally been used to study bacterial replication , and as a result have remained undiscovered until recently , despite being widely conserved throughout the bacterial kingdom . Therefore , this study contributes to a developing new paradigm of bacterial DNA replication .
Mtb Erdman strain was grown at 37°C in Middlebrook 7H9 supplemented with 0 . 5% glycerol , 0 . 05% Tween 80 , and 10% oleic acid/albumin/dextrose/catalase . M . smegmatis mc2155 and its derivatives were grown on LB agar plates supplemented with 0 . 5% dextrose and 0 . 5% glycerol and in LB broth supplemented with 0 . 5% dextrose , 0 . 5% glycerol , and 0 . 05% Tween 80 except in the [5 , 6-3H] uracil experiment , in which cells were grown in Middlebrook 7H9 supplemented with 0 . 5% dextrose , 0 . 5% glycerol , and 0 . 05% Tween 80 and for live-cell imaging , in which M . smegmatis was grown in Middlebrook 7H9 supplemented with 0 . 05% Tween 80 , 10% ADC ( albumin , dextrose , catalase ) , and 0 . 2% glycerol . All bacterial strains , plasmids , and primers used in this study are described in detail in S1–S3 Tables . RNA was extracted from mycobacteria using Trizol ( Invitrogen ) followed by high salt and isopropanol precipitation . Contaminating genomic DNA was removed using the TURBO DNA-free kit ( ThermoFisher Scientific ) , cDNA was synthesized using Superscript III ( Invitrogen ) , and iTaq Universal SYBR Green Supermix ( Bio-Rad ) was used in qRT-PCR reactions . Primers used to amplify 16S rRNA from Mtb and M . smegmatis , dciaAMtb ( Rv0004 ) , dciAMsm ( MSMEG_0004 ) , and M . smegmatis dnaA , dnaN , MSMEG_0002 , recF , gyrB and sigA are found in S3 Table . Levels of dciAMsm , dciaAMtb , dnaA , dnaN , MSMEG_0002 , recF , and gyrB , were normalized to either 16S rrnA or sigA transcript levels as previously described [68] . M . smegmatis was collected , washed once with equal volume Phosphate Buffered Saline ( PBS ) , resuspended in equal volume 1 μg/ml FM1-43FX ( Thermo Fisher ) diluted in PBS and incubated for 20 minutes at 37°C shaking . Cells were then fixed with 3% paraformaldehyde in PBS for 30 minutes shaking at 37°C . Fixed cells were applied to 0 . 1% poly-L-lysine ( Sigma ) treated multitest slides ( MP Biomedicals ) and then washed once with PBS . Cells were permeabilized by treatment with 2 mg/ml lysozyme ( Sigma ) at 37°C for 30 minutes followed by 0 . 1% Triton-X 100 ( Sigma ) for exactly five minutes at room temperature [69] . Cells were rinsed with PBS and stained with DAPI ( Thermo Fisher ) diluted to 5μg/ml with Slow Fade Antifade Equilibration Buffer ( ThermoFisher ) and mounted using the Slow Fade Antifade Kit according to the manufacturer’s instructions . Slides were visualized with a Zeiss Axioskop 2 Mot Plus equipped with an Axiocam MRm monochrome camera and a 100X , 1 . 4 numerical aperature Zeiss Plan Apochromat oil objective and images were acquired using Axiovision 4 . 6 software , or using an upright Zeiss Axio Imager M2 fluorescence microscope and the Zen Blue image acquisition software . M . smegmatis cells were grown to log phase overnight with shaking at 37°C . M . smegmatis cells were filtered through a 10 μm filter to remove clumps before being loaded into a custom polydimethylsiloxane ( PDMS ) microfluidic device , as before [26] . The viewing device incorporates a main microfluidic channel for continuous flow for growth media , with a height of approximately 10–17 μm , and viewing chambers with a diameter of 60 μm and a height of 0 . 8−0 . 9 μm . 2% DMSO and 0 . 0625 mg/ml FM4-64 are present in the flowed medium , to stain septal membranes . The microfluidics device was placed on an automated microscope stage housed within an environmental chamber maintained at 37°C . M . smegmatis cells were imaged for up to 40 hours using a widefield DeltaVision PersonalDV ( Applied Precision , Inc ) with a hardware-based autofocus . Cells were illuminated with an InsightSSI Solid State Illumination system every 15 minutes: FM4-64 was visualized with 475nm excitation and 679 nm emission wavelengths , and cells were imaged using transmitted light brightfield imaging . Each set of images was illuminated with identical imaging conditions that were optimized to decrease phototoxicity inherent in long-term fluorescent imaging of live cells . Tet-DciA was grown in the presence of hygromycin and ATc overnight and for the duration of the DciAMtb replete control movies ( S1 Movie ) . Tet-DciA was grown in the presence of hygromycin and ATc overnight followed by growth in the presence of hygromycin alone for 10 . 5 hours prior to imaging and for the duration of imaging ( S2 Movie ) . For analysis , all Tet-DciA cells grown in depleting conditions were born at least 24 hours after ATc was removed from the media . M . smegmatis was fixed in 2% paraformaldehyde/2 . 5% glutaraldehyde ( Polysciences Inc . ) in 100 mM sodium cocadylate buffer , pH 7 . 2 for 1 hour at room temperature . Samples were washed in sodium cacodylate buffer and postfixed in 1% osmium tetroxide ( Polysciences Inc . ) for 1 hr . Samples were then rinsed extensively in dH20 prior to en bloc staining with 1% aqueous uranyl acetate ( Ted Pella Inc . ) for 1 hr . Following several rinses in dH20 , samples were dehydrated in a graded series of ethanol and embedded in Eponate 12 resin ( Ted Pella Inc . ) . Sections of 95 nm were cut with a Leica Ultracut UCT ultramicrotome ( Leica Microsystems Inc . ) , stained with uranyl acetate and lead citrate , and viewed on a JEOL 1200 EX transmission electron microscope ( JEOL USA Inc . ) equipped with an AMT 8 megapixel digital camera and AMT Image Capture Engine V602 software ( Advanced Microscopy Techniques ) . Assays were carried out similarly to published reports [25 , 28] , with a few modifications . M . smegmatis cells were collected at ODλ600 = 0 . 3 , washed once with 7H9 media , and inoculated into fresh 7H9 media containing 1μCi/ml [5 , 6-3H]-uracil ( Perkin Elmer ) . [5 , 6-3H]-uracil is converted to [5 , 6-3H]-thymidine and incorporated into DNA in mycobacteria [25 , 28] . After 20 and 60 minutes of labeling , 3 ml aliquots of cells were taken for processing to measure counts per minute ( cpm ) of 3H incorporated into DNA or for enumeration of bacterial counts by colony forming units ( cfu ) . A 7H9 media plus [5 , 6-3H]-uracil sample was also collected . Aliquots for 3H cpm were treated with 0 . 3M KOH to hydrolyze RNA and incubated at 37°C for 24 hours . Macromolecules were then precipitated with ice cold 10% trichloroacetic acid ( Sigma ) and filtered onto glass 25 mm GF/C filters ( GE Healthcare ) . Filters were dried under a heat lamp and submerged in 15 ml of Ultima Gold liquid scintillation fluid ( Perkin Elmer ) . 3H counts per minute ( cpm ) were measured using a Beckman LS6000IC scintillation counter that was programmed to read each sample for 5 minutes . All presented data indicates the cpm of that sample minus the cpm from the processed media + [5 , 6-3H]-uracil control to subtract background , relative to the cfu of that sample . The efficiency of RNA hydrolysis by KOH treatment was confirmed by including a KOH-untreated control . Cells were fixed and permeabilized as described for fluorescent microscopy [69] with an additional RNaseIf treatment ( NEB ) , and stained with 100 μM DAPI diluted in water for 15 minutes at room temperature . Cells were then resuspended in PBS , sonicated , and passed through a 30μm filter to remove clumps before flow cytometry analysis . Samples were analyzed using a FACSAria ( Becton Dickinson ) and data were processed with FlowJo ( Treestar ) . All cells ( events ) were included in the analysis of DAPI intensity . Mtb dciAMtb , HA-dciAMtb , dciAMtbW113A , HA-dciAMtbW113A , truncations of HA-dciAMtb and HA-dciAMtbW113A , dnaA , and HA-dnaA were cloned into pGEX-6P ( GE Healthcare Life Sciences , S2 and S3 Tables ) . The plasmids were transformed into BL21 ( DE3 ) ( Novagen ) . Mtb DnaB and DnaB-FLAG were amplified from Mtb genomic DNA following the published cloning scheme[35] to exclude the native intein , cloned into pET SUMO ( Invitrogen ) and transformed into BL21 ( DE3 ) cells ( Novagen ) . RelMtb1-394 was cloned into pET SUMO [70] . Transformed cells were grown to mid-logarithmic phase ( ODλ600 0 . 6–0 . 8 ) and induced with 1mM isopropyl β-D-1-thiogalactopyranoside ( IPTG ) ( GoldBio ) for 3 hours at 37°C . Cells were harvested by centrifugation and resuspended in lysis buffer supplemented with 1 mg/ml lysozyme and flash frozen . The lysis buffer for DnaB constructs was 50mM Tris-HCl pH8 , 300mM NaCl , 5mM imidazole pH8 and 1mM β-mercaptoethanol ( BME ) . The lysis buffer for DnaA was PBS , and for the DciAMtb constructs the lysis buffer was 0 . 25M Urea , 750mM NaCl , 2 . 7mM KCl , 10mM sodium phosphate dibasic heptahydrate , 2mM potassium phosphate supplemented with . 01% Nonidet P-40 ( Sigma ) and . 1% TritonX 100 . After thaw , cells were sonicated , treated with 9 units per ml Benzonase ( Sigma ) and 6mM MgCl2 when indicated , and spun at 10 , 000g to clarify the lysate . For the purification of DnaB and RelMtb1-394 lysate was incubated with Ni-NTA agarose ( Qiagen ) , washed with 50mM Tris-HCl pH8 , 300mM NaCl , 20mM imidazole pH8 , and 1mM BME , and proteins were eluted with 50mM Tris-HCl , pH8 , 300mM NaCl , 250mM imidazole pH8 , and 1mM BME . The His-SUMO tags were cleaved with His-tagged Ulp1 enzyme . A second incubation with Ni-NTA agarose was used to bind His-Ulp1 and the cleaved His-SUMO tag , and un-tagged recombinant protein was collected as flow-through . For the purification of DnaA and DciAMtb constructs , lysate was incubated with Protino Glutathione Agarose 4B ( Macherey-Nagel ) , washed with wash buffer ( PBS or DciAMtb lysis buffer plus . 01% Nonidet P-40 ) , and cleaved on beads with GST-tagged PreScission Protease ( GE Healthcare ) in 50mM Tris-HCl pH7 , 150mM NaCl , 1mM EDTA , and 1mM DTT , with un-tagged recombinant protein collected as flow-through . DNA fragments ( see S3 Table ) containing oriCMtb [33] and rrnAPL [19] were amplified by PCR and products were gel-purified using QIAquick column ( Qiagen ) . 250 ng of gel-purified dsDNA or 3x FLAG ssDNA oligo ( see S3 Table ) were labeled with T4 polynucleotide kinase ( NEB ) and [γ-32P]-ATP , and unincorporated [γ-32P]-ATP was removed using Illustra ProbeQuant G-50 microcolumns ( GE Healthcare ) . 20 , 000 cpm of labeled probe , 10 μg BSA , and the indicated amounts of DciAMtb or DciAMtbW113A were mixed with buffer ( 50mM Tris-HCl pH7 , 150mM NaCl , 1mM EDTA , 1mM DTT ) in a total volume of 12μl and incubated for 20 minutes at room temperature . Samples underwent native electrophoresis in 4–20% nondenaturing TBE polyacrylamide gels ( Invitrogen ) . The gels were dried and exposed to film for detection by autoradiography . For immunoprecipitation from M . smegmatis cell lysates , 1 liter cultures of mid-logarithmic phase M . smegmatis were washed with PBS and frozen at -80°C until processing . Cell pellets were resuspended in 10mls NP-40 Buffer ( 10 mM sodium phosphate , pH8 , 150 mM NaCl , 1% Nonidet P-40 , and Roche EDTA-free Complete protease inhibitor cocktail ) and lysed using a Constant Systems Cell Disruptor ( 4 passes at 40k psi ) , and spun at 50 , 000g for 30 minutes to generate lysate . When indicated , some lysates were treated with DNase I ( NEB ) according to manufacturers instructions . 1 ml of lysate was added to 50 μL monoclonal anti-HA agarose ( Sigma ) and rotated at 4°C overnight . The anti-HA agarose was then washed 3 times with NP-40 buffer and immune complexes were eluted with 500 μg/ml HA peptide ( Roche ) in IP elution buffer ( 50 mM Tris-HCl , pH 7 . 5 , 50 mM NaCl with Roche EDTA-free Complete protease inhibitor cocktail ) . For pull-down experiments with purified protein , 0 . 3114 nmol of purified bait protein in 100 μL NP-40 buffer was bound to either 50 μL anti-HA agarose ( Sigma ) or 40 μL anti-FLAG M2 affinity gel ( Sigma ) and rotated for 5–6 hours 4°C . The bait-bound matrix was then washed 3 times with NP-40 buffer and twice molar excess ( 0 . 6228 nmol ) of prey protein was added ( unless otherwise indicated ) in 275 μL NP-40 buffer and rotated at 4°C overnight . The matrix was then washed three times and immune complexes were eluted with either 500 μg/ml HA peptide or 150 μg/ml FLAG peptide in IP elution buffer as described above . For western blot analyses , FLAG-tagged proteins were detected with mouse monoclonal anti-FLAG clone M2 antibody ( Sigma ) and HA-tagged proteins were detected with mouse monoclonal anti-HA clone HA-7 ( Sigma ) . To detect DciAMtb , we used a rabbit polyclonal anti- DciAMtb antibody that was generated by Cocalico Biologicals , Inc . by raising rabbit anti-sera against purified DciAMtb protein . Western blots were visualized on a ChemiDoc Touch Imaging System ( Biorad ) as well as film and quantified using ImageLab software version 5 . 2 . 1 software ( Biorad ) . 50 mL cultures of M . smegmatis expressing either HA-tagged DciAMtb ( HA-DciAMtb ) , HA-tagged CarD ( HA-CarD ) , or untagged DciAMtb ( No tag ) were grown to an ODλ600 of 0 . 6 . Protein-nucleic acid complexes were crosslinked with 2% formaldehyde for 30 min at room temperature , crosslinking was quenched with 0 . 125M glycine , and cells were lysed in ChIP lysis buffer ( 50mM HEPES-KOH pH7 . 5 , 140mM NaCl , 1mM EDTA , and 1% TritonX 100 , 1X protease inhibitors ( Roche ) , and 2mg/ml lysozyme ) . After sonication , 0 . 5 ml of lysate was kept as input fractions and the remaining lysate was rotated with anti-HA agarose ( Sigma ) overnight at 4°C . Anti-HA agarose was washed 2 times each with ChIP lysis buffer , ChIP lysis buffer plus 360mM additional NaCl , ChIP wash buffer ( 10mM Tris-HCl pH8 , 250mM LiCl , 0 . 5% NP-40 , 0 . 5% sodium deoxycholate , and 1mM EDTA ) , and TE buffer ( 10mM Tris-HCl pH8 and 1mM EDTA ) . All buffers were supplemented with 1X protease inhibitors . Protein-nucleic acid complexes were eluted from the anti-HA agarose twice with ChIP elution buffer ( 50mMTris-HCl pH8 , 10mM EDTA , 1% sodium dodecyl sulfate , 1X protease inhibitors ) for 10 min at 65°C with agitation . Inputs and eluates were incubated at 65°C overnight to reverse crosslinking . Two phenol-chloroform extractions were performed consecutively on inputs and eluates and DNA was precipitated with ethanol and resuspended in TE buffer . Inputs and eluates were diluted to 0 . 5896 ng/μl and 0 . 426 ng/μl , respectively . Quantitative PCR ( qPCR ) was performed with 1 μl of the diluted DNA using primers to amplify three DNA fragments in oriC ( oriC1 , oriC2 , oriC3 ) , a fragment in the promoter of rplN , and a fragment within the sigA coding region ( sigAIN ) ( See S3 Table and S8A Fig ) . Recombinant DnaB and DciAMtb were purified from E . coli as described above . An additional dialysis step to remove Tris from the buffer was performed into 150mM NaCl , 10mM NaPO4 pH8 , 1mM BME . DnaB was non-specifically biotinylated in vitro using EZ-Link NHS-PEG4-Biotin ( ThermoFisher ) . Five molar excess of biotin relative to DnaB was used and incubated for 30 minutes at room temperature . Excess biotin was removed using Zeba Spin Desalting columns ( ThermoFisher ) . The Octet RED96 System was used to attain biolayer interferometry ( BLI ) progress curves . Assay buffer for all steps consisted of 150mM NaCl , 0 . 02% Tween-20 , 0 . 1% bovine serum albumin , and 10mM NaPO4 pH8 . Briefly , streptavidin ( SA ) biosensor pins ( ForteBio ) were first equilibrated by being dipped into assay buffer for a 180 second baseline step , and then captured 200nM biotinylated DnaB during a 200 second loading step , followed by a 180 second baseline step in assay buffer . After performing an additional 60 second baseline step in assay buffer , pins were dipped into DciAMtb protein samples for a 300 second association step , followed by a 300 second dissociation step in assay buffer . This series of 60 second baseline , 300 second association , and 300 second dissociation steps was performed for each concentration of DciAMtb . Curves were corrected by subtracting double reference of both biotin-coated pins dipped into the DciAMtb wells and DnaB-coated pins dipped into buffer only . Data was analyzed on ForteBio Data Analysis 6 . 4 . Processed data were fit globally for all concentrations of DciAMtb in a 1:1 kinetic binding model . Bands generated by SDS-PAGE followed by staining with ProteoSilver Plus Silver Stain Kit ( Sigma ) , were cut out and destained according to manufacturer’s instructions , and submitted to the Proteomics & Mass Spectrometry Facility at the Danforth Plant Science Center for trypsin digestion followed by LC-MS/MS analysis . Prism6 ( Graphpad Software , Inc . ) was used to determine statistical significance of differences . Unpaired two-tailed Student’s t-test was used to compare two groups with similar variances . Unpaired two-tailed Student’s t-test with Welch’s correction was used to compare two groups with different variances . One-way analysis of variance ( ANOVA ) and Tukey’s multiple comparison test were used to determine significance when more than two groups were compared . When utilized , center values and error bars represent mean ± SEM . * p <0 . 05 , ** p <0 . 01 , *** p<0 . 001 , **** p <0 . 0001 . | DNA is the molecule that encodes all of the genetic information of an organism . In order to pass genes onto the next generation , DNA has to first be copied through a process called DNA replication . Most of the initial studies on bacterial DNA replication were performed in Escherichia coli and Bacillus subtilis . While these studies were very informative , there is an increasing appreciation that more distantly related bacteria have diverged from these organisms in even the most fundamental processes . Mycobacteria , a group of bacteria that includes the human pathogen Mycobacterium tuberculosis , are distantly related to E . coli and B . subtilis and lack some of the proteins used for DNA replication in those model organisms . In this study , we discover that a previously uncharacterized protein in Mycobacteria , named Rv0004 , is essential for bacterial viability and involved in DNA replication . Rv0004 is conserved in most bacteria but is absent from E . coli and B . subtilis . Since Rv0004 is essential for mycobacterial viability , this study both identifies a future target for antibiotic therapy and expands our knowledge on the diversity of bacterial DNA replication strategies , which may be applicable to other organisms . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"medicine",
"and",
"health",
"sciences",
"nuclear",
"staining",
"pathology",
"and",
"laboratory",
"medicine",
"enzymes",
"cell",
"cycle",
"and",
"cell",
"division",
"pathogens",
"cell",
"processes",
"bacillus",
"dna-binding",
"proteins",
"microbiology",
"enzymology",
"membrane",
"staining",
"prokaryotic",
"models",
"dna",
"replication",
"experimental",
"organism",
"systems",
"dna",
"bacteria",
"bacterial",
"pathogens",
"research",
"and",
"analysis",
"methods",
"specimen",
"preparation",
"and",
"treatment",
"staining",
"proteins",
"medical",
"microbiology",
"microbial",
"pathogens",
"actinobacteria",
"biochemistry",
"dapi",
"staining",
"helicases",
"cell",
"biology",
"nucleic",
"acids",
"mycobacterium",
"tuberculosis",
"bacillus",
"subtilis",
"genetics",
"biology",
"and",
"life",
"sciences",
"organisms"
] | 2017 | Rv0004 is a new essential member of the mycobacterial DNA replication machinery |
Plants , as sessile organisms , survive environmental changes by prioritizing their responses to the most life-threatening stress by allocating limited resources . Previous studies showed that pathogen resistance was suppressed under abiotic stresses . Here , we show the mechanism underlying this phenomenon . Phosphorylation of WRKY45 , the central transcription factor in salicylic-acid ( SA ) -signalling-dependent pathogen defence in rice , via the OsMKK10-2–OsMPK6 cascade , was required to fully activate WRKY45 . The activation of WRKY45 by benzothiadiazole ( BTH ) was reduced under low temperature and high salinity , probably through abscisic acid ( ABA ) signalling . An ABA treatment dephosphorylated/inactivated OsMPK6 via protein tyrosine phosphatases , OsPTP1/2 , leading to the impaired activation of WRKY45 and a reduction in Magnaporthe oryzae resistance , even after BTH treatment . BTH induced a strong M . oryzae resistance in OsPTP1/2 knockdown rice , even under cold and high salinity , indicating that OsPTP1/2 is the node of SA-ABA signalling crosstalk and its down-regulation makes rice disease resistant , even under abiotic stresses . These results points to one of the directions to further improve crops by managing the tradeoffs between different stress responses of plants .
Plants , as sessile organisms , are continuously exposed to various environmental stresses in nature . To cope with such conditions using limited resources , plants have evolved various mechanisms that enable resource allocation to the most life-threatening stress [1] [2] . Such tradeoffs between the responses to different stresses are often regulated by crosstalk between signalling pathways [3] [4] [5] . A number of studies have reported various signalling components that appear to influence signalling crosstalk . However , the precise molecular mechanisms that regulate the crosstalk remain poorly understood in most cases [6] [4] [7] . The salicylic acid ( SA ) signalling pathway plays a crucial role in pathogen defence . In Arabidopsis , NPR1 , the transcriptional cofactor , plays a major role in the SA defence signalling pathway [8] . In rice ( Oryzae sativa ) , in addition to OsNPR1/NH1 , the rice ortholog of NPR1 , the transcription factor ( TF ) WRKY45 plays a crucial role in the branched SA pathway [9] [10] [11] [12] . Up-regulation of WRKY45 by chemical defence inducers , such as benzothiadiazole ( BTH ) , or its overexpression , renders rice plants resistant against several pathogens , including fungus , such as Magnaporthe oryzae causing blast disease , and bacterium , such as Xanthomonas oryzae pv . oryzae causing leaf blight disease [9] [13] [14] , without major negative effects on plant growth . WRKY45 auto-regulates the transcription of its own gene [12] and is regulated by the ubiquitin-proteasome system [15] . Abscisic acid ( ABA ) signalling is mainly involved in plant responses to abiotic stresses , such as the cold , drought , and high salinity [16] [17] . However , ABA also acts as a modulator of defence responses against pathogens , both positively and negatively , with its negative role being more prevalent [18] [3] [19] [20] [4] [21] [5] [22] . Recent studies have shown that ABA antagonizes SA-signalling , thereby interfering with defence responses in tomato , Arabidopsis , and rice [23] [24] [25] . The WRKY TFs can be phosphorylated and activated by MAP kinases , as is the case with Arabidopsis WRKY33 [26] and Nicotiana benthamiana WRKY8 [27] . The negative regulation of MAP kinases through dephosphorylation by protein phosphatases , including Ser/Thr-specific phosphatases , dual-specificity phosphatases , and Tyr-specific phosphatases ( PTPases ) , has been reported [28] [29] . We have previously reported that activated MAP kinases can phosphorylate WRKY45 in vitro , and MAP kinases can be activated in response to SA [30] . However , details of WRKY45 phosphorylation , and the biological significance of the phosphorylation have remained unknown . Rice blast is one of the most serious crop diseases in the world . It has been reported that rice plants are more blast susceptible under abiotic stresses , such as low temperature and drought [31] [32] [33] . In rice , ABA treatment severely compromised M . oryzae resistance [34] , which is mediated by suppression of WRKY45 and OsNPR1/NH1 genes via ABA signalling [25] [35] . These authors suggest that ABA signalling plays a role in increased blast susceptibility under low temperature . In this report , we show the mechanism underlying blast resistance through the activation of WRKY45 by MAP kinase ( MAPK ) -dependent phosphorylation in the SA pathway . Moreover , we showed that the tyrosine dephosphorylation of the MAPK by PTPases , OsPTP1/2 , is responsible for the MAPK inactivation under abiotic stresses or in the presence of exogenous ABA . Additionally , the knockdown of OsPTP1/2 uncoupled the induced blast resistance from the abiotic stresses . These findings should enable the development of technologies to protect rice from diseases even under the influence of environmental factors .
WRKY45 was phosphorylated by OsMPK6 in vitro in the presence of a constitutively active form of OsMKK10-2 ( MKK10-2D , Os03g0225100 , LOC_Os03g12390 . 1 ) , which is a rice MAPK kinase that phosphorylates and activates OsMPK6 in vitro [30]; however , its biological significance was unknown . To further analyse the WRKY45 phosphorylation , we therefore determined the sites in WRKY45 that are phosphorylated by OsMPK6 . Incubation in vitro of fragmented WRKY45 polypeptides fused to a maltose binding protein ( MBP ) with OsMPK6 revealed that only two regions ( amino acids 239–292 and 281–326 ) of WRKY45 were phosphorylated ( S1 Fig ) . By a consequent substitution study on candidate phosphorylation sites ( Ser or Thr ) , we found that Thr266 , Ser294 , and Ser299 were required for phosphorylation of WRKY45 by OsMPK6 in vitro ( S2 Fig ) . To assess the effects of the phosphorylation on transcriptional activity , we generated mutant forms of WRKY45 in which all the three amino acids were replaced by Asn ( NNN ) or Asp ( DDD ) to mimic dephosphorylation and phosphorylation of all three sites , respectively , and tested them by transient reporter assays ( Fig 1A ) . The transactivation activity of DDD was 2–4-fold higher than that of NNN ( Fig 1B ) , indicating that the phosphomimetic mutation elevated the transcriptional activity of WRKY45 . These results suggest that WRKY45 phosphorylation results in its activation in vivo . The level of wild-type ( WT ) WRKY45 activity was between those of NNN and DDD ( Fig 1B ) , consistent with its partial phosphorylation state . We attempted to generate transgenic plants overexpressing NNN and DDD mutants tagged with the myc sequence at their N-termini , but we obtained neither of them . Then , we focused on the carboxyl-terminal region that contained two of the three phosphorylation sites , the closely located Ser residues , Ser294 and Ser299 , because the carboxyl-terminal region of WRKY45 is critical for transcriptional activity [15] . We generated mutants in which the two serines were substituted with Asp ( TDD ) or Asn ( TNN ) , and attempted to overexpress the mutant cDNAs in rice transformants ( Fig 1A ) . While we were unable to obtain the transformants for TDD , we obtained several for TNN . Whereas the TNN transformants accumulated the transgene-derived proteins to levels comparable , or even slightly higher , than those in WT WRKY45-overexpressing ( ox ) transformants , they showed no significant enhancement of blast resistance ( Fig 1C ) , indicating that the substitution of these two serines compromised the function of WRKY45 . To examine the contribution of the phosphorylation at these two Ser to transcriptional activity , we generated new mutants , DNN and NDD , and analysed them in the transient system ( Fig 1A ) . The relative activity of NDD was as high as that of DDD , while that of DNN was as low as that of the NNN mutant ( Fig 1D ) . These results suggest that the phosphorylation of Ser294 and/or Ser299 , but not that of Thr266 , is important for the transcriptional activity of WRKY45 , consistent with the results of blast resistance test ( Fig 1C ) . To examine whether these two serines in WRKY45 are phosphorylated in vivo , we treated the extracts from WT and TNN WRKY45-ox plants with lambda protein phosphatase ( PPase ) . The electrophoretic mobility of the TNN form treated with the PPase was indistinguishable from that of WT WRKY45 ( Fig 1E ) , indicating that the effect of Ser-to-Asn substitutions at the two sites on the electrophoretic mobility is negligible . Without the PPase treatment , two bands were seen in both WT and TNN extracts ( Fig 1E ) , suggesting that both WT and the mutant WRKY45 proteins were phosphorylated in vivo . The slower mobility ( upper ) bands in WT extracts were broad and much more intense than the faster mobility ( lower ) band for the unphosphorylated WRKY45 protein . By contrast , the upper band in the TNN extract was less intense than the lower band and thinner . These reproducible results suggest that Ser294 and/or Ser299 are/is actually phosphorylated in plant cells . This idea was also supported by the results of Phos-Tag polyacrylamide gel electrophoresis [36] ( S3 Fig ) . Taken together , the phosphorylation of Ser294 and/or Ser299 of WRKY45 and consequent activation of its transcriptional activity are required for the full functioning of WRKY45-dependent blast resistance . We have previously shown that OsMPK6 becomes active in response to SA [30] . Here , we monitored the activation state of OsMPK6 by examining the dual phosphorylation of MAPK at Thr and Tyr in the TEY-signature ( position 225–227 in OsMPK6 ) [37] in calli . An immunoblot using anti-pTEpY antibody showed that a band , which is missing in osmpk6 mutant and corresponds to that for OsMPK6 in gel mobility , was intensified by the SA-treatment in WT calli ( Fig 2 ) . The increase of the band intensity in response to SA was rather weak , which we interpret to be due to sporadically elevated basal OsMPK6 phosphorylation level because of high SA levels in rice . In the osmpk6 mutant calli , a faster migrating band appeared , consistent with our previous observations in an in-gel kinase assay [30] . In leaves , OsMPK6 was also dually phosphorylated and another band appeared ( Fig 2 ) . These results suggest that OsMPK6 is a major MAPK activated by SA . To mimic the activation of the SA pathway that leads to the activation of OsMPK6 and then WRKY45 , we performed in vitro kinase assays using MKK10-2D and mutant forms of OsMPK6 as substrates ( Figs 3A and S4 ) . A kinase-dead form of OsMPK6 , K96R , was phosphorylated by MKK10-2D in vitro; however , another kinase-dead form , in which , in addition to the K96R mutation , the Thr and Tyr in the TEY-signature were replaced by Asp ( K96R/T225D/Y227D ) , was not ( Fig 3A ) . In addition , OsMPK6 mutants in which the Tyr and Thr were independently substituted , T225D and Y227A , were less phosphorylated than the WT OsMPK6 ( S4 Fig ) . These results indicate that MKK10-2D phosphorylates the TEY-signature of OsMPK6 specifically and suggest that OsMPK6 is activated by OsMKK10-2 through the specific phosphorylation . Then , we expressed MKK10-2D in rice plants using the dexamethasone ( Dex ) -inducible system ( GVG-MKK10-2D ) [38] and monitored the activation of OsMPK6 by pTEpY antibody . The dual phosphorylation was induced after the Dex-treatment in two independent GVG-MKK10-2D lines ( Fig 3B , α-pTEpY ) . In these plants , WRKY45 expression ( Fig 3C ) and blast resistance ( Fig 3D ) were also induced after the Dex treatment . These results demonstrate that the activation of OsMPK6 by MKK10-2D is sufficient for the induction of WRKY45 expression and blast resistance without exogenous SA or BTH . OsMKK4 can also phosphorylate and activate OsMPK6 in vitro and in vivo [39] [30] . However , the induced expression of the OsMKK4 constitutively active form using the Dex induction system failed to induce WRKY45 expression , while phenylalanine ammonia lyase gene , as a positive control , was induced ( S5 Fig ) . These results suggest that OsMKK10-2 , but not OsMKK4 , is involved in the MAPK cascade in the SA signalling pathway leading to WRKY45 up-regulation and blast resistance . Then , we investigated the effects of ABA on the OsMPK6–WRKY45 pathway in GVG-MKK10-2D lines ( Fig 4A and S6 Fig ) . We induced OsMKK10-2D in the GVG-MKK10-2D transformants by Dex treatment in the presence of ABA . Interestingly , the accumulation of WRKY45 transcripts was severely reduced by the ABA treatment , although the induction of the MKK10-2D transgene was unaffected by ABA ( Fig 4A ) . These results imply a possibility that ABA-signalling interfered with the SA pathway by affecting some molecular event ( s ) between OsMKK10-2 activation and WRKY45 transcription . We examined whether ABA-signalling affects the phosphorylation status of OsMPK6 . Strikingly , the ABA treatment significantly lowered the dual phosphorylation level of OsMPK6 in parallel with the repression of WRKY45 transcription ( Fig 4A and S6 Fig ) . These results suggest that ABA-signalling dephosphorylates OsMPK6 or inhibits its phosphorylation by MKK10-2D . In Arabidopsis , MAPK can be dephosphorylated by Ser/Thr-phosphatases , dual-specificity phosphatases , and PTPases [28] [29] . This information , taken together with our results , led us to presume that OsMPK6 could be dephosphorylated in response to ABA , rather than ABA inhibiting the MAPK kinase . To further investigate the phosphorylation state of OsMPK6 , we used anti-phospho-Tyr and anti-phospho-Thr specific antibodies in an immunoblot analysis . While the phospho-Thr signal was unaffected , the phospho-Tyr signal became faint in parallel with that of dual phosphorylation ( Fig 4A and S6 Fig ) , suggesting that the Tyr , but not Thr , of OsMPK6 was dephosphorylated in response to ABA . Based on these results , we predicted the involvement of dephosphorylation of OsMPK6 by PTPases in the action of ABA . To assess the possible involvement of PTPases , we treated rice leaf segments with SA and ABA in the presence or absence of the PTPase inhibitors , Bay11-7082 [40] and vanadate . In the absence of these inhibitors , WRKY45 transcripts dramatically increased after the SA treatment; however , the co-treatment with ABA largely compromised the induction ( S7 Fig ) [25] . Meanwhile , the reduction of WRKY45 transcript levels by ABA was significantly less in the presence of these inhibitors ( Fig 4B and S7 Fig ) . Taken together with the results of immunodetection described above , we postulated that ABA-responsive PTPase ( s ) dephosphorylated/inactivated OsMPK6 , which in turn deactivated WRKY45 by under-phosphorylation , leading to the reduction of WRKY45 transcripts . The rice genome has two genes encoding putative Tyr-specific PTPases , OsPTP1 ( Os12g0174800 , LOC_Os12g07590 ) and OsPTP2 ( Os11g0180200 , LOC_Os11g07850 . 1 ) ( S8 Fig ) , which have high homology to AtPTP1 , a Tyr phosphatase that dephosphorylates MPK6 [41 , 42] . Active site signature of Tyrosine phosphatase ( S8 Fig ) , which is conserved in PTPs from all the organisms [43] , is also present in OsPTP1 and -2 , further lending support for these proteins being Tyr-specific phosphatases . To test this hypothesis , we generated rice transformants in which both PTPase genes were knocked down using the construct shown in Fig 5A ( PTP-wkd ) . Then , we investigated the dual phosphorylation of OsMPK6 in WT [Nipponbare ( NB ) ] and PTP-wkd rice plants after SA treatment in the absence or presence of ABA ( Fig 5B ) . In NB , the dual phosphorylation level increased after the SA treatment , but the increase was completely cancelled in the presence of ABA . In PTP-wkd rice , ABA did not suppress the level of dual phosphorylation , which further increased after SA treatment . These results strongly suggest that OsPTP1/2 is involved in ABA-dependent dephosphorylation of OsMPK6 . In the absence of ABA in PTP-wkd rice , the dual phosphorylation level was relatively high even without SA treatment; thus , OsPTP1/2 could also play a role in reducing the basal level of OsMPK6 phosphorylation . To test whether OsPTP1/2 directly dephosphorylate OsMPK6 , we performed in vitro dephosphorylation assays ( Fig 6 ) . Phospho-Tyr , but not phosphor-Thr , of OsMPK6 , due to phosphorylation by OsMKK10-2 , decreased when incubated with WT OsPTP1 fused with MBP ( Fig 6 , upper panels ) . However , the decrease of signals was not significant when incubated with a mutant OsPTP1 in which catalytically essential cysteine [44] ( positions 258 , S8 Fig ) was replaced with serine ( Fig 6 ) . We also monitored the ability of OsMPK6 to phosphorylate WRKY45 by the addition of recombinant WRKY45 and [γ-32P] ATP to the system preincubated with unlabelled ATP to activate OsMPK6 . The WRKY45 phosphorylation was completely abolished in the presence of WT OsPTP1 , whereas the phosphorylation level remained unchanged in the presence of the mutant PTPase ( Fig 6 , lower panels ) . These results indicate that OsPTP1 directly dephosphorylates OsMPK6 and inactivates its WRKY45 phosphorylation activity . These results support the notion that OsPTP1 mediates the suppression of the SA-OsMPK6-WRKY45 pathway via ABA-signalling . We did not detect evident activities for OsPTP2 using the same reaction mixtures , possibly because of suboptimal conditions . BTH , a chemical defence inducer , enhances disease resistance by acting on the SA signalling pathway in various plants , including rice [45] [46] . In Arabidopsis , the effect of BTH on defence responses is compromised by high salt conditions acting through ABA signalling [24] . To test the effects of abiotic stresses on BTH-induced blast resistance in rice , we pretreated WT rice plants , NB , with 10 μM ABA , low temperature ( 15°C/9°C , day/night ) , or high salinity ( 250 mM NaCl ) in the presence or absence of 10 μM BTH , inoculated them with M . oryzae and monitored for fungal growth ( Fig 7 ) . BTH confers a strong resistance against blast disease in rice [9] . Co-treatment with ABA compromised the resistance as reported previously [25] . Interestingly , BTH-treated plants under the cold and high salinity conditions were more highly susceptible to blast than the control plants , similar to the ABA/BTH co-treated plants ( Fig 7 ) . These results imply a possibility that the abiotic stresses mediated by ABA signalling negatively affected the BTH-induced blast resistance in rice . Activation of OsMPK6 by MKK10-2D , which mimics the activation of the SA pathway by SA/BTH , conferred rice plants with blast resistance . OsPTP1/2 dephosphorylated and inactivated OsMPK6 probably in response to ABA . These results led us to examine whether PTP-wkd plants were less sensitive to the ABA-mediated abiotic stresses , in regard to BTH-induced disease resistance ( Fig 7 ) . Under normal conditions , BTH induced as strong a blast resistance in PTP-wkd plants as in NB ( Fig 7 ) . Unlike NB; however , PTP-wkd plants exhibited strong blast resistance even in the presence of ABA ( Fig 7 ) . Moreover , PTP-wkd plants showed strong blast resistance under abiotic stress conditions , low temperature ( 15/9°C , day/night ) , and high salinity ( 250 mM NaCl ) ( Fig 7 ) . These results indicate the involvement of OsPTP1/2 in the suppression of SA/BTH-dependent blast resistance by ABA . Visible morphological and growth phenotypes were not observed in PTP-wkd under normal or stress conditions . SA/BTH-dependent defence system involves two independent sub-pathways , WRKY45 and OsNPR1 sub-pathways [9–12] . To investigate whether OsPTP1/2 act ( s ) on either sub-pathway specifically , we examined the gene expression patterns in PTP-wkd lines after the treatment with SA in the absence or presence of ABA ( Fig 8 and S10 Fig ) . In the absence of ABA , WRKY45 transcript levels in NB and PTP-wkd lines were within a similar range ( Fig 8A ) . In the presence of ABA , the transcript levels were significantly decreased in NB . By contrast , the transcript levels in PTP-wkd plants were increased to the levels without ABA ( Fig 8A ) . WRKY62 is a direct target gene of WRKY45 , and we have previously observed that this gene behaves similarly to WRKY45 in response to SA/BTH [9] [14] . In this experiment ( Fig 8B ) , the expression pattern of WRKY62 paralleled that of WRKY45 . We have previously reported that the expression of OsNPR1 , as well as that of WRKY45 , were suppressed by ABA [25] . Interestingly; however , no effect of the OsPTP1/2 double knockdown was observed on the suppression of OsNPR1 expression by ABA ( Fig 8C ) . The SalT gene is an ABA-inducible gene [47] [25]; however , the effect of the OsPTP1/2 double knockdown was not observed on the expression of this gene either ( Fig 8D ) . These results suggest that OsPTP1/2 acts specifically on the WRKY45 sub-pathway of SA-signalling under abiotic stress conditions ( Fig 9 ) . The expression of OsPTP1/2 genes was not positively affected by ABA ( Fig 8E and 8F ) , high salinity , or low temperature condition ( S9 Fig ) , suggesting that these genes are regulated post-transcriptionally by ABA and abiotic stresses .
Tyr-specific PTPases are also encoded in dicot genomes . Of these , only Arabidopsis AtPTP1 has been functionally characterized , and the tyrosine dephosphorylating activity of AtMPK6 ( MPK6 ) has been reported [42] [41] . Elevated levels of SA , accompanied by PR-gene expression , have been reported in a double mutant of the AtPTP1 and MKP1 phosphatase genes [41] . However , no phenotype has been found in a single mutant of the AtPTP1 gene . Our finding for OsPTP1/2 is the first to report a function that is specific to plant PTPs; therefore , whether or not PTPs in other plant species have similar functions remains to be determined . In addition , this function is probably largely dependent on the pathosystem ( combination of host and pathogen ) . Why have ( rice ) plants developed this antagonistic signalling crosstalk ? Presumably , this is the mechanism behind the trade-off between the responses to biotic and abiotic stresses , which prioritizes the most life-threatening stress through the allocation of limited resources under various situations . In nature , such a system is likely to help plants to survive changing environments in a cost-efficient manner . Serious rice losses due to blast disease occurred during the cold summers of 1993 and 2003 in Japan . Other research reported that low temperatures and drought render rice plants more susceptible to blast disease [32] [33] . Moreover , high salinity conditions can breakdown the disease resistance to Fusarium and Phytophthora in tomato [48] . In our experiments , the rice plants prioritized the responses to the cold or high salinity over disease resistance , which eventually compromised the BTH-induced blast resistance . Owing to this rice response under multiple stresses , most of the rice plants died of blast disease . Considering this consequence , this trade-off mechanism does not appear to have been beneficial to plants in our experiments . Crop cultivation is often conducted under resource-rich fertile conditions , and so were our experiments . Under such conditions , the trade-off seems to be unnecessary and even harmful for plants , as well as for farmers . Therefore , down-regulating such crosstalk seems to be favourable for agriculture as long as there is no unexpected adverse side effect . Theoretically , provided that the PTPases act specifically on the SA–ABA crosstalk , there is unlikely to be any such side effects . Indeed , we have not observed any significant adverse effects on the growth of PTP-wkd plants so far . The same could also hold for the trade-offs between particular stress responses and/or between a particular stress and plant growth/yield . On the basis of this speculation , down-regulating a particular signalling crosstalk could be an important goal of crop breeding . So far , little is known about the molecules , besides OsPTP1/2 , that directly mediate the crosstalk . To date , several signaling components have been reported to mediate signalling crosstalk; however , most of them play indirect roles in the crosstalk . Modifications of such molecules would change the balance of responses to different stresses , but they are unlikely to eliminate the crosstalk , which would maximize the defence or tolerance to both stresses [22] . Currently , only a few molecules , such as BZR1 , which mediates the crosstalk between brassinosteroid- and gibberellin-signaling , as well as innate immunity [49] [50] , are known to play direct roles in the signalling crosstalk . Once researchers identify more of these molecules , it should become possible to develop multi-stress-tolerant crops without penalties on yields .
Rice plants ( Oryza sativa subsp . japonica cv . Nipponbare ) were grown in a greenhouse in soil ( Bonsol No . 2; Sumitomo Chemical Co . , Tokyo , Japan ) at 30°C/26°C ( day/night ) with a relative humidity ( RH ) of approximately 60% . Culturing of the blast fungus M . oryzae ( race 003 ) and fungal inoculations of rice plants were carried out essentially as described previously [25] , with slight modifications . Briefly , M . oryzae conidia suspended in 0 . 02% Tween20 at a density of 150 , 000/ml were sprayed onto rice leaves , which had been pretreated with chemicals in 0 . 1 × MS in 50-ml tubes sealed with surgical tape ( 3M , St . Paul , MN , USA; cat# 3530–1 ) . Detailed condition for each experiment is described below . Normal-temperature conditions: after chemical pretreatments for 1 d in a growth chamber ( 30°C , 14-h day and 26°C , 10-h night; 60% RH ) , the solutions were removed , and the leaves were spray-inoculated with fungal conidia . The leaves were incubated in a dew chamber at 24°C for 24 h , and then for a further 5 d period in a growth chamber . Low-temperature conditions: rice leaves were pretreated with chemicals for 2 d in a growth chamber ( 15°C , 14-h days and 9°C , 10-h nights; 60% RH ) . After fungal inoculations followed by incubation for 3 d under low-temperature conditions , solutions were removed , and leaves were further incubated for 5 d under normal-temperature conditions . High-salinity conditions: before detaching leaves , whole plants were soaked in 250 mM NaCl for 4 d . Other procedures are the same as those of the normal condition . Disease development was evaluated by quantifying M . oryzae 28S rDNA by qPCR [51] . Three or more biological repeats were performed for each disease resistance assay . Site-directed mutagenesis was performed with a QuickChange Multi Site-directed mutagenesis kit ( Stratagene , La Jolla , CA , USA ) according to the manufacturer’s instructions . For constructs using NanoLuc , cDNA encoding NanoLuc was subcloned from the pNL vector ( Promega , Madison , WI , USA ) . For overexpression of WT and mutant WRKY45 , cDNA amino-terminally fused with 3X myc tag was cloned into the pZH vector under the control of the maize ubiquitin promoter . For Dex-inducible MKK10-2D , OsMKK10-2D cDNA [30] was subcloned into the pINDEX vector [52] . For PTP-wkd , the 3′-untranslated regions of OsPTP1 and -2 genes were amplified using u7 ( 5′- CACCCGGGTATCCCTAAGGCAGGA-3′ ) and u8 ( 5′- AAATGATTCAGTTTAAACCTACTAACTCTCTTTAATTCCGT-3′ ) , and u9 ( 5′- TTAGTAGGTTTAAACTGAATCATTTCTATGGAACAATCAGT-3′ ) and u10 ( 5′- AGGCCTGGGTGGGCAGGAGAAGCG-3′ ) primers , respectively . Then , we performed an overlapping second PCR using the u7 and u10 primers , and the products of the first reactions . Each amplified fusion gene was cloned into pENTR/D-TOPO ( Invitrogen , Carlsbad , CA , USA ) and subsequently transferred into the pANDA vector [53] . A reporter assay in the rice leaf sheath was essentially performed as described previously [14] . The mixture of plasmids for the expression of effectors consisted of the plasmids encoding wild-type or mutant proteins amino-terminally fused with and without NanoLuc . The effectors with or without the N-terminal NanoLuc were mixed at the ratio of 1:10 . A total of 3 μg of the LUC reporter plasmid and 0 . 5 μg of the effecter plasmid mixture were used per assay . LUC and Nanoluc activities were determined using luciferin and furimazine as substrates , respectively , and their ratios ( LUC/Nanoluc ) were compared . Agrobacterium-mediated transformations of rice calli were performed as described previously [54] [55] . Plants were regenerated from transformed calli by selecting for hygromycin resistance . Chemical treatments of leaves were conducted essentially as described previously [25] . Leaf blades from rice plants at the four-leaf stage were cut into segments approximately 0 . 5 cm long and submerged in a solution containing the chemicals prepared in 0 . 002% Silwet L-77 . The leaf segments were incubated in the light at 30°C for the periods indicated . We applied ABA 1 h before SA or Dex treatments . The PTPase inhibitors were applied 1 h before the ABA treatment . Immunoblot assays were carried out essentially as described previously [15] . Proteins were extracted with 50 mM Hepes-KOH , pH 7 . 5 , containing Complete Protease Inhibitor Cocktail ( Roche Diagnostics , Mannheim , Germany ) , 1 mM PMSF , and protein phosphatase inhibitor cocktails ( phosphatase inhibitor cocktail 1 and 2; Sigma , St Louis , MO , USA ) or alternatively PhosTop ( Roche Diagnostics ) . For the treatment with PPase , the protein phosphatase inhibitor cocktails were excluded . After centrifugation , supernatants ( containing 6–20 μg protein ) were subjected to sodium dodecyl sulfate polyacrylamide gel electrophoresis ( SDS-PAGE; 10% 29:1acrylamide:bis-acrylamide ) , followed by electroblotting . In the case of the Phos-Tag SDS-PAGE , 10 μM Phos-Tag acrylamide ( Wako , Tokyo , Japan ) and 20 μM Zn ( NO3 ) 2 were included in the 7 . 5% polyacrylamide gel ( 29:1 acrylamide:bis-acrylamide ) . Other procedures were performed as described previously [36] . Immunodetection was performed using SNAPid ( Millipore , Billerica , MA , USA ) according to the manufacturer’s instructions . The antibodies used were as follows: anti-pTEpY ( Promega , cat# V8031 ) at a 1/1 , 500 dilution; anti-OsMPK6 [39] at a 1/1 , 500 dilution; anti-phosphoenolpyruvate carboxylase [56] at a 1/30 , 000 dilution; anti-pY ( Millipore , clone 4G10 platinum ) at a 1/1 , 500 dilution; anti-pT [Promega , anti-pT183 MAPK pAb ( rabbit ) ] at a 1/4 , 000 dilution; and anti-WRKY45 [15] at a 1/300 dilution . Total RNA was isolated from rice leaves treated with chemicals as described above using Trizol reagent ( Invitrogen ) . cDNA was synthesized using ReverTraAce ( Toyobo , Tokyo , Japan ) . Quantitative PCR was run on a Thermal Cycler Dice TP800 system ( Takara Bio , Tokyo , Japan ) using the SYBR premix ExTaq mixture ( Takara Bio ) as described previously [9] . Sequences of primers used for RT-qPCR are listed in S1 Table . Phosphorylation and dephosphorylation assays were carried out as described previously [57] [39] , with modifications . GST-MKK10-2D and MBP-MPK6 ( WT or mutant ) were incubated in reaction buffer ( 10 mM Hepes-KOH , pH 7 . 5 , 5 mM EGTA , 20 mM MgCl2 , 1 mM DTT ) containing 0 . 5 mM ATP at 25°C for 20 min . For dephosphorylation assays , MBP-PTP1 or -2 was added and the reaction mixture was incubated for an additional 20 min . For WRKY45 phosphorylation activity assays , the same reaction mixtures were pre-incubated for 20 min , and reactions were initiated by adding MBP-WRKY45 and 37 kBq [γ-32P]ATP . The mixtures were incubated for an additional 20 min , and then terminated by adding Laemmle’s sample buffer and boiling . Labelled proteins were analysed by SDS-PAGE . Coomassie brilliant blue staining was performed as loading controls . WRKY45: Os05g0322900 , LOC_Os05g25770 OsMKK10-2: Os03g0225100 , LOC_Os03g12390 . 1 OsMPK6: Os06g0154500 , LOC_Os06g06090 OsPTP1: Os12g0174800 , LOC_Os12g07590 OsPTP2: Os11g0180200 , LOC_Os11g07850 . 1 | Chemical defence inducers make plants resistant to diseases such as rice blast . However , plants sometimes become more pathogen susceptible under abiotic stresses even in their presence . Because such regulation prioritizes the responses to the most life-threatening stress , it could be necessary for plants to survive in nature . However , it seems dispensable or even disadvantageous for crops cultivated under fertile conditions . Here , we show the molecular mechanism underlying one of such phenomena in rice . WRKY45 is a central transcription factor that regulates strong defence signalling mediated by salicylic acid . We found that WRKY45 is activated through phosphorylation by a protein kinase , OsMPK6 , which is activated by dual phosphorylation in response to the defence signalling . We also found that OsMPK6 can be inactivated by tyrosine dephosphorylation in response to abiotic stresses such as low temperature and high salinity probably mediated by abscisic acid , leading to reduction of WRKY45-dependent disease resistance . Moreover , we found that specific tyrosine protein phosphatases dephosphorylate/inactivate OsMPK6 in response to abiotic stresses . Knockdown of their genes rendered rice plants resistant against blast disease even under the abiotic stresses , pointing to the way to further improve rice . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [] | 2015 | Abiotic Stresses Antagonize the Rice Defence Pathway through the Tyrosine-Dephosphorylation of OsMPK6 |
During neurogenesis , transcription factors combinatorially specify neuronal fates and then differentiate subtype identities by inducing subtype-specific gene expression profiles . But how is neuronal subtype identity maintained in mature neurons ? Modeling this question in two Drosophila neuronal subtypes ( Tv1 and Tv4 ) , we test whether the subtype transcription factor networks that direct differentiation during development are required persistently for long-term maintenance of subtype identity . By conditional transcription factor knockdown in adult Tv neurons after normal development , we find that most transcription factors within the Tv1/Tv4 subtype transcription networks are indeed required to maintain Tv1/Tv4 subtype-specific gene expression in adults . Thus , gene expression profiles are not simply “locked-in , ” but must be actively maintained by persistent developmental transcription factor networks . We also examined the cross-regulatory relationships between all transcription factors that persisted in adult Tv1/Tv4 neurons . We show that certain critical cross-regulatory relationships that had existed between these transcription factors during development were no longer present in the mature adult neuron . This points to key differences between developmental and maintenance transcriptional regulatory networks in individual neurons . Together , our results provide novel insight showing that the maintenance of subtype identity is an active process underpinned by persistently active , combinatorially-acting , developmental transcription factors . These findings have implications for understanding the maintenance of all long-lived cell types and the functional degeneration of neurons in the aging brain .
Tremendous progress has been made in delineating the transcriptional mechanisms that diversify neuronal subtype identities during development . Spatiotemporally-patterned transcription factor cascades act within increasingly diversified progenitor populations to specify postmitotic neuron subtype fate . Within those postmitotic neurons , subtype-specific sets of transcription factors act combinatorially to differentiate subtype identity by initiating expression of the genes that define subtype form and function [1] , [2] , [3] . These so-called terminal differentiation genes include subtype-specific neuropeptides , neurotransmitter enzymes and ion channels [4] . Developmental transcriptional cascades are progressive and typically nonlinear; many transcription factors act at multiple levels , they exhibit extensive cross-regulation , and their expression undergoes considerable refinement in developing postmitotic neurons [1] , [5] , [6] , [7] . Here , we apply the term ‘subtype transcription network’ to refer to the transcription factors that direct subtype specification and differentiation , their cross-regulatory relationships ( or configuration ) , and the manner in which they direct the expression of subtype-specific sets of terminal differentiation genes . After subtype-specific gene expression profiles are established by differentiation , continued neuronal function throughout life depends upon their maintenance of subtype gene expression profiles . However , we currently have only a rudimentary understanding of the mechanisms of long-term , subtype-specific gene maintenance . Two extreme models would posit that subtype identity is either actively maintained by persistent subtype transcription network activity or passively maintained by for example stabilized chromatin structure , independent of a subtype transcription network . Here , we test the active model in two Drosophila neuronal subtypes to address the following largely unanswered questions: Do developmental subtype transcription networks persist in adult neurons or are they dispensed with ? Are they required to maintain the expression of subtype-specific sets of terminal differentiation genes ? If they are required , does maintenance of terminal differentiation genes require the same complex combinatorial codes of transcription factors as for their initiation , or a simplified code involving fewer transcription factors ? Finally , do persisting developmental transcription factors retain the same cross-regulatory relationships that regulated their expression during development ? We model these questions in Drosophila Tv1 and Tv4 neurons . There are six clusters of Tv neurons in the Drosophila ventral nerve cord , each comprising four distinct subtypes ( Tv1–Tv4 ) . Tv1 and Tv4 express subtype-specific terminal differentiation genes , the neuropeptides Nplp1 ( Tv1 ) and FMRFa ( Tv4 ) and the neuropeptide amidase PHM ( peptidylglycine alpha-hydroxylating monooxygenase; in Tv1/Tv4 ) . Using these genes as markers for subtype-specific differentiation , previous work had revealed the elaborate subtype transcription networks that direct Tv1/4 subtype specification and differentiation [6] , [8] , [9] , [10] , [11] , [12] , [13] . The expression of FMRFa [14] , Nplp1 and PHM ( herein ) are stably maintained in Tv1/4 neurons throughout Drosophila life . Thus , our detailed understanding of their subtype-specific initiation in the embryo provides an ideal background to investigate how such terminal differentiation genes , and hence subtype identity , are maintained in the adult . We previously established that persistent retrograde BMP signaling is required to initiate and maintain FMRFa in Tv4 neurons [14] . Here , we examined the adult function of Tv1 and Tv4 subtype transcription networks in the maintenance of Nplp1 , FMRFa and PHM . We further examined whether the cross-regulatory interactions observed between the Tv1/Tv4 network transcription factors during development were maintained in adults . We found that each subtype transcription network is largely retained in adult Tv neurons and is required to actively maintain subtype-specific gene expression . Thus , the combinatorial transcription codes for subtype-specific gene expression are not ‘simplified’ or dispensed with for maintenance . Further , we find that certain critical developmental cross-regulatory interactions between transcription factors are no longer utilized in adults for transcription factor maintenance . Thus , we observe a post-developmental switch to a distinct maintenance configuration between individual transcription factors . Collectively , these data provide novel insight relevant to understanding how long-lived cell types maintain their subtype identity .
We previously reported that adult Tv4 neurons maintain FMRFa , ap and eya expression , as well as retrograde BMP-signaling . We also demonstrated that FMRFa maintenance in adult Tv4 neurons requires persistent retrograde BMP-signaling [14] . Here , we determined the adult expression of the other transcriptional regulators implicated in Tv1 and Tv4 development ( Figure 1C–1J ) . We found that adult Tv4 neurons retained ap , eya , dimm , dac and sqz ( Figure 1C–1E , 1J ) , but no longer expressed grh , cas or nab ( Figure 1G–1I ) . Additionally , adult Tv1 neurons retained expression of Nplp1 , as well as ap , eya , dimm and col ( Figure 1C , 1D , 1F ) . Previously , cas expression was shown to be lost in all Tv neurons prior to neuropeptide initiation [9] , and here we find that it does not become re-expressed in adult Tv1 or Tv4 neurons ( Figure 1H ) . These data are summarized ( Figure 1K , 1L ) . Further analysis found that the adult complement of transcription factors was established by the start of the L1 larva stage ( Figure S1 ) , shortly after Tv1/4 terminal differentiation . Which transcription factors persist in adult neurons is intriguing . All those previously implicated in the postmitotic differentiation of Tv1 and/or Tv4 neurons persist . In contrast , transcription factors that act within the neuroblast and newborn postmitotic neuron to specify the fate of Tv1 ( cas ) or Tv4 ( cas , col , grh , nab ) neurons are not retained in the adult . The exceptions to this are col and sqz . Both are implicated in Tv subtype specification , but it is notable that both transcription factors have also been implicated , by loss and gain of function genetics , as part of the combinatorial transcription factor codes that initiate Nplp1 or FMRFa expression [6] , [9] , [12] . Thus , we find that only regulators implicated in postmitotic subtype differentiation are maintained in adult neurons . To test the function of each transcription factor in maintaining terminal differentiation gene expression in Tv1 and Tv4 neuronal subtypes , we used apGAL4 ( except where noted ) to express UAS-dsRNAi transgenes ( abbreviated to dsRNAi ) targeted to each transcription factor . We also overexpressed UAS-Dicer2 in all experiments to enhance dsRNAi efficacy [21] . To selectively induce dsRNAi in adults , we utilized the TARGET system wherein a temperature-sensitive GAL4-repressor GAL80 ( GAL80TS ) controls the activity of GAL4 [22] . Flies were raised at 18°C to allow functional GAL80TS to repress GAL4 activity throughout development . Then at adult day 1 ( A1 ) , flies were switched to 29°C and kept at that temperature for the remainder of each experiment . At this temperature , GAL80TS becomes dysfunctional and thus GAL4 is allowed to induce dsRNAi expression ( Figure 2A ) [14] . We induced eyadsRNAi expression at adult day 1 ( A1 ) and quantified FMRFa transcript levels by fluorescent in situ hybridization , relative to the mean of controls . In adults , eyadsRNAi dramatically reduced FMRFa transcript to 19 . 7±4 . 4% of control ( p<0 . 0001 ) by adult day A15 ( Figure 2B–2D ) . We observed a similar downregulation in immunoreactivity to the mature amidated FMRFa peptide ( Table S1 ) . To confirm dsRNAi specificity , we tested for enhancement of the FMRFa phenotype when expressing eyadsRNAi in an eya heterozygous background . Indeed , we found that this further reduced FMRFa transcript to 4 . 6±0 . 5% of control by A15 ( p<0 . 0001 to eyadsRNAi alone or eya heterozygosity alone ) ( Figure 2D ) . Eya immunoreactivity was eliminated in all cases ( Table S4 ) . Previous studies demonstrated that FMRFa is severely downregulated in eya mutants by late embryogenesis [11] . Our data now show that FMRFa maintains this critical dependence on eya in adults . We next tested the role of apterous ( ap ) . As apGAL4 is a strong hypomorphic ap allele [23] , we used dacGAL4 to express apdsRNAi in wildtype and heterozygous ap backgrounds . We observed a significant downregulation of FMRFa transcript when expressing apdsRNAi in ap heterozygotes , falling to 66 . 8±4 . 2% of control ( p<0 . 0001 from control or ap heterozygote alone , and p<0 . 005 from apdsRNAi alone ) . No loss of FMRFa was observed in either ap heterozygotes or apdsRNAi alone ( Figure 2H–2J ) . Similar results were obtained for downregulation of the mature FMRFa amidated peptide ( Table S1 ) . We also examined FMRFa expression using the strong apGAL4 driver to overexpress apdsRNAi , and observed a significant reduction of mature amidated FMRFa peptide to 40 . 0±2 . 2% of control by A20 ( w1118 control n = 40 , apdsRNAi n = 38; p<0 . 0001 ) . Finally , we also confirmed that an apdsRNAi targeting different ap sequences also significantly downregulated FMRFa ( Table S2 ) . The downregulation of FMRFa that we observed in adults is comparable to that reported for embryonic ap null mutants [12] , [17] . Thus , we conclude that ap maintains a persistent role in FMRFa regulation . We were unable to determine the extent of Ap knockdown by either apdsRNAi transgene due to a lack of suitable Ap-specific antibodies . Therefore , we tested apdsRNAi efficacy by examining another ap phenotype . In ap mutants , the wings fail to form [24] . We found that expression of apdsRNAi in the developing wing , using apGAL4 , could precisely phenocopy this ap phenotype ( Figure S2 ) . Thus , we conclude that apdsRNAi is specific and highly effective . However , as we could not directly quantify Ap downregulation in Tv neurons , we cannot formally discount the possibility that FMRFa would be further downregulated if Ap were entirely eliminated . To test the role of sqz in adult Tv4 neurons , we expressed sqzdsRNAi at A1 and observed a partial downregulation of FMRFa expression in sqz heterozygotes to 53 . 0±4 . 8% of control ( p<0 . 0001 from control , sqz heterozygote or sqzdsRNAi alone ) ( Figure 2E–2G ) . Similar results were obtained for immunoreactivity to the mature amidated FMRFa peptide ( Table S1 ) . Previous reports established that FMRFa is partially downregulated in embryonic sqz mutants [12] . Thus , our data indicate that sqz maintains its partial requirement for FMRFa expression . Due to ubiquitous but weak Sqz expression in the thoracic nerve cord , we were not able to adequately quantitate Sqz downregulation in Tv neurons . Thus , we do not discount the possibility that Sqz may not have been entirely eliminated , and therefore we may be underestimating its effect on FMRFa expression . Taken together , our data demonstrate that eya , ap and sqz are required to maintain wildtype FMRFa levels in the adult . We induced dimmdsRNAi at A1 and found that immunoreactivity to the mature amidated FMRFa peptide was rapidly and profoundly reduced by dimmdsRNAi to 24 . 0±3 . 2% of control by A10 ( p<0 . 0001 ) , and this was enhanced to 9 . 8±1 . 6% of control in dimm heterozygotes ( p<0 . 0001 to control and dimm heterozygotes , p<0 . 001 to dimmdsRNAi alone ) ( Figure 3A–3C , Figure S3 ) . Immunoreactivity to Dimm demonstrated that it had been eliminated ( Table S4 ) . In contrast , FMRFa transcript in adults was downregulated in dimm heterozygotes to 67 . 1±2 . 9% of control at A20 ( p<0 . 0001 ) ( Figure 3D ) . Similar effects were observed using a dimmdsRNAi that targets different dimm sequences ( Table S2 ) . It is notable that downregulation of the transcript was only observed after 20 days of dimmdsRNAi induction but the peptide was profoundly reduced after only 10 days of induction . In late Stage 17 embryonic dimm mutants , immunoreactivity to the mature amidated FMRFa peptide was profoundly reduced , but the extent to which FMRFa transcript was affected had not been quantified [10] , [18] . Here , we find that FMRFa transcript was only modestly downregulated in late Stage 17 embryonic dimm mutants to 71 . 6±3 . 9% of controls ( wild type control n = 54 , dimm mutant n = 34 ( p<0 . 0001 ) ) . Thus , we conclude that dimm retains its role in the initiation and maintenance of both FMRFa transcript and mature peptide . Why is the mature peptide more responsive to dimmdsRNAi than is the transcript ? We postulated that this was due to dimm's regulation of proprotein convertases and peptide amidases in secretory neurons , both of which are required to process the FMRFa prepropeptide into amidated neuropeptides [10] , [20] , [25] . We tested this in adults by examining expression of peptidylglycine α-hydroxylating monooxygenase after dimmdsRNAi induction ( PHM ) . Confirming our hypothesis , dimmdsRNAi entirely eliminated PHM immunoreactivity in Tv4 neurons ( Figure 3E , 3F ) . Thus , the maintenance of neuropeptide-processing enzyme expression and biosynthesis of the amidated FMRFa peptide is highly dependent upon persistent dimm function in adult Tv neurons . Previous studies found that FMRFa was only modestly downregulated in dac mutants during development [11] . In confirmation , we found here that in L1 larvae , FMRFa immunofluorescence per Tv4 neuron was 68 . 3±8 . 5% of control ( Figure 3J; P<0 . 02 ) . We tested dac function in adults and found that dacdsRNAi dramatically downregulated FMRFa immunoreactivity in adults to 14 . 8±2 . 5% of controls , as early as A10 ( p<0 . 0001 ) ( Figure 3G–3I ) . Correspondingly , FMRFa transcript was reduced to 24 . 9±4 . 2% of controls ( p<0 . 0001 to controls ) , and this was enhanced in dac heterozygotes to 6 . 5±0 . 8% ( p<0 . 001 to dacdsRNAi alone ) ( Figure S3 ) . Notably , by A15 , FMRFa peptide and transcript were entirely eliminated ( not shown ) . In all cases , we found that Dac immunoreactivity was eliminated ( Table S4 ) . Moreover , similar effects were observed using a dacdsRNAi that targets different dac sequences ( Figure S2 ) . Thus , dac appears to be unique amongst the Tv4 subtype transcription network factors in that it assumes an increasingly essential role in maintenance compared to developmental initiation . The Tv4 subtype transcription network acts through hierarchical and feedforward transcription factor activity , which we refer to here as the network's configuration ( summarized in Figure 1A , 1B ) [6] , [10] . Initiation of grh , dac , sqz and col requires transient cas activity [6] . Expression of ap and eya requires transient col expression [9] . The induction of dimm then requires eya , ap and grh [6] , [9] , [10] . BMP signaling is dependent upon eya [11] . Finally , ap , eya , dimm , dac , sqz , grh and BMP signaling are all required for FMRFa initiation [6] , [11] , [12] , [17] , [18] . This cascade represents a progressive and dynamic set of interactions during Tv4 neuron specification and differentiation . However , for long-term maintenance of subtype gene expression , the subtype transcription network presumably resolves into a stable configuration . As grh , cas and col are lost by early L1 ( Figure S1 ) , network configuration must change as the remaining transcription factors become independent of those that initiated their expression . However , we wished to ask whether the developmental cross-regulatory interactions between the persisting transcription factors are retained in the adult to help stabilize the network post-developmentally . Thus , we examined the configuration ( cross-regulatory interactions ) of all Tv4 subtype transcription network factors . BMP signaling in embryonic Tv4 neurons is dramatically reduced in eya mutants [11] . We expressed eyadsRNAi in adults until A15 and found that nuclear pMad , an indicator of BMP activity [12] , was significantly downregulated to 47 . 9%±2 . 9 of control ( Figure 4A , 4B , 4D ) . As BMP signaling is required for FMRFa expression in embryos and adults , we asked whether eya-dependence of FMRFa in adults is due to reduced BMP signaling . To do this , we simultaneously expressed eyadsRNAi and restored BMP signaling , using constitutively-activated type I BMP-receptors , thickveins and saxophone . Even though nuclear pMad was robustly activated in all Tv neurons , eyadsRNAi-induced FMRFa downregulation was not rescued ( Figure 4C , 4E ) . Thus , in adults , eya independently maintains both BMP signaling and FMRFa expression . In the embryo , initiation of dimm expression in Tv4 is absolutely dependent upon eya and grh [6] , [9] and partially dependent upon ap [10] . As grh is not expressed in adult Tv4 neurons , dimm maintenance must become independent of grh . However , as eya and ap are retained , we tested their role in dimm maintenance . We expressed apdsRNAi in adults until A20 using apGAL4 ( a strong hypomorphic allele ) and found that Dimm immunoreactivity was significantly downregulated to 58 . 5%±6 . 6 of control ( Figure 4I–4K ) . In contrast , we found that Dimm expression in adult Tv4 neurons was entirely unaffected by eyadsRNAi ( Figure 4F–4H ) . These data indicate that Dimm becomes independent of eya in adult Tv4 neurons , even though eya expression persists in adults and eya is absolutely required for the initiation of dimm expression [9] . We conclude that maintenance of dimm remains dependent on ap but becomes independent of eya and grh post-developmentally . We also examined all other potential cross-regulatory relationships within the Tv4 subtype transcription network ( Table S3 ) , but found no instances of a transcription factor requiring the presence of another for its maintenance . Is the maintained role found for the Tv4 subtype transcription network common to other subtype transcription networks ? To determine this , we examined the output and configuration of the adult Tv1 subtype transcription network . In the neuroblast lineage that gives rise to the Tv1 neuron , cas induces col . Upon birth of the postmitotic Tv1 neuron col initiates eya and ap expression . Initiation of dimm is then absolutely dependent on each of col , ap and eya . Then all four regulators are required for Nplp1 initiation [6] , [9] . In adult Tv1 neurons , col , eya , ap and dimm were maintained ( Figure 1 ) . Induction of coldsRNAi [9] at A1 reduced Nplp1 to 10 . 7±1 . 7% of control by A15 ( Figure 5A–5C ) . We also found that induction of apdsRNAi , eyadsRNAi or dimmdsRNAi significantly reduced Nplp1 expression levels to 22 . 7±2 . 7% , 45 . 3±4 . 2% and 19 . 0±2 . 4% of control , respectively ( all p<0 . 0001 to control ) ( Figure 5G , 5J , 5O ) . We verified that Col , Eya and Dimm immunoreactivity in Tv1 were eliminated by their respective dsRNAi ( Table S4 ) . In addition , we found that dimmdsRNAi also eliminated PHM expression in Tv1 neurons ( Figure 5N ) . Thus , the Tv1 subtype transcription network is required to maintain the expression of Tv1-specific terminal differentiation gene expression . Next , we examined the configuration of the adult Tv1 subtype transcription network . Intriguingly , even through col is essential for eya , ap and dimm expression in the embryo , coldsRNAi had no effect on dimm , ap , or eya expression in Tv1 ( Figure 5D–5F ) . In contrast , expression of either apdsRNAi or eyadsRNAi led to a significant downregulation of Dimm levels in Tv1 to 52 . 4±5 . 1% and 42 . 9±3 . 7% ( Figure 5H , 5K ) . These data are highly intriguing . The loss of col-dependence of ap , eya and dimm on col was unexpected , notably because Nplp1 retains its col-dependence . Moreover , it is intriguing that dimm retains its eya-dependence in adult Tv1 neurons , but not in adult Tv4 neurons . Thus , the cross-regulatory interactions between persistent transcription factors can be significantly altered after the process of differentiation . Genetic studies in the embryo had established that col , ap , eya and dimm act non-redundantly to initiate Nplp1 expression during development [6] . We tested whether these transcription factors also act non-redundantly in the adult . As coldsRNAi dramatically downregulated Nplp1 but did not affect ap , eya or dimm , we conclude that col acts non-redundantly in this case . However , dimm is dependent on both ap and eya in Tv1 . Therefore , to test for redundancy between these transcription factors , we restored dimm ( UAS-dimm ) in either apdsRNAi or eyadsRNAi backgrounds . UAS-dimm expression was found to only partially rescue Nplp1 expression in an apdsRNAi background , from 22 . 7±2 . 7% to 57 . 0±5 . 4% ( p<0 . 0001 compared to apdsRNAi and also w1118 control ) ( Figure 5I ) . However , UAS-dimm expression failed to rescue Nplp1 expression in an eyadsRNAi background ( Figure 5L ) . Thus , as during development , all regulators are required combinatorially for normal Nplp1 expression in adult Tv1 neurons . During development , late-acting subtype transcription networks can override earlier-acting transcriptional codes to dominantly activate ectopic expression of their target genes and/or subtype identity [6] . For example , in embryos , overexpression of col in all Tv neurons ectopically activated Nplp1 in Tv4 , presumably reconstituting the col/ap/eya/dimm Tv1 subtype transcription network ( Baumgardt et al . , 2007 ) . Interestingly , this did not disrupt native FMRFa expression in Tv4 neurons , nor its known subtype transcription network profile . Here , we verify these data in embryos ( Figure 6A , 6C ) . Additionally , we demonstrate here that the reciprocal subtype transcription network reconstitution , dac and BMP activation in all Tv neurons , is sufficient to initiate FMRFa expression in embryonic Tv1 neurons . Interestingly , we found that this also occurred without a concomitant disruption of Nplp1 expression in Tv1 ( Figure 6A , 6B ) . We tested whether Tv1 and Tv4 subtype transcription networks retained this capacity in adult neurons . This would prove that subtype transcription networks are capable of inducing expression of their pertinent target gene in a mature cell that never had expressed that gene . We found that activation of BMP signaling and dac in adult Tv1 neurons for 5 days robustly activated ectopic FMRFa expression in 100% of Tv1 neurons ( Figure 6D , 6E ) , without affecting Nplp1 in Tv1 . We next ectopically expressed col in adult Tv neurons , but this failed to induce ectopic Nplp1 expression ( n = 21 Tv4 neurons ) ( data not shown ) . However , co-expression of col and dimm in adult Tv neurons was sufficient to trigger ectopic Nplp1 expression in 100% of Tv4 neurons ( n = 42 ) ( Figure 6D , 6F ) . These data show that subtype transcription networks are sufficient to initiate pertinent target gene expression , even in adult neurons that had never expressed the gene .
Our data provide novel insight supporting the view of Blau and Baltimore [26] that cellular differentiation is a persistent process that requires active maintenance , rather than being passively ‘locked-in’ or unalterable . We make two primary findings regarding the long-term maintenance of neuronal identity . First , we find that all known developmental transcription factors acting in postmitotic Tv1 and Tv4 neurons to initiate the expression of subtype terminal differentiation genes are then persistently required to maintain their expression . Second , we found that key developmental cross-regulatory relationships that initiated the expression of certain transcription factors were no longer required for their maintained expression in adults . Notably , we found this to be the case even between transcription factors whose expression persists in adults . In this study , all transcription factors implicated in the initiation of subtype-specific neuropeptide expression in Tv1 and Tv4 neurons were found to maintain subtype terminal differentiation gene expression in adults ( summarized in Figure 7 ) . In Tv1 , col , eya , ap and dimm are required for Nplp1initiation during development ( Figure 1A ) . In this study , knockdown of each transcription factor in adult Tv1 neurons was shown to dramatically downregulate Nplp1 . In Tv4 neurons , FMRFa initiation during development requires eya , ap , sqz , dac , dimm and retrograde BMP signaling ( Figure 1B ) . Together with our previous work showing that BMP signaling maintains FMRFa expression in adults [14] , we now demonstrate that all six regulatory inputs are required for FMRFa maintenance . Most transcription factors , except for dac , also retained their relative regulatory input for FMRFa and Nplp1 expression . In addition , individual transcription factors also retained their developmental subroutines . For example , as found during development [10] , [11] , [20] , dimm was required in adults to maintain PHM ( independently of other regulators ) and FMRFa/Nplp1 expression ( combinatorially with other regulators ) . The few genetic studies that test a persistent role for developmental transcription factors support their role in initiating and maintaining terminal differentiation gene expression . In C . elegans , where just one or two transcription factors initiate most neuronal subtype-specific terminal differentiation genes , they then also appear to maintain their target terminal differentiation genes . In ASE and dopaminergic neurons respectively , CHE-1 and AST-1 initiate and maintain expression of pertinent subtype-specific terminal differentiation genes [27] , [28] . In vertebrate neurons , where there is increased complexity in the combinatorial activity of transcription factors in subtype-specific gene expression , certain transcription factors have been demonstrated to be required for maintenance of subtype identity . These are Hand2 that initiates and maintains tyrosine hydroxylase and dopa β-hydroxylase expression in mouse sympathetic neurons [29] , Pet-1 , Gata3 and Lmx1b for serotonergic marker expression in mouse serotonergic neurons [30] , [31] , and Nurr1 for dopaminergic marker expression in murine dopaminergic neurons [32] . However , while these studies confirm a role for certain developmental transcription factors in subtype maintenance , it had remained unclear whether the elaborate developmental subtype transcription networks , that mediate neuronal differentiation in Drosophila and vertebrates , are retained in their entirety for maintenance , or whether they become greatly simplified . Our analysis of all known subtype transcription network factors in Tv1 and Tv4 neurons now indicates that the majority of a developmental subtype transcription network is indeed retained and required for maintenance . Why would an entire network of transcription factors be required to maintain subtype-specific gene expression ? The combinatorial nature of subtype-specific gene expression entails cooperative transcription factor binding at clustered cognate DNA sequences and/or synergism in their activation of transcription . In such cases , our data would indicate that this is not dispensed with for maintaining terminal differentiation gene expression in mature neurons . How the transcription factors of the subtype transcription networks are maintained is less well understood . An elegant model has emerged from studies in C . elegans , wherein transcription factors stably auto-maintain their own expression and can then maintain the expression of subtype terminal differentiation genes [33] . The transcription factor CHE-1 is a key transcription factor that initiates and maintains subtype identity in ASE neurons . CHE-1 binds to a cognate DNA sequence motif ( the ASE motif ) in most terminal differentiation genes expressed in ASE neurons , as well as in its own cis-regulatory region . Notably , a promoter fusion of the che-1 transcription factor failed to express in che-1 mutants , indicative of CHE-1 autoregulation [27] , [34] . Similar observations were made for AST-1 [28] , and for the cooperatively-acting TTX-3 and CEH-10 transcription factors in AIY neurons [35] . Thus , subtype maintenance in C . elegans is anchored by auto-maintenance of the transcription factors that initiate and maintain terminal differentiation gene expression . In contrast , all available evidence in Tv1 and Tv4 neurons fails to support such an autoregulatory mechanism . An ap reporter ( apC-τ-lacZ ) is expressed normally in ap mutants [12] , [36] , and in this study apdsRNAi was not found to alter apGAL4 reporter activity ( Figure 2I ) . Moreover , col transcription was unaffected in col mutants that express a non-functional Col protein [9] . This leaves unresolved the question of how the majority of the transcription factors are stably maintained . For transcription factors that are initiated by transiently expressed inputs , a shift to distinct maintenance mechanisms have been invoked and in certain cases shown [35] . In this study , this was found for the loss of cas expression in Tv1 ( required for col initiation ) and the loss of cas , col and grh in Tv4 ( required for eya , ap , dimm , sqz , dac initiation ) . However , we were surprised to find that the cross-regulatory relationships between persistently-expressed transcription factors were also significantly altered in adults . Notably , eya initiated but did not maintain dimm in Tv4 . In Tv1 , col initiated but did not maintain eya , ap or dimm . This was particularly unexpected as eya remained critical for FMRFa maintenance and col remained critical for Nplp1 maintenance . Indeed , although we tested for cross-regulatory interactions between all transcription factors in both the Tv1 and Tv4 subtype transcription networks , only Dimm was found to remain dependent upon its developmental input; Eya and Ap in Tv1 as well as Ap in Tv4 . However , even in this case , the regulation of Dimm was changed; it no longer required eya in Tv4 , and in Tv1 it no longer required col , in spite of the fact that both col and eya are retained in these neurons . We anticipate that such changes in transcription factor cross-regulatory relationships will be found in other Drosophila and vertebrate neurons , which exhibit high complexity in their subtype transcription networks [1] , [5] . Indeed , recent evidence has found that in murine serotonergic neurons , the initiation of Pet-1 requires Lmx-1b , but ablation of Lmx-1b in adults did not perturb the maintenance of Pet-1 expression [31] . We are pursuing the potential role of autoregulation for the other factors in the Tv1/Tv4 subtype transcription networks . However , we consider there to be three additional , potentially overlapping , models for subtype transcription network maintenance . First , regulators may act increasingly redundantly upon one another . Second , unknown regulators may become increasingly sufficient for transcription factor maintenance . Third , transcription factors may be maintained by dedicated maintenance mechanisms , as has been shown for the role of trithorax group genes in the maintenance of Hox genes and Engrailed [37] , [38] . Moreover , chromatin modification is undoubtedly involved and likely required to maintain high-level transcription of Tv transcription factors as well as FMRFa , Nplp1 and PHM . However , the extent to which these are instructive as opposed to permissive has yet to be established [39] . In this light , it is intriguing that MYST-HAT complexes , in addition to the subtype transcription factors Che-1 and Die-1 , are required for maintenance of ASE-Left subtype identity in C . elegans [31] . Taken together , our studies have identified two apparent types of maintenance mechanism that are operational in adult neurons . On one hand , there are sets of genes that are maintained by their initiating set of transcription factors . These include the terminal differentiation genes and the transcription factor dimm . On the other , most transcription factors appear to no longer require regulatory input from their initiating transcription factor ( s ) . Further work will be required to better understand whether these differences represent truly distinct modes of gene maintenance or reflect the existence of yet unidentified regulatory inputs onto these transcription factors . One issue to consider here is that the expression of certain terminal differentiation genes in neurons , but perhaps not subtype transcription factors , can be plastic throughout life , with changes commonly occurring in response to a developmental switch or physiological stimulus [40] , [41] , [42] . Thus , terminal differentiation genes may retain complex transcriptional control in order to remain responsive to change . It is notable , however , that FMRFa , Nplp1 and PHM appear to be stably expressed at high levels in Tv1/4 neurons , and we have not found any conditions that alter their expression throughout life . Thus , we consider these to be stable terminal differentiation genes akin to serotonergic or dopaminergic markers in their respective neurons that define those cells' functional identity and , where tested , are actively maintained by their developmental inputs [30] , [32] . Tv1/4 neurons undoubtedly express a battery of terminal differentiation genes , and sets of unknown transcription factors are likely required for their subtype-specific expression . We consider subtype transcription networks to encompass all regulators required for differentiating the expression of all subtype-specific terminal differentiation genes . Further , we view differentiation of subtype identity as the completion of a multitude of distinct gene regulatory events in which each gene is regulated by a subset of the overall subtype transcription network . As highly restricted terminal differentiation genes expressed in Tv1 and Tv4 neurons , we believe that Nplp1 , FMRFa and PHM provide a suitable model for the maintenance of overall identity , with the understanding that other unknown terminal differentiation genes expressed in Tv1 and Tv4 may not be perturbed by knockdown of the transcription factors tested herein . In the future , it will be important to incorporate a more comprehensive list of regulators and terminal differentiation genes for each neuronal subtype . However , we believe that the principles uncovered here for FMRFa , Nplp1 and PHM maintenance will hold for other terminal differentiation genes . Finally , we propose that the active mechanisms utilized for maintenance of subtype differentiation represent an Achilles heel that renders long-lived neurons susceptible to degenerative disorders . Nurr1 ablation in adult mDA neurons reduced dopaminergic markers and promoted cell death [32] . Notably , Nurr1 mutation is associated with Parkinson's disease [43] , and its downregulation is observed in Parkinson's disease mDA neurons [44] . Adult mDA are also susceptible to degeneration in foxa2 heterozygotes , another regulator of mDA neuron differentiation that is maintained in adult mDA neurons [45] . Studies in other long-lived cell types draw similar conclusions . Adult conditional knockout of Pdx1 reduced insulin and β-cell mass [46] , [47] and , importantly , heterozygosity for Pdx1 leads to a rare monogenic form of non-immune diabetes , MODY4 [48] . Similarly , NeuroD1 haploinsufficiency is linked to MODY6 [49] and adult ablation of NeuroD in β-islet cells results in β-cell dysfunction and diabetes [50] . These data , together with our results here , underscore the need to further explore the transcriptional networks that actively maintain subtype identity , and hence the function , of adult and aging cells .
Flies were maintained on standard cornmeal food and maintained at stable temperatures in environment rooms set at 70% humidity at 18°C , 25°C or 29°C . apterousmd544 ( referred to as apGAL4 ) ; apP44; sqzIE; sqzGAL4; UAS-thickveins activated ( UAS-tkvA ) ; UAS-saxophone activated ( UAS-saxA ) [12]; dacGAL4 [51]; rev4; UAS-dimm [18]; dac3 [52]; ; eyaCliIID [53]; grhGAL4 [6] . tubP>GAL80TS ( temperature-sensitive GAL80 under the control of the Drosophila tubulin 84B promoter ) ( McGuire et al . , 2003 ) ; UAS-nEGFP ( nuclear localized EGFP ) ; UAS-dicer2 [21] . Strains used for primary data: UAS-col#24E [9]; UAS-apdsRNAi 8376R-2; UAS-dacdsRNAi 4952R-2; UAS-sqzdsRNAi 5557R-2 ( NIG-FLY ) ; UAS-dimmdsRNAi GD44470; UAS-eyadsRNAi GD43911 ( VDRC ) . Strains used secondarily to verify data: UAS-dimmdsRNAi KK103356; UAS-eyadsRNAi 108071KK ( VDRC ) ; UAS-eyadsRNAi JF03160; UAS-dacdsRNAi JF02322 ( TRiP ) . Flies for TARGET-mediated transgene induction were generated by crossing utility flies ( UAS-dicer2/UAS-dicer2; apGal4; tubP>GAL80TS , UAS-nEGFP/SM6-TM6 , Tb ) or ( UAS-dicer/UAS-dicer; dacGal4; tubP>GAL80TS , UAS-nEGFP/SM6-TM6 , Tb ) to UAS-dsRNAi ( experimental group ) or w1118 flies ( control ) . Experiments were performed on resulting progeny bearing appropriate genotypes ( screened by loss of SM6-TM6 , Tb balancer chromosome ) . All experimental and control flies were raised at 18°C until eclosion ( hatching from the pupal case ) . On adult day 1 ( A1 ) , flies were switched to 29°C for the duration of the induction period indicated . Throughout the text , we present dsRNAi data for the induction period at which we observe the maximal phenotype of Nplp1 and FMRFa expression . Further details for each dsRNAi are provided in the text . Standard in situ and immunohistochemistry protocols were carried out as described [14] . All tissues compared for fluorescence intensity were processed at the same time using the same aliquots of all solutions under the same conditions . They were then mounted on the same slide and confocal settings were calibrated to control staining levels . All images acquired on an Olympus FV1000 confocal microscope . Fluorescent intensity of individual neurons was measured using Image J ( US National Institutes of Health ) . The mean pixel intensity for each neuron was measured from compressed Z-slices , and corrected for background fluorescence . Analysis was performed on every identifiable Tv1 and Tv4 neuron in segments T1 and T3 . The resulting value for each Tv neuron was then incorporated as a single datum point towards the mean intensity for each experiment . Each datum point is represented as a percentage of the mean of the w1118 control for that experiment . Data are presented as Mean ± SEM . Representative images of Tv neurons that were directly compared in figures were processed in an identical way , simultaneously , using Adobe Photoshop CS4 . Normally distributed unpaired data were compared using a two-tailed T-test assuming equal variance , to identify significant differences between means . All statistical analysis and graphs data were performed using Prism 5 software . ( Graphpad ) . | For neurons to function properly , they must establish and then maintain their unique , subtype-specific gene expression profiles . These unique gene expression profiles are established during development by networks of DNA–binding proteins , termed transcription factors ( TFs ) . However , how neurons maintain their unique gene expression profiles in the mature and aging brain is largely unknown . Recent advances in inducible genetic technologies now allow us to manipulate gene expression in adult neurons , after normal development . Applying such techniques , we examined the effect of knocking down TF expression in two adult neuronal subtypes . We show that the TF networks that establish unique gene expression profiles during development are then required to maintain them thereafter . Thus , gene expression profiles are not simply “locked-in , ” but must be actively maintained by persistent developmental TF networks . However , we found that critical cross-regulatory relationships that had existed between TFs during development were not present in the adult , even between persisting TFs . This highlights important differences between developmental and maintenance transcriptional networks in individual neurons . The dependence of subtype gene expression on active mechanisms represents a potential Achilles heel for long-lived cells , as deterioration of those active mechanisms could lead to functional degeneration of neurons with advancing age . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"developmental",
"biology",
"organism",
"development",
"genetics",
"biology",
"neuroscience",
"genetics",
"and",
"genomics"
] | 2012 | Developmental Transcriptional Networks Are Required to Maintain Neuronal Subtype Identity in the Mature Nervous System |
Trypanosoma brucei is a master of antigenic variation and immune response evasion . Utilizing a genomic repertoire of more than 1000 Variant Surface Glycoprotein-encoding genes ( VSGs ) , T . brucei can change its protein coat by “switching” from the expression of one VSG to another . Each active VSG is monoallelically expressed from only one of approximately 15 subtelomeric sites . Switching VSG expression occurs by three predominant mechanisms , arguably the most significant of which is the non-reciprocal exchange of VSG containing DNA by duplicative gene conversion ( GC ) . How T . brucei orchestrates its complex switching mechanisms remains to be elucidated . Recent work has demonstrated that an exogenous DNA break in the active site could initiate a GC based switch , yet the source of the switch-initiating DNA lesion under natural conditions is still unknown . Here we investigated the hypothesis that telomere length directly affects VSG switching . We demonstrate that telomerase deficient strains with short telomeres switch more frequently than genetically identical strains with long telomeres and that , when the telomere is short , switching preferentially occurs by GC . Our data supports the hypothesis that a short telomere at the active VSG expression site results in an increase in subtelomeric DNA breaks , which can initiate GC based switching . In addition to their significance for T . brucei and telomere biology , the findings presented here have implications for the many diverse pathogens that organize their antigenic genes in subtelomeric regions .
Trypanosoma brucei is an extracellular human pathogen with an unparalleled capacity to evade host humoral immunity . The causative agent of African sleeping sickness in humans and nagana in cattle , T . brucei is transmitted into the bloodstream of its host by a tsetse vector and can grow to densities as high as 109 cells per milliliter of blood . The parasitemia is cyclically diminished to nearly undetectable levels , to be followed by another wave of immense growth [1] . These rounds of parasitemia reflect the battle between the host immune system and the pathogen's elegant mechanisms of immune evasion . At the forefront of this battle is the T . brucei cell surface , which is primarily composed of about 107 copies of a single , densely packed Variant Surface Glycoprotein ( VSG ) [2] . VSG is highly immunogenic , yet T . brucei escapes immune recognition by switching the monoallelic expression of one VSG-encoding gene to another [1] , [3] . This process of surface antigen variation is made possible by a genomic repertoire of more than 1000 highly divergent VSG-encoding genes and pseudogenes [4] , [5] , [6] . Furthermore , existing VSGs can recombine to form novel mosaic VSGs , making the depth of the repertoire potentially limitless [7] . VSGs are encoded throughout the T . brucei genome , which consists of 11 megabase chromosomes , numerous intermediate chromosomes , and ∼100 minichromosomes [6] , [8] . Although most VSGs are found in VSG arrays within megabase chromosomes or singly on minichromosomes , they can only be transcribed by bloodstream-form parasites from one of the ∼15 Bloodstream Expression Sites ( BES ) at a time . Proper expression of VSG on the cell surface is essential for survival [1] . Each BES is composed of a collection of Expression Site Associated Genes ( ESAGs ) , a long repetitive element ( 70-bp repeats ) , and a terminal VSG encoding gene , which are transcribed from one upstream promoter and spliced into separate mRNAs before translation [4] , [9] . All known BESs are located within 60 kb from the end of the chromosome positioning the expressed VSG within >2 kb of repetitive telomeric DNA [4] . The organization of surface antigen encoding genes in subtelomeric regions is a common theme among pathogens that employ antigenic variation to evade host defenses [10] , [11] . There are three predominant mechanisms by which T . brucei VSG switching can occur: In Situ ( IS ) transcriptional VSG switching — the inactivation of one BES coupled with activation of transcription from a new BES [12] , [13] , [14] , [15] , Reciprocal Telomeric Exchange ( TE ) — a homologous recombination event between two chromosome ends resulting in the balanced transfer of a new VSG to the active BES and the previously active VSG to a silent BES [16] , [17] , and Duplicative Gene Conversion ( GC ) — a non-reciprocal transfer of a VSG containing DNA to the active BES that results in loss of the previously active VSG from the genome [18] , [19] , [20] , [21] . GC is predicted to account for the majority of VSG switching under natural conditions because it is the only mechanism that permits the expression of non-telomerically-encoded VSG genes . How each of these mechanisms is orchestrated is largely unknown . Although the mechanistic basis of each type of VSG switch is unknown , it had long been predicted that GC would be initiated by a DNA break . Recent studies have shown that the artificial induction of a DNA double-stranded break ( DSB ) proximal to the BES repetitive region upstream of the active VSG increases the frequency of VSG switching by as much as 250-fold , recapitulating the rate estimated in natural isolates [22] , [23] . As predicted , VSG switching under these conditions occurred mainly by GC [22] . Furthermore it was shown that DNA breaks accumulate in the repetitive region of the active BES [22] . However , the natural source of DNA breaks that precipitate GC remains a mystery . A proposed source of GC initiating DSBs is related to the proximity of the BES encoded VSGs to the telomere ( usually within >2 Kb ) [4] , [24] . The actively transcribed BES frequently experiences large stochastic terminal deletions , which are hypothesized to result from the very high levels of transcription at the end of the chromosome [25] , [26] . Thus it has been suggested that when the telomere of the chromosome harboring the active BES is short the DNA breaks precipitating telomeric deletions would occur upstream of the VSG , resulting in an antigenic switch [24] . This claim was further correlated to the fact that strains of T . brucei that have been recently isolated from nature , who switch at a rate of approximately 10−2–10−3 , have shorter telomeres than laboratory-adapted strains , whose rate of switching can be are 100–10 , 000-times lower [27] , [28] , [29] , [30] . In all , this suggested an inverse correlation between telomere length and the amount antigenic switching in T . brucei [24] , yet there were no data in direct support of this hypothesis . T . brucei strains with the protein component of telomerase deleted ( TERT−/− ) are unable to repair telomeric breaks ( such as those that occur frequently at the actively transcribed BES ) and undergo progressive shortening of all the telomeres in the genome by 3–6 bp/Population Doubling ( PD ) [31] . Previously it was shown that when the telomere of a TERT−/− isolate is short , the actively expressed VSG is lost over the course of several weeks and replaced by a new VSG gene , which suggested an increase in VSG switching [32] . However , because of the time span in which those experiments took place , that study could not differentiate between the equivalent possibilities of an increase in switching versus the death of short-telomere clones being replaced by a subpopulation of switchers arising at the normal in vitro frequency [32] . In this study we directly tested the proposed correlation between VSG switching and telomere length . Using updated techniques , we compared the frequency of VSG switching between strains with wild-type length or shortened telomeres at the active BES . Large populations of switched clones were analyzed to identify their mechanism of switching and determine if those with short telomeres switch by way of a preferred mechanism . The findings presented here provide experimental support for the hypothesis that telomere length directly affects the frequency of VSG switching , and answer long-standing questions about the relationship between VSG switching , telomere length , and gene conversion .
To address the effect of telomere length on T . brucei antigenic switching , we first isolated strains with various telomere lengths at their active BES ( BES1 expressing VSG427-2 [221] ) . The telomere of BES1 in a population wild-type isolates ( WT ) can range from ∼10–15 kb ( FIG . 1 . A ) . TERT−/− clones with short ( ∼1 . 5 kb ) , medium ( ∼5 . 0 kb ) , and long ( >10 kb ) BES1 telomeres were isolated and characterized by Southern blot ( FIG . 1 . B ) . The active BES telomere in a TERT−/− strain is not only prone to progressive shortening ( 3–6 bp/PD ) but also massive truncations [31] . Thus , medium- and long-telomere clones can only be handled for a minimal number of passages before they shorten ( as evidenced by the smear under the primary band in FIG . 1 . B – “Long” ) . In contrast , critically short telomeres are stabilized by an unknown , telomerase-independent , mechanism that appears to be unique to T . brucei [33] . Thus resulting in the short-telomere clone used here ( FIG . 1 . B – “Short” ) , which can be stably maintained at a length of ∼1 . 5 kb for numerous passages . Populations of trypanosomes , as with any organism , are heterogeneous and this can affect both the expressed VSG and telomere length . Furthermore , accurate determination of the frequency of VSG switching requires that the populations being compared undergo a comparable number of population doublings ( PD ) during the experiment . Therefore , WT , TERT−/− short- and long-telomere clones with similar growth rates ( FIG . S1 ) were grown from single-cells to ∼5×107 in vitro , thereby performing a modified Luria-Delbrück fluctuation analysis [34] ( FIG . 1C ) . The VSG switching frequency of the resulting populations of trypanosomes was determined using the previously published magnetic-activated cell sorting ( MACS ) depletion of the initiating VSG followed by flow cytometry quantification [22] . TERT−/− short clone populations switched their expressed VSG at a significantly ( P<0 . 0001 ) higher frequency ( 11 . 3×10−5±4 . 6 ) than both WT ( 1 . 8×10−5±1 . 1 ) and TERT−/− long-telomere clones ( 2 . 5×10−5±1 . 9 ) , which were not significantly different from each other ( P = 0 . 3136 ) ( FIG . 1D ) . The TERT−/− long-telomere clones serve as a proxy for TERT complementation in this study because a previous study showed that ectopic TERT expression results in rapid elongation of the active site telomere ( ∼160 bp/PD ) [33] , which prevents the analysis of a short-telomere TERT complemented clone by this method . The switching frequencies of the TERT−/− short-telomere clones covered a broad range of values ( 3 . 2–22×10−5 ) , which correspond to a 6- to 36-fold increase in switching compared to WT . These data might be explained by a stochastic increase in subtelomeric DNA breakage that occurs when the telomere of the active site is short . Alternatively , in the absence of TERT complementation data , the increase in switching could arise from a combinatorial effect of having a short telomere in the context of a telomerase mutant . In either case these data support the notion that subtelomeric breaks promote antigenic switching . The output of the VSG switching assay represents the proportion of the input population that is no longer expressing the starting VSG type , but this value does not account for replication of progeny , the number of VSG switching events intervening before the final measurement , or potential genotypic diversity of phenotypically identical cells in the resulting population ( FIG . 1C ) . We therefore adopted the nomenclature Observed Switching Frequency ( henceforth referred to as switching frequency or OSF ) to indicate the limitations of this measured value ( FIG . 1D ) . Hypothetically , populations that switch more often will contain a greater diversity of expressed VSGs . Thus we predicted that populations with higher OSF values would contain a larger diversity of VSG transcripts . To investigate this prediction , the MACS eluates from 12 of the 18 TERT−/− short-telomere switched populations ( FIG . 1D ) were grown to a sufficient extent for RNA extraction to identify the expressed VSGs , which were then compared with the determined OSF for that population ( FIG . 2A , y-axis values correspond directly with FIG . 1D TERT−/− short data ) . This type of analysis is only possible due to the extensive nucleotide sequence diversity of VSG-encoding genes , which allows each VSG to be accurately distinguished from another [35] . Although there was a subtle trend of populations with OSF>10 to express a greater diversity of VSGs ( ∼3–6 ) than those with an OSF<10 ( 1–2 VSGs ) ( FIG . 2B ) , we did not observe a linear relationship between the switching frequency of these populations and the depth of their VSG diversity ( FIG . 2 ) . This is highlighted by the fact that the highest OSF ( 22×10−5 ) of this set contained only one expressed VSG ( VSG427-3 [224] ) ( FIG . 2A & 2B ) . It is worth noting that the diversity of VSGs identified in TERT−/− short-telomere VSG switching assays was similar to those seen in other studies [1] . VSG427-3 ( 224 ) was the most commonly observed VSG , which was present in 10 of the 12 populations analyzed , and the sole VSG expressed in 3 of those populations ( FIG . 2A & 2C ) . The second most commonly expressed was VSG427-1 ( 060 ) in 7 populations , followed by VSG427-8 ( OD1 ) in 4 populations ( FIG . 2C ) . These data support previous in vivo and in vitro studies suggesting that certain VSGs are favored donors and that VSG switching follows a semi-predictable order [36] . In addition to the predictably expressed VSGs , we isolated two VSGs that were recently demonstrated to reside on minichromosomes ( 427-23 & 427-24 ) [22] and a VSG that had not previously been annotated ( FIG . 3 , “NOVEL” ) . The lack of a clear linear relationship between the switching frequency and the depth VSG diversity ( FIG . 2 ) could arise from two alternative biological situations: extensive propagation of an early switch event ( FIG . 1C , top ) or multiple switch events that result in expression of the same VSG ( FIG . 1C , bottom ) . As noted , the population of TERT−/− short telomere with the highest OSF ( 22×10−5 ) contained only VSG427-3 ( 224 ) ( FIG . 2A ) . Did this population arise from one or multiple VSG switch events ? To address this question , we isolated TERT−/− short-telomere trypanosomes that had switched from VSG427-2 ( expressed from BES1 ) to VSG427-3 ( originating from BES7 ) from three separate fluctuation analysis experiments . The genotypic differences among isolates from the same starting clone were then compared by PCR and Southern blot analysis ( FIG . 3 ) . Eighteen switchers expressing VSG427-3 were isolated from three populations ( 3 clones from two populations [#2 & #3] and 12 from the third [#5] ) . All 18 clones had switched to VSG427-3 by GC ( expressed from BES1 ) , as evidenced by the loss of the VSG pseudogene from BES1 ( FIG . 3A; BES1 map VSGΨ green box ) . Using PCR primers unique to ESAG1 [4] in BES1 , we determined that VSG427-3 switchers from populations #2 and #5 contained a mixture of clones that had either lost or retained BES1 ESAG1 during GC ( 2 of 3 in population 2 and 2 of 12 in population 5 had lost ESAG1 ) . Loss of BES1 ESAG1 indicates that resolution of GC occurred upstream of ESAG1 . Therefore , clones expressing VSG427-3 in these populations arose from at least two distinct VSG switching events . To further analyze the genotypes of the VSG427-3 switchers , we used a unique region in BES7 to probe a HindIII-digested large-fragment separation gel and Southern blot ( FIG . 3A ) . There are two HindIII sites in BES1 between the repeat region and ESAG4 that distinguish it from BES7 ( FIG . 3A ) . Thus , GC resolution downstream of ESAG1 results in the formation of a >10 kb fragment , upstream of ESAG1 but downstram of ESAG2 results in a >14 kb fragment , and upstream of ESAG2 but downstream of ESAG4 results in a fragment >24 kb ( FIG . 3C ) . Because BES7 is intact in the genome regardless of switching , all strains produce the >24 kb and ∼2 . 4 kb fragments . Exact prediction of product sizes for fragments resolved beyond the BES7 repeat region ( FIG . 3A , yellow boxes ) is not possible due to missing sequence data [4] . Southern blot data largely agreed with PCR data ( FIG . 3B ) , such that the VSG427-3 switchers that retained BES1 ESAG1 by PCR produced the restriction fragment associated with resection downstream of ESAG1 ( >10 kb fragment present for #2 = 2/3 , #3 = 2/3 , & #5 = 3/12 [only partial data shown] ) . However , one isolate from clone 3 retained ESAG1 but produced a fragment that was predicted to result in the loss of ESAG1 ( #3 D2 , >14 kb ) . This combination of data could arise from a recombination event of unpredicted complexity . These results further demonstrate that each population of phenotypically identical VSG427-3 switchers analyzed are composed of at least 2 genotypes , and therefore must arise from multiple genetic events . Fine mapping or sequencing the point of resection for these strains could show further switching diversity , but is not possible due to the >90% sequence similarity between BES1 and BES7 [4] . Therefore , the OSF values displayed in figure 1 under-represent genetic switching frequency by not accounting for the number of VSGs expressed in the population ( FIG . 2 ) and the genetic diversity of the switched population ( FIG . 3 ) . To determine if telomere length affects the mechanism of switching as well as frequency , populations arising from single cells clones of wild-type and TERT−/− short-telomere strains were MACs sorted , their OSF determined ( FIG . 4B – using the same methodology used to produced the data in FIG . 1 with new starting populations ) , and plated to limiting dilution . The secondary clones were then analyzed for the expression of VSG427-2 ( 221 ) by high-throughput screening ( HTS ) flow cytometry , and all cultures not expressing VSG427-2 were deemed “switched clones . ” From 12 populations of WT , we screened 1617 secondary clones , identifying 189 WT switched clones ( 12% of the post-MACS screened population ) , whereas screening of 255 secondary clones from 6 populations of TERT−/− short-telomere clones resulted in a similar number ( 188 ) of switched clones to be further analyzed ( 74% of the post-MACS screened population ) . The mechanism of switching for all secondary clones was then determined , based on three criteria: ( 1 ) resistance or sensitivity to a BES1 promoter-proximal antibiotic marker ( WT marked with hygromycin [HYG] & TERT−/− marked with blasticidin [BSD] ) , ( 2 ) the presence or absence of VSG427-2 ( 221 ) in the genome , and ( 3 ) the presence or absence of the promoter-proximal resistance marker in the genome . Thus the switch type can be counted in the following way: IS events correspond to MarkerS , 427-2 ( 221 ) + , Marker+; TE events are MarkerR , 427-2 ( 221 ) + , Marker+; GC results in MarkerR , 427-2 ( 221 ) − , Marker+; ES GC ( Expression Site Gene Conversion ) , a subtype of GC in which the entire active BES is replaced by the donor BES , are MarkerS , 427-2 ( 221 ) − , Marker−; and UD , for Undetermined , when switched clones did not fit the other criteria ( FIG . 4A & TABLE S3 ) . These data were also used to establish a minimal number of independent switch events . By creating a matrix based on the 12 known combinations of phenotype and genotype analyzed ( VSG427-3 [224] expression as determined flow cytometry [data not shown] ) , switch type [IS , TE , GC , ES GC or UD] , & BES1 VSG Ψ+/− ) against the source populations ( 12 for WT & 6 for Short ) , we determined that the 189 WT switched clones arose from at least 47 independent switchers and the 188 Short switched clones from at least 25 ( TABLE S3 ) . The higher number for WT is an artifact of the increased complexity of its matrix ( i . e . more originating populations ) . Determining the mechanism of switching in TERT−/− long-telomere clones was not possible because of their propensity to break at the active BES , which results in the rapid formation of a heterogenic population of telomere lengths . WT VSG switched clones produced similar average levels of IS ( 33% ) , GC ( 24% ) , and TE ( 37% ) . This was in contrast to the TERT−/− short-telomere VSG switchers , which showed a clear preference for GC-based switching ( 88% ) , only a small amount of IS ( 7% ) and negligible TE ( 1% ) ( FIG . 4C [“GC” shown is the sum of both GC and ES GC] & TABLE S1 , S2 , S3 ) . Switching by GC for a subset of TERT−/− short-telomere clones was further confirmed by pulsed-field gel separation of chromosomes followed by Southern blot analysis using specific VSG probes , which showed both the loss of the initiating VSG ( 427-2 ) and the duplication of the newly expressed VSG to the active BES ( FIG . 4D ) . These data , in conjunction with published data that induction of a DNA break in the active site initiates a GC based switch [22] , suggest that a short telomere at the active site can increase switch initiating subtelomeric DNA breaks . Further mathematical analysis of switched clone data suggested that the heightened level of GC could account for the overall increase in switching frequency observed when the telomere is short ( mathematical validation shown in Experimental Procedures & TABLE S3 ) . The same analytical process showed that the frequency of IS switching ( FIS = average OSF×% IS , for both WT and TERT−/− short-telomere clones ) was not significantly different between WT and TERT−/− short-telomere clones ( TABLE S3 ) . To further analyze the nature of the GC events in WT and TERT−/− short , all isolates identified as GC were assayed for the presence of BES1 VSG pseudogene ( PSD or Ψ ) in the genome , a parameter that indicates the location of GC resolution . BES1 VSG PSD was lost from TERT−/− short-telomere switchers ( 90% ) ∼30% more often than WT ( 58% ) ( FIG . 4C , lined bars “Ψ− GC+” ) . This indicates that the resolution of GC events in TERT−/− switchers often occurs farther upstream than in WT VSG switchers and suggests that the initiating lesion itself may occur farther upstream when the telomere is short . The body of isolated switcher data was further analyzed to determine if individual populations , OSF values , or isolation dates , affected the propensity to switch by a given mechanism ( TABLE S3 , FIG . S2 & S3 ) . Each initiating clone results in a population of secondary switched clones with a somewhat variable percentage of each switching mechanism . Generally , population variations did not affect the results , but there were two WT populations ( #2-5 & #2-6 ) with a large number of switchers ( 34 & 25 , respectively ) that had a high percentage of TE ( 76% & 80% , respectively ) ( FIG . S2 & TABLE S3 ) , which gave the average WT TE value of 37% . If reanalyzed , the median values for WT were IS = 37% , GC = 20% , and TE = 19% , thus reducing the overall significance of TE . Analysis of the percent switch mechanism in comparison with the switching frequency for each population showed a very subtle trend for populations with higher OSF values to contain a higher level of diversity in the mechanisms represented , which was similarly true for both WT and short-telomere isolates ( TABLE S3 ) . Due to different growth rates following limiting dilution , VSG switch isolates were selected 7 , 9 , or 11 days after initial plating . Analysis of the VSG switch mechanism by isolation date for WT and TERT−/− short-telomere clones displayed a trend of increased IS and decreased GC with time ( this pattern was more apparent in WT ) , while TE did not ( FIG . S3 ) . This suggests that isolates that switch by IS initially grow more slowly than those that switch by GC; the basis of this difference is unknown . None of the alternative analysis of the data in this section detract from the central result , namely that TERT−/− short-telomere clones switch preferentially by GC .
Telomeres are structures of DNA and protein that protect the ends of chromosomes from DNA loss and damage . Yet , somewhat counterintuitively , T . brucei is only one of the many pathogens whose critical antigenic diversity genes are organized in potentially fragile subtelomeric regions [10] , [37] . In 2007 Dreesen et al . proposed a model in which telomere length ( at the active BES ) had an inverse correlation with the frequency of VSG switching in T . brucei . The model also predicted that switching at short telomeres would occur by GC , resulting from an increase in BES internal DSBs when the telomere is short [24] . The 2007 proposal was based on a correlation between two main bodies of data: 1 ) the telomeres of laboratory-adapted strains are longer than those of strains recently isolated from nature [30] & 2 ) populations of TERT−/− short-telomere strains progressively loose the initially expressed VSG , which suggested , but did not demonstrate an increase in the frequency of VSG switching [33] . Although this model gained popular support , its predictions were unsubstantiated . Here we have used updated techniques to rigorously test the proposed correlation between telomere length and VSG switching . Our data demonstrate , for the first time , that antigenic switching in T . brucei increases in a TERT−/− mutant when the telomere of the active BES is short ( in direct contrast to TERT−/− strains with long a telomere whose switching is like WT ) ( FIG . 1 ) . Furthermore , we show that the increase in switching can be accounted for by a significant increase in GC ( in comparison with other measured switching mechanisms ) , which is likely due to an increase in DNA breaks in the region upstream of the VSG in the active/short-telomere BES ( evidenced by an increased loss of VSG pseudogene [Ψ] ) ( FIG . 4 ) . There are two equally valid mechanistic interpretations of these data . The first is that telomere shortening per se leads to an increase in the frequency of DNA breakage in locations that surround the VSG , thus precipitating a switching event , as hypothesized originally by Cross and colleagues [24] . Alternatively , the phenotype we observe might be due to a telomere-capping defect , which would be the result of the combined effect of shortened telomeres in the context of a telomerase-null mutant . Indeed , such a capping defect has been shown to increase gene conversion in the budding yeast Kluyveromyces lactis [38] . To distinguish between these two mechanistic possibilities , we have made concerted efforts to complement telomerase expression in TERT−/− short-telomere T . brucei so that their switching frequency could be compared to un-complemented TERT−/− short-telomere strains . However , we have been unable to set up a strictly regulated inducible system to reconstitute TERT expression while retaining a short telomere at the active expression site ( even at exceedingly low expression levels ) . This is due to the inherent leakiness of inducible systems available for T . brucei , together with the impressive efficiency of catalytically active telomerase to rapidly elongate a short telomere at the active expression site [30] , [31] . Also , the impact of telomerase mutations , if any , on t-loop formation and telomere capping in T . brucei is not known at this time . Thus , at the moment , we cannot distinguish between the possibility that the phenotype we observe in the TERT−/− short-telomere strains is due to a direct effect of telomere shortening , as originally proposed , or a capping defect ( combined with telomere shortening ) due to lack of telomerase function , along the lines of K . lactis [38] . Nevertheless , the general appropriation of conserved mechanisms of chromosome end protection ( breakage , lengthening and capping ) for the purposes of increased antigenic variation is an intriguing possibility as an aspect of antigenic variation in T . brucei . T . brucei strains that have been recently isolated from nature are distinct from their laboratory propagated cousins in that they switch more frequently ( approximately 10−2–10−3 ) [27] , have shorter telomeres [30] , and preferentially switch by GC [39] . Thus , it would appear that by producing a T . brucei strain with an artificially shortened telomere we have , at least partially , recapitulated these characteristics of a natural isolate . The 10- to 100-fold increase OSF measured in the TERT−/− short-telomere strain does not cover the approximately 100- to 10 , 000-fold difference in the rate of switching published between recent isolates ( approximately 10−2–10−3 ) and laboratory-adapted strains in vitro ( approximately 10−5–10−6 ) [28] , [29] . Comparative analysis of switch rate approximations has been a central challenge in the field due to significant variability among the methodologies used and the very small sets of VSG switchers ( often less than 10 ) from which these rate approximations were often derived [27] , [28] , [29] . The OSF value is simply a measurement of the proportion of cells in the population that are no longer coated in the parental VSG , which does not account for the phenotypic diversity of VSGs in the population ( which was included in previous approximate rate derivations ) ( considered in FIG . 2 ) or the genotypic diversity ( for which no method has ever accounted ) ( considered in FIG . 3 ) . The findings presented here suggest that the shorter telomeres of natural isolates could contribute to their comparatively high level of VSG switching . The question that persists is , how naturally occurring populations keep their telomeres short ? In addition , T . brucei telomeres grow by 6–8 bp/PD during growth under laboratory conditions ( a process that has only been observed in this organism ) [25] , [30] . What is distinct between the natural T . brucei lifecycle and their laboratory propagation that could account for the observed differences in telomere lengths ? Although telomerase regulation in T . brucei is not understood , in other organisms telomerase is activity is regulated by the cell cycle such that telomeres only lengthen in the transition from S phase to G2 [40] , [41] , [42] . If this were also the case for trypanosomes , the organisms increased in vivo growth rate ( which is more than 2 times higher than in vitro ) could result in a reduced duration of telomerase activity and thus progressively shorter telomeres . An alternative possibility is that a regulated component of telomere structure or stability , or possibly telomerase itself , is affected when trypanosomes are not permitted to undergo their natural life cycle , which includes passage through the tsetse . In support of this hypothesis , laboratory adapted T . brucei can recover its in vivo switch rate following passage through the tsetse [29] . Although little is known about the regulation of telomerase and other telomere-associated proteins in trypanosomes , perhaps this is a missing connection between the switching behavior of natural isolates and extensively adapted T . brucei strains .
Cell lines were generated from Lister427 bloodstream-form trypanosomes derived from the ‘single marker’ line [43] , with a hygromycin resistance marker at the BES1 promoter [22] ( “wild-type” [WT] ) , or homozygous telomerase ( Gene ID: 3664223 & protein accession: XP_829083 ) deletion mutant with blasticidin resistance marker at the BES1 promoter ( TERT−/− ) [31] , [32] . TERT−/− short and long active-site telomere BES clones were isolated from single cell cultures of TERT−/− and telomere lengths were determined by Southern Blot analysis ( below ) [31] . Trypanosomes were cultured in vitro in HMI-9 medium at 37°C [44] . DNA restriction fragments were separated by standard agarose gel electrophoresis ( 1–15 kb ) , Field Inversion Gel Electrophoresis ( FIGE ) [45] ( 1–25 kb , BIORAD “Program 1” ) , Rotating Agarose Gel Electrophoresis [46] ( “Classical Program” for separation of megabase , intermediate , and minichromosomes ) , using published methods . Southern blots were produced using established methods of capillary blotting by neutral transfer ( GE Scientific ) . DNA probes were made by PCR amplification using previously published primer sequences [4] , [22] , 32P radiolabeled using Prime-It II Random Labeling Kit ( Stratagene ) , and purified over G-50 microcolumns ( GE Healthcare ) . Blots were probe-hybridized , washed , and visualized by phosphorimaging as described ( GE Healthcare ) . Strains expressing VSG427-2 were single cell cloned by limiting dilution and expanded to a total of ∼5×107 cells prior to MACS isolation and flow cytometry-based quantification of VSG switching frequency as previously described [22] . The VSG switching frequency , here termed the “Observed Switching Frequency ( OSF ) ” , is a direct measure of the proportion of living ( measured by propidium iodide staining ) trypanosomes in a population that no longer express VSG427-2 on their surface , compared to the total input population , both of which are normalized to a control sample ( CountBright Beads purchased from and used according to Invitrogen instructions ) . Following MACS depletion of VSG472-2 expressing cells , half of the effluent was removed prior to OSF determination by flow cytomertry and grown in 50 mL HMI-9 with pen-strep until confluent . RNA was extracted , cDNA synthesized , VSG RT-PCR amplified and subcloned ( pGEM-T Easy , Promega ) using established methods ( tryps . rockefeller . edu/trypsru2_protocols_index . html , “VSG cloning for mRNA” ) , prior to sequencing and expressed VSG determination by NCBI BLAST . Following MACS depletion of VSG427-2 expressing cells ( from single cell cultures grown for the specific purpose of VSG427-3 switcher isolation and not OSF determination ) , the effluent was labeled with anti-224 antibody and bound to a second MACS column . Following standard MACS binding and wash steps the bound trypanosomes ( anticipated VSG427-3+ switchers ) were plunged from the column , cloned and screened for VSG427-3 expression by flow cytometry . VSG427-3 expressing clones were grown for genomic DNA isolation [46] , which was analyzed by PCR using BES1 VSG pseudogene primers ( pseudo5-F: 5′- GCGCCGAATTTAATGCAATATGCAACG & pseudo5-R: 5′- GCAGGCCGTCTTTTGAGTTGTAGTAAG ) & BES1 ESAG1 primers ( ESAG1 Fw1: 5′-GAGCAAACTGATAGGTTGGAAAAG & ESAG1 Rv1 5′-GCACTGGCGGCCACTCCATTGTC ) and HindIII FIGE Southern blot analysis using a BES7 specific probe ( FIG . 4A – red bar ) produced by PCR using unique primers ( BES7 UR1 Fw: 5′-GCAACTAACTACTGTAATTCCC & BES7 UR Rv: 5′-GCTACTAATGTGTTTCAATATGCG ) . Following expansion of VSG427-2 ( 221 ) expressing WT and TERT−/− short-telomere cells from single cells to ∼5×107 total cells and MACS depletion of VSG472-2 expressing cells as described above [22] , the effluent was split into two samples . Half of the resulting cells were used to determine the OSF and the other half was cloned by limiting dilution in 96-well tissue-culture plates . Clone growth was observed for two weeks following initial plating . At 7 , 9 , and 11 ( WT only ) days after plating , confluent cultures were split into new 96-well plates and allowed to become confluent . Final identification of VSG427-2− switchers was performed by high-throughput-screening ( HTS ) flow cytometry using anti-VSG221 antibody on an LSRII with HTS adaptor for 96 well plate analysis ( BD Biosciences ) . VSG427-2 negative clones were then analyzed for BES1 promoter activity by antibiotic selection of the BES1 promoter proximal marker ( WT: hygromycin & TERT−/−: blasticidin ) and the Southern dot blots for the presence of VSG427-2 ( 221 ) , the promoter proximal antibiotic resistance gene and BES1 VSG pseudogene in the genome ( using radiolabeled probes to VSG221 , hygromycin , blasticidin , pseudogene ) . Southern dot blot was adapted for trypanosomes from manufacturers protocols ( GE Healthcare ) by adding ∼1×106 cells/well to the membrane , cell lysis with denaturation buffer , neutralization , and fixation with UV prior to radiolabeled probe hybridization , washing , and visualization as described above for Southern blot analysis . Primers for PCR production of Southern dot blot probes: 221 Prb Fw ( 5′- GTAACAGCTACTGCAACAGCGAGC ) & 221 Prb Rv ( 5′- GCTTCTTCAACAAGCTTGGTAACG ) , HYGRO Prb Fw ( 5′-GCTCTCGATGAGCTGATGCTTTGG ) & HYGRO Prb Rv ( 5′-GATAGAGTTGGTCAAGACCAATGC ) , BES1 PSEUDO Fw ( 5′- CATTAAATTCAAGCGTCTAGACCGCAGC ) & BES1 PSUEDO Rv ( 5′- GCGCGTTGTTCCGTATCTGCTGAGC ) If ΣFShort≈ΣFWT+ΔFGC , then GC accounts for the increase in short-telomere OSF . Where FMech = Average OSF×% Mechanism and ΔFGC = FGC , Short−FGC , WT . First the F for each mechanism for both WT and Short were determined , the sum of which is equal to the average OSF for each strain ( ΣFWT = 1 . 38 & ΣFShort = 8 . 84 ) . Then the ΔFGC was determined and used to show that ΣFWT+ΔFGC = 8 . 86 which is fundamentally identically to the ΣFShort ( 8 . 84 ) , thus making ΣFShort≈ΣFWT+ΔFGC a valid statement ( TABLE S3 ) . | A broad array of human pathogens ( including bacteria , fungi and parasites ) vary the proteins on their cell surface to escape the immune response of their hosts . This process , called antigenic variation , relies on a repertoire of variant protein encoding genes in the genome and the organism's ability to accurately switch from the expression of one variant gene to another . A common theme in both the diversification of these variant genes and the mechanisms required for their expression is that they are often located near the ends of chromosomes . The ends of chromosomes are protected by structures called telomeres . Regions near the telomere are referred to as subtelomeric and are commonly thought to be comparatively unstable DNA sites . It is therefore intriguing that organisms that rely on antigenic variation for survival would organize their critical survival genes in these sites . Trypanosoma brucei is a model organism for the study of antigenic variation . The causative agent of African sleeping sickness , this unicellular parasite possesses an antigenic repertoire of unparalleled diversity , which can only be expressed from specific subtelomeric sites . Here we use the power of the T . brucei model to investigate the effect of telomere length on antigenic variation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genetics",
"biology",
"microbiology",
"genetics",
"and",
"genomics"
] | 2012 | Telomere Length Affects the Frequency and Mechanism of Antigenic Variation in Trypanosoma brucei |
The increasing prevalence of multi-drug–resistant ( MDR ) strains of Pseudomonas aeruginosa among critically ill humans is of significant concern . In the current study , we show that MDR clinical isolates of P . aeruginosa representing three distinct genotypes that display high virulence against intestinal epithelial cells , form novel appendage-like structures on their cell surfaces . These appendages contain PstS , an extracellular phosphate binding protein . Using anti-PstS antibodies , we determined that the PstS-rich appendages in MDR strains are involved in adherence to and disruption of the integrity of cultured intestinal epithelial cell monolayers . The outer surface–expressed PstS protein was also identified to be present in P . aeruginosa MPAO1 , although to a lesser degree , and its role in conferring an adhesive and barrier disruptive phenotype against intestinal epithelial cells was confirmed using an isogenic ΔPstS mutant . Formation of the PstS rich appendages was induced during phosphate limitation and completely suppressed in phosphate-rich media . Injection of MDR strains directly into the intestinal tract of surgically injured mice , a known model of phosphate limitation , caused high mortality rates ( 60%–100% ) . Repletion of intestinal phosphate in this model completely prevented mortality . Finally , significantly less outer surface PstS was observed in the MPAO1 mutant ΔHxcR thus establishing a role for the alternative type II secretion system Hxc in outer surface PstS expression . Gene expression analysis performed by RT-PCR confirmed this finding and further demonstrated abundant expression of pstS analogous to pa5369 , pstS analogous to pa0688/pa14–55410 , and hxcX in MDR strains . Taken together , these studies provide evidence that outer surface PstS expression confers a highly virulent phenotype of MDR isolates against the intestinal epithelium that alters their adhesive and barrier disrupting properties against the intestinal epithelium .
Infection due to P . aeruginosa continues to be a major cause of mortality among critically ill and immuno-compromised patients despite the development of newer and more powerful antibiotics . Both the immunoevasive nature of P . aeruginosa as well as its acquisition of multi-drug resistance makes elimination of this organism a particular challenge . Multi-drug–resistant ( MDR ) strains of P . aeruginosa , defined as resistant to at least three of the following antibiotics: ceftazidime , imipenem , gentamicin or ciprofloxacin , are often isolated from patients exposed to prolonged intensive care-type therapies [1] . Yet antibiotic resistance itself does not confer enhanced virulence [2] , and therefore the ability to discriminate between virulent versus non-virulent phenotypes among colonizing multi-drug resistant isolates would be a major step in predicting the particular threat of a colonizing strain of P . aeruginosa . The primary site of colonization and a frequent source of subsequent infection of P . aeruginosa is the gastrointestinal tract reservoir , where as many as 50% of critically ill patients are colonized within 3 days of admission with as many as 30% of strains displaying antibiotic resistance [3] . Yet little is known about the behavior of these pathogens in this site , especially those that are multi-drug resistant . We recently screened several strains of MDR isolates from hospitalized patients and characterized their virulence against the intestinal epithelium using an in vitro model of cultured intestinal epithelial monolayers [2] . The majority of strains ( 60% ) were found to be either attenuated or have no effect in their ability to adhere to or disrupt the integrity of the intestinal epithelium . However several strains representing three distinct genotypes , showed extremely high adherence capacity and a profound ability to disrupt the barrier function of cultured intestinal epithelial cells . These strains harbored the exoU gene , known to encode the most toxic effector protein of the type III secretion apparatus , thus possibly explaining their extreme toxicity against cultured intestinal epithelial cells . However , exoU expression is dependent on contact to host epithelial cells , and as recently shown with the exoU positive strain P . aeruginosa PA103 [4] , lack of adherence leads to a loss of cytotoxicity against cultured epithelial cells [5] despite an intact exoU gene . Therefore we studied selected exoU positive MDR strains of P . aeruginosa displaying unusually high adherence and disrupting properties against the intestinal epithelium and determined whether surface structures might exist to explain their enhanced adhesiveness to cultured intestinal epithelial cells . In this report we show that these strains express previously un-described appendages that contain significant quantities of PstS , a high affinity phosphate binding protein . We characterized the structural and functional aspects of these PstS rich appendages and determined that they play a significant role in the adherence to and disruption of intestinal epithelial cells . Outer surface expression of PstS rich appendages was induced under low phosphate conditions and suppressed in high phosphate media . Lethality assays in a mouse model of gut-derived sepsis in which low phosphate conditions are known to exist , demonstrated high lethality rates that were completely abrogated when mice were supplemented with intestinal phosphate . Taken together , these data provide evidence that low phosphate conditions increase the presence of PstS rich appendages on MDR P . aeruginosa whose presence facilitates binding to the intestinal epithelium and whose expression in vivo may play a significant role in the development of gut-derived sepsis in critically ill patients .
In our previous work we screened consecutive MDR P . aeruginosa clinical isolates and identified a subset of strains that displayed a highly destructive phenotype against cultured intestinal epithelial cells ( Caco-2 bbe ) [2] . Among this subset of isolates , high swimming motility , increased adhesiveness to Caco-2 monolayers , and the presence of the exoU gene predicted a cytotoxic phenotype against the intestinal epithelium [2] . In the present study , we screened these highly adhesive MDR clinical isolates by their cell surface morphology using electron microscopy , and identified appendage-like structures on the surfaces of the most cytotoxic isolates ( 1 , 13 , and those of genotype 20 ) ( Figure 1A–1D and Table 1 ) . The identified appendages were 20 nm in diameter , up to 500 nm in length , and were visually distinct from flagella ( Figure 1A and 1D ) . The identified appendages were not detected on any of the remaining clinical isolates of the previously reported series of strains ( see Figures S1 and S2 ) . To identify proteins involved in the formation of the visualized appendages , surface-associated proteins were obtained by extensive vortexing of bacterial cells grown on Pseudomonas isolation agar ( PIA ) , denatured by boiling with sample buffer and then separated by 10% Tris-glycine SDS-PAGE . Figures 2A and 2B show the presence of abundant protein bands at an approximate MW of 32 kDa from surface sheared proteins in strain MDR25 ( Figure 2A , lane 2 ) ; 32 and 40 kDa bands from strain MDR1 ( Figure 2A , lane 3 ) ; and a 40 kDa band from strain MDR13 ( Figure 2B , lane 2 ) . Proteins were transferred to a PVDF membrane and N-terminal sequencing of the 32 and 40 kDa proteins in strain MDR1 and the 40 kDa protein in strain MDR13 were performed with ABI-Procise cLC Protein Sequencer ( Mayo Proteomics Research Center ) . The N-terminal peptide sequence of the 32 and 40 kDa proteins in strain MDR1 were found to be AIDPALPEYQK and EINGGGATLPQQLXQEPGV , respectively . The N-terminal peptide sequence of 40 kDa protein in strain MDR13 was identified as DINGGGATLPQQLYQ . The peptide sequences were searched with BLAST , and the best hit of the 32 kDa band sequence was found to be the PstS protein PA5369 in P . aeruginosa PAO1 . The sequence AIDPALPEYQK was located to aa 25–35 on ORF PA5369 . PA5369 was demonstrated to contain a cleavable type I signal peptide of 24 aa [6] , therefore the 32 kDa protein in strain MDR1 might correspond to periplasmic orthologous protein PA5369 in PAO1 . The best hit of the 40 kDa band sequence in both strains 1 and 13 was found to be PA55410 from P . aeruginosa PA14 ( http://v2 . pseudomonas . com/getAnnotation . do ? locusID=PA14_55410 ) . The sequence DINGGGATLPQQLYQ was located to aa 24–38 on ORF PA55410 . Orthologous to PA55410 , PA0688 protein in PAO1 was also demonstrated to contain a cleavable type I signal peptide of 23 aa , MFKRSLIAASLSVAALVSAQAMA [6] , which was 100% identical to N-terminus of PA55410 . Therefore the 40 kDa proteins in strains 1 and 13 might be orthologous to PA55410 in P . aeruginosa PA14 [7] , the strain known to be highly virulent , and PA0688 from P . aeruginosa PAO1 . We next amplified and sequenced genes analogous to pa5369 in MDR strains 1 , 13 , and 25 ( GenBank Accession numbers EF601157 , EF601158 , and EF601159 ) . We determined them to be very conserved with few differences in nucleotide sequences that did not affect amino acid sequences which were 100% identical to PA5369 in P . aeruginosa PAO1 ( http://www . pseudomonas . com/ ) . We therefore created anti-PA5369 antibodies against the specific peptide 192–212 KEEALCKGDFRPNVNEQPGS that was chosen based on hydrophobicity , surface probability , flexibility , and antigenic index , as well as the Advanced BLAST Search for the absence of significant homology to other P . aeruginosa proteins . Antibodies were subjected to affinitive purification using the native peptide 192–212 , and then used to detect appendage-like structures in the clinical isolates . We first performed immunobloting of cell surface structures using anti-192–212 peptide antibodies , now referred to as anti-PA5369 antibodies , and found high antibody affinity to cell surface proteins of strain MDR25 , moderate affinity to strains MDR1 and MPAO1 , and minimal affinity to strain MDR13 ( Figure 2C ) . These data corresponded to the results of the SDS-PAGE demonstrating an abundance of the 32 kDa protein band in strain MDR25 but not in strain MDR13 . The specificity of the anti-PA5369 antibody was confirmed by examining both sheared appendages and bacterial pellets in wild-type MPAO1 , PA5369 mutant , and clinical strain MDR25 ( Figure 2D ) . Results demonstrated that anti-PA5369 recognized abundant amounts of protein in both the cell pellet and sheared appendages in strain 25 . In strain MPAO1 , antibodies recognized an abundant amount of protein in the cell pellet but a low amount in sheared appendages . No recognizable protein in either the cell pellet or sheared surface fractions in the MPAO1 mutant ΔPA5369 was observed . ELISA assays performed with sheared proteins from different clinical isolates ( Figure 2E ) demonstrated the presence of highly abundant PA5369-like protein in MDR clinical strains 25 , 27 , and 28 , all of which share the same genotype 20 . Interestingly , anti-PA5369 antibodies recognized significantly lower amount of proteins in sheared fractions isolated from other clinical isolates previously shown to be unable to alter the epithelial resistance of Caco-2 monolayers ( see Table 1 ) [2] . Another interesting finding was the presence of PA5369 in sheared fractions of MPAO1 by both ELISA ( Figure 2E ) and immunobloting ( Figure 2D ) , although much lower in abundance compared to the highly adherent strains MDR25 and MDR1 . PA5369 has been predicted by COG ( Clusters of Orthologous Groups , http://www . pseudomonas . com/ ) to be PstS , a phosphate transport system substrate-binding protein whose expression in P . aeruginosa is induced at phosphate concentrations < 1 mM [8–12] . In order to determine if the formation of appendages in clinical isolate 25 was phosphate dependent , we suspended a single colony in 10% glycerol , and plated equal amounts on either PIA that we measured to contain 300 μM of phosphate or PIA supplemented with 1 mM K-phosphate buffer , pH 7 . 0 . Cells grown on these plates were analyzed for the presence of appendages by transmission electron microscopy ( TEM ) and immunobloting . TEM images clearly demonstrated the absence of appendages in strain MDR25 grown on phosphate-enriched PIA ( Figure 2F ) and an abundance of appendages in the same strain grown on PIA only ( Figure 2G ) . We also noted the differences in colony phenotype when smooth surface colonies appeared on high Pi media ( Figure 2Fa ) , and rough surface colonies appeared on low Pi media ( Figure 2Ga ) . Immunoblot analysis ( Figure 2H and 2I ) demonstrated the absence of proteins recognized by anti-PA5369 antibody on phosphate rich media versus their abundance on phosphate poor media in both sheared cell surface fractions ( Figure 2H ) and cell pellets ( Figure 2I ) . Finally , we performed immuno-gold electron microscopy of strain MDR25 to confirm the presence of PstS protein on the appendages . Whole cells of strain MDR25 were directly harvested from PIA plates and incubated with anti-PA5369 antibody followed by incubation with gold-labeled goat anti-rabbit antibody . Figure 2J demonstrates gold spots localized on the cell surface structures in strain MDR25 . Gold spot localization was not observed in negative controls performed in the absence of primary anti-PA5369 antibodies ( data not shown ) . We noted the fragility of these appendages on EM as detached and fragmented appendages were observed ( see Figure S3 ) . In order to determine the contributory role of the PA5369-like protein on the ability of MDR P . aeruginosa to adhere to and disrupt barrier function of cultured intestinal epithelial cells , we examined the effect of anti-PA5369 antibodies on the adhesiveness and barrier disrupting capability . In order to avoid the non-specific interference of whole antibodies , we purified Fab fragments of anti-PA5369 antibodies to use in these experiments . Using stain MDR25 , we performed adhesion assays to Caco-2 monolayers and determined the transepithelial resistance ( TER ) of Caco-2 cells , a measure of barrier function , in the presence or absence of Fab fragments of anti-PA5369 antibodies . Both the adhesiveness of strain MDR25 to Caco-2 monolayers ( Figure 3A ) and the ability of strain MDR25 to disrupt the TER of Caco-2 monolayers ( Figure 3B ) were significantly attenuated when pre-incubated with the Fab fragments of anti-PA5369 antibodies . In order to determine if PstS could influence the ability of non-multi-drug resistant strains to alter intestinal barrier function , we performed complementary experiments using P . aeruginosa MPAO1 and its derivative mutant ΔPA5369 [13] . The mutant ΔPA5369 was complemented with the pa5369 gene on a multi-copy plasmid pUCP24 ( Δ5369/pa5369 ) . First strains were verified for the presence of surface-associated PstS by ELISA using specific anti-PA5369 antibodies ( Figure 3C ) . In order to determine if PstS contributed to the adhesiveness of MPAO1 , we apically inoculated Caco-2 monolayers with P . aeruginosa strains and assessed the degree of adhesiveness after one hour of co-incubation . We observed the adhesiveness of MPAO1 to cultured intestinal epithelial cells to be as low as 1% of the initial inoculum; an effect that was further decreased with the mutant strain Δ5369 ( Figure 3D ) . Strain Δ5369/pa5369 demonstrated increased adhesiveness to Caco-2 cells compared to both the wild type and 5369 mutant ( Figure 3D ) . Reiterative experiments were then performed to assess the ability of the strains to alter epithelial barrier function , as measured by TER of Caco-2 cells . We have previously reported strain MPAO1 to display low virulence against Caco-2 monolayers ( ∼5% decrease in TER at 3 hours ) compared to clinical strain MDR25 ( ∼70% decrease in TER at 3 hours ) ( see Table 1 ) . However at later time points ( 7 hours ) strain MPAO1 decreased resistance of Caco-2 monolayers by 60%–70% . Therefore , we measured TER after 7 hours of co-incubation of Caco-2 cells using MPAO1 and its derivatives and found that strain Δ5369 was significantly attenuated in its ability to decrease the TER of Caco-2 monolayers ( Figure 3E ) . Complementation of Δ5369 with pa5369 gene restored its effect to decrease the TER of Caco-2 cells similar to the wild type PAO1 ( Figure 3E ) . We incidentally noticed the spontaneous appearance of smooth colonies among rough-edged colonies of MDR25 when grown on PIA where the phosphate level ( Pi ) was determined to be ∼300 μM ( Figure 4A , black arrows ) . When smooth colonies were isolated and re-plated on PIA , the rough-edged colonies re-appeared interspersed among smooth colonies ( Figure 4B , shown by white arrow ) suggesting the possibility of colony phase variation . We also noted that rough ( MDR25R ) and smooth ( MDR25S ) colonies were distinct in their production of biofilm , whereby MDR25S produced significantly greater amounts of biofilm compared to MDR25R ( Figure 4C ) . Growth curves for MDR25S and MDR25R were similar in liquid Pseudomonas broth ( see Figure S4 ) . We next examined smooth and rough colonies for their PstS content on surface sheared fractions by ELISA using anti-PA5369 antibodies and determined that smooth edge colonies produced significantly less PstS compared to rough colonies ( Figure 4D ) . Finally , we determined if MDR25R and MDR25S differentially induced mortality in mice using an established model of lethal gut-derived sepsis [14 , 15] . This model involves creating a surgical stress with a 30% hepatectomy and simultaneous intestinal exposure to P . aeruginosa via direct injection into the cecum [15] . This model is of particular clinical relevance as it is well established that surgical hepatectomy results in severe hypophosphatemia [16] . Rough and smooth colonies were suspended in 10% glycerol at OD 0 . 25 ( 600 nm ) and injected into the cecum at the time of hepatectomy . Mice were followed for 48 hours for mortality . Results demonstrated that mice injected with the smooth edged , PstS poor strain MDR25S displayed 10% mortality at 48 hours whereas mice injected with the rough edged PstS rich strain MDR25R , displayed 60% mortality at 48 hours ( Figure 4E ) . Data were analyzed using Kaplan-Meier survival curves with SPSS software , n = 10/group , p = 0 . 021 . We determined if phosphate supplementation in mice subjected to a 30% hepatectomy , could prevent lethality due to MDR25R or MDR1 by performing reiterative experiments in which mice were fed varying concentrations of phosphate ( [Pi] ) . Group 1 ( n = 8 ) were fed water only , Group 2 ( n = 8 ) were fed 0 . 2x PBS ( phosphate buffered saline , pH 7 . 4 , [Pi] = 2 mM ) as their drinking water , and Group 3 ( n = 5 ) were fed 1x PBS ( [Pi] =10 mM ) as their drinking water for 36 hours before surgical hepatectomy and injection of bacteria . In addition , prior to injection of MDR25R into the cecum , bacteria were suspended in either water containing 10% glycerol ( group 1 ) or 0 . 2x PBS ( group 2 ) or 1x PBS ( group 3 ) . Results shown in Figure 5A demonstrate that MDR25R caused 100% mortality within 48 hours when mice drank water only ( Group1 ) whereas mice drinking a water solution containing 2 mM phosphate ( Group 2 ) had significantly decreased mortality ( 50% ) while mice drinking a water solution containing 10 mM phosphate ( Group 3 ) had no mortality ( 100% survival ) . Data were analyzed using Kaplan-Meier surviving curves in SPSS software , p = 0 . 004 . Similar results were found in reiterative experiments with the clinical isolate MDR1 ( Figure 5B ) ( n = 10 , p = 0 . 001 ) . The outer surface expression of PstS PA5369 observed in the current study is at variance with its previously reported characterization as a periplasmic protein . In order to clarify this we hypothesized that knockout of adjacent low phosphate responsive elements might impair the surface expression of PstS . PstS PA5369 is clustered to the phosphate ABC transporter locus ( Figure 6A ) . Based on the KEGG SSDB ( Kyoto Encyclopedia of Genes and Genomes Sequence Similarity DataBase http://www . genome . jp/kegg/ssdb/ ) search , PstS PA5369 can be considered as a paralogous protein of PstS PA0688 in P . aeruginosa PAO1 that is clustered to the alternative type II secretion locus ( Figure 6B ) . According to recent data [17] , PA0688 is characterized as alkaline phosphatase that is secreted by the alternative type II secretion system Hxc induced under low phosphate conditions . We therefore hypothesized that hxc might also play a role in the outer surface expression of PstS PA5369 and performed immunoblot analysis of sheared appendages in strains P . aeruginosa PAO1 and its derivative mutant ΔHxcR ( PA0686 ) . As shown in Figure 6D , the ΔHxcR mutant was attenuated in the production of the outer surface but not intracellular PA5369 suggesting at least partial involvement of hxc system to present PstS on outer surface appendages . Moreover , complementation of the mutant with hxcR restored its ability to express outer surface PstS ( Figure 6D ) , confirming involvement of the Hxc system . N-terminal sequence of proteins from sheared fractions of the MDR isolates 1 and 13 by Blast Search correspond to PA14–55410 which is orthologous to PA0688 and clusters to the hxc system ( Figure 6C ) . We amplified and sequenced the corresponding gene in strain MDR1 , and found that the protein encoded by this gene had 90% identity to PA14 55410 from P . aeruginosa PA14 , 45 . 3% identity to PA0688 from P . aeruginosa PAO1 , and 64 . 3% identity to the human plasma phosphate-binding protein HPBP which is classified as a DING protein [18–20] . We therefore named the protein from MDR1 as DING and its respective gene dinG ( GenBank Accession number EF616488 ) . We next performed experiments to determine the expression level of pstS and related genes from the phosphate ABC transporter system , pstS pa5369 , phoB and phoU , as well as pstS and related genes from hxc system , pa0688 , dinG , and hxcX , a gene of hxc operon ( Figure 6A–6C ) . We examined four strains: MPAO1 , MDR1 , MDR25R , and MDR25S . Strains were grown overnight on PIA and PIA complemented with 10 mM K-Ph buffer , pH 7 . 0 , and RNA was directly isolated from cell colonies . Results are presented in Figure 6E . While the expression of the housekeeping enzyme citrate synthase demonstrated no significant change in response to phosphate limitation , the expression of all genes tested was increased in response to low phosphate media with each expressing a distinct pattern . While a similar increase in phoU expression was observed between all strains , only a modest increase in phoB expression was detected in MDR25R compared to other strains . Similarly only a modest increase in pstS pa5369 expression was observed in the MDR isolates compared to MPAO1 . The most intriguing finding however was that , although pstS pa5369 expression was similar between the rough ( high outer surface PstS ) and smooth ( low outer surface PstS ) colony variants , a dramatic difference in expression was observed in the hxc operon . In fact , hxcX expression was ten times higher in MDR25R compared to MDR25S . The low response of hxcX to phosphate limitation was also observed in MPAO1 strain compared to MDR25R and MDR1 . Although the expression of pa0688 , the orthologous of DING protein in MPAO1 , was 20-fold higher at low phosphate concentrations , this effect was small compared to that observed for MDR strains where a 150-fold increase was observed with MDR25S , a 1 , 400-fold in MDR25R , and 5 , 000-fold increase in expression in MDR1 .
Numerous reports have documented that the rise in MDR nosocomial pathogens continues to threaten hospitalized patients despite the implementation of various countermeasures including isolation techniques and antibiotic de-escalation measures [21] . While the mere culture of a MDR resistant pathogen such as P . aeruginosa is perceived to be a real and present danger to patients primarily because it cannot be readily eliminated by antibiotics , the evidence that antibiotic resistance itself confers a more virulent phenotype is highly variable . Our previous work on screening consecutively isolated MDR strains of P . aeruginosa from critically ill hospitalized patients demonstrated that strains express extremely polar virulence phenotypes against the intestinal epithelium from those that are essentially inert , to those that are highly motile , adhesive , and destructive [2] . In fact among the consecutively collected strains in this series , only a minority of strains displayed a virulent phenotype against the intestinal epithelium ( ∼15% ) . A better understanding of the virulence determinants of MDR P . aeruginosa and their mechanism of action against the intestinal epithelium is important given the high prevalence of colonization of this organism in the intestinal tract of critically ill and immuno-compromised patients [22–24] . Human critical illness represents a unique ecological niche for P . aeruginosa because of its prolonged exposure to antibiotics and physiologic disturbances that have no historical precedent in terms of host survival . Extensive life sustaining measures employed during the care of the critically ill such as the use of gastric acid suppression therapy , vasoactive agents that result in profound luminal hypoxia , continued use of opioids that impair the ability of the intestinal tract to excrete non-commensal pathogens , and the delivery of highly processed artificial nutrition , all favor the exposure of pathogens like P . aeruginosa to a composite of environmental cues that can directly activate its virulence circuitry [25–29] . In this regard a major environmental cue within the intestinal tract that could shift the virulence of P . aeruginosa to that of a more virulent phenotype against the epithelium may be low extracellular phosphate which is often present during severe critical illness [16 , 30–32] . Hypophosphatemia is reported to be present following a variety of physiologic stress states such as myocardial infarction [33] , ischemia-reperfusion injury [33] , major liver resection [16] , use of insulin to control hyperglycemia [34 , 35] , use of intravenous nutrition [36 , 37] , and during sepsis [38] . In such circumstances , phosphate depletion appears to be severe and an independent predictor of mortality due to infection and sepsis [38] . While the mechanisms for this observation are unknown , it is possible that colonizing strains of P . aeruginosa present in the intestinal lumen of critically ill patients become activated to express a more virulent phenotype against the intestinal epithelium in response to low phosphate concentrations resulting from surgical injury and catabolic stress . In the present study , we determined that MDR P . aeruginosa clinical strains displaying a high degree of virulence against cultured intestinal epithelial cells express an extraordinary amount of surface-associated PstS proteins orthologous to PA5369 from P . aeruginosa PAO1 and PA14 55410 protein from P . aeruginosa PA14 . The observation that PstS on appendages contributes to intestinal epithelial adherence coupled with its known role as a phosphate binding protein , raises the possibility that the PstS present on appendages might facilitate the ability of MDR P . aeruginosa to acquire phosphate from intracellular stores within the host . This latter effect appeared to be dependent on the presence and expression of the alternative type II secretion system Hxc , which itself is activated in the presence of low phosphate [17] . This finding establishes a link between the phosphate binding ABC transporter and the Hxc system in P . aeruginosa . Simultaneous expression of both systems is necessary for the formation of outer surface PstS-rich appendages as neither ΔPstS nor ΔHxcR mutants produce them in P . aeruginosa MPAO1 . Both the PstS and the Hxc systems are highly inducible in MDR clinical isolates that express a particularly adhesive and barrier disrupting phenotype against intestinal epithelial cells . In this regard , certain MDR P . aeruginosa strains may have adapted unique genetic changes in response to unusually harsh selective pressures that typify critically ill humans including multiple antibiotic use , severe hypoxia , and the ability to sustain life with prolonged intravenous nutrition . Among such changes may be the ability to acquire phosphate and other nutrients from within host cells given that physiologic stress and tissue injury are know to shift phosphate into the intracellular compartment resulting in severe hypophosphatemia [39] . Under such circumstances outer surface expression of PstS on appendages may confer an evolutionary advantage to P . aeruginosa by expressing phosphate acquiring structures capable of scavenging intracellular phosphate at arm's length from the host immune system . Data from the present study are not the first to demonstrate that PstS is secreted by bacteria during nutrient manipulation of the media . For example , it has been recently reported that Streptomyces lividans secretes PstS into liquid cultures containing high concentrations ( >3 % ) of certain sugars , such as fructose , galactose , and mannose [40] . Another example in which PstS has been shown to be secreted is PA0688 protein in P . aeruginosa PAO1 . Inquiry of the PA0688 protein into the “Clusters of Orthologous Groups ( COG ) Program” predicted it to be PstS . However PA0688 clusters to the hxc loci and has been shown to be secreted by the Hxc system under phosphate depleted conditions and functions as alkaline phosphatase [17] . In the current study , the proteins identified in MDR P . aeruginosa orthologous to PA0688 were expressed several hundred fold higher than in MPAO1 . That sequence analysis revealed this protein in MDR1 to belong to DING proteins is intriguing , given that the origin and function of DING proteins have remained a focus of speculation . DING proteins are characterized by their N-terminal DINGGGATL-sequence and are highly conserved in both animal and plants , although they are more variable as microbial proteins [19 , 41–43] . There are some functional similarities between DING proteins from pro- and eukaryotes including structural homology with phosphate-binding proteins [19 , 43] . It has been hypothesized that pathogenic or symbiotic bacteria might acquire the DING gene via horizontal gene transfer from eukaryotes in order to sense and respond to host signals or to modify intercellular signaling pathways in host cells [42] . Others have suggested that DING proteins do not exist in eukaryotes at all , and that their detection in human tissues has been a result of microbial contamination or infection [41] . Based on the codon usage analysis of DNA , it has been assumed that DING sequences found in eukaryotes are of Pseudomonas origin [41] . Further work is in progress to characterize the role of the DING-protein related appendages found in this series of MDR clinical isolates , the results of which may add to our understanding of the impact of bacterial DING proteins on the modulation of signal transduction in animals . The gene pstS pa5369 is part of the pst operon encoding a specific phosphate transport system that is activated under low phosphate conditions . The high affinity phosphate transport system pst belongs to the Pho regulon that is controlled by the two-component regulatory system PhoB/PhoR , which responds to local phosphate concentration . PhoB/PhoR is highly conserved and widely present in Gram negative and Gram positive microorganisms . In addition , PhoB/PhoR controls the expression of multiple genes [44 , 45] many of which are involved in phosphate uptake and metabolism and various other metabolic pathways such as the de novo biosynthesis of NAD [46] , the initiation of chromosome replication [47] , the acid shock response [48] , the RpoS-mediated stress response [49] , and AMP hydrolysis [44 , 50] . The phosphate regulon might be also involved in the activation of quorum sensing in P . aeruginosa as evidenced by the recent observation that the transcriptional activation of rhlR and production of PQS and pyocyanin develop during phosphate limitation [12] . Furthermore , a link between the expression of the ABC phosphate transporter and penicillin resistance in Streptococcus pneumoniae has been reported thereby proposing a novel role for PstS [51] . These investigators reported that the pstS gene product was overproduced in resistant isolates , the inactivation of which resulted in penicillin sensitivity [51] . Further evidence linking PstS to antibiotic resistance has been demonstrated in fluoroquinolone resistant Mycobacterium smegmatis [52] where amplification of the phosphate specific transporter suggested that the efflux mediated fluoroquinolone resistance might be an intrinsic function of the Pst system [52–54] . Thus is it plausible that the development of multi-drug resistance in P . aeruginosa clinical isolates MDR1 , MDR13 , and MDR25 might be related to the overproduction of PstS proteins . Mouse lethality experiments from the present study strongly suggest a significant role for PstS in the virulence of MDR25R P . aeruginosa virulence in vivo . The importance of PstS in in vivo virulence has been previously addressed in various models including a mouse infection model using Mycobacterium tuberculosis and pstS1 and pstS2 knockout strains [55] , a fish infection model using Edwardsiella tarda , a facultative aerobic enterobacterium that causes hemorrhagic septicemia in fish and gastrointestinal infections in humans [56] , and a chicken infection model using Escherichia coli O78 , an organism associated with extraintestinal infections and septicemia in poultry , livestock , and humans [57] . In summary , we have identified PstS-rich appendage-like structures on the outer surfaces of selected strains of MDR P . aeruginosa that confer a highly adhesive and virulent phenotype against cultured intestinal epithelial cells . Further characterization of these appendages and better understanding of their molecular regulation are needed to fully define their role in the virulence of multi-drug resistant P . aeruginosa . The observation that critical virulence factors such as PstS in P . aeruginosa are highly responsive to environmental phosphate , in conjunction with the observation that intestinal phosphate repletion completely prevented mortality in surgically injured mice exposed to MDR strains , underscores the importance of recognizing intestinal phosphate depletion following catabolic stress and a possible strategy of intestinal phosphate loading as a countermeasure against colonizing strains of P . aeruginosa that are resistant to all conventional antibiotics .
The consecutively collected clinical strains of MDR P . aeruginosa used in the present study ( Table 1 ) have been characterized and described previously [2] . P . aeruginosa strains MPAO1 , MPAO1 mutant ΔPA5369 ( PA5369:: ISphoA/hah , ID 29772 , and MPAO1 mutant ΔPA0686 ( PA0686:: ISphoA/hah , ID 957 ) were obtained from the P . aeruginosa mutant library [13] . The mutant ΔPA5369 was complemented with pa5369 gene to create strain ΔPA5369/ pa5369 , and the mutant ΔPA0686 was complemented with DNA comprising pa0686 plus pa0687 genes to create the strain ΔPA0686/ pa0686-pa0687 . The MDR clinical isolates were routinely subcultured from frozen stocks on Pseudomonas isolation agar ( PIA ) containing Gm , 50 μg/ml . Note that strain 25 , herein referred to as MDR25 , and all strains of genotype 20 ( see Table 1 ) did not grow on rich media ( LB- liquid or agarized ) or TSB ( liquid or agarized ) , and did not grow in the specially designed phosphate limited liquid media described by Hancock [58] . We observed strain MDR25 growth was best supported in Pseudomonas broth that we determined to contain 2 mM Pi , and PIA determined to contain ∼0 . 3 mM Pi . Human intestinal epithelial cells Caco-2bbe were grown to confluence in 0 . 3 cm2 transwells ( Costar ) , and their barrier function was assessed by measuring the transepithelial electrical resistance ( TER ) to a fixed current across cells as previously described [2] . All experiments were performed in triplicate . Adhesiveness of P . aeruginosa to Caco-2 bbe cells was determined as previously described [2] . All experiments were performed in triplicate . Biofilm formation was assayed as described with modifications [59] . Briefly , P . aeruginosa MDR clinical strains were grown overnight in 2 ml of PB , Gm 50 μg/ml in 15 ml culture tubes at 37°C , 200 rpm ( C24 Incubator Shaker , New Brunswick Scientific , Edison , NJ ) . The wells were then rinsed thoroughly with water and the attached material was stained with 3 ml of 0 . 1% crystal violet , washed with water , and solubilized in 3 ml of ethanol . Solubilized fractions were collected and absorbance measured at 550 nm . All experiments were performed in triplicate . P . aeruginosa strains were grown on PIA plate , than bacteria were harvested , suspended in PBS , and surface-associated structures were sheared by vigorous vortexing for 2 min . After centrifuging at 5 , 000g , for 5 min , proteins in the supernatant were separated using 10% Tris-glycine SDS-polyacrylamide gel and detected by Coomassie brilliant blue staining . For amino-terminal peptide sequence analysis , the proteins were electroblotted onto polyvinylidene fluoride ( PVDF ) membranes , and sequenced by Edman degradation chemistry using an Applied Biosystems Procise 492 HT Protein Sequencer ( Applied Biosystems , Foster City , CA ) at the Mayo Proteomics Research Center ( Mayo Clinic College of Medicine , Rochester , MN ) . Polyclonal rabbit antiserum against 192–212 peptide KEEALCKGDFRPNVNEQPGS of PA5369 ( anti-PA5369 ) was produced in rabbits ( SynPep Corporation , Dublin , CA ) . Anti- PA5369 antibodies were affinity purified by AminoLink Plus Immobilization Kit ( Pierce ) using 192–212 peptide to create an affinity column . For immunoblot analysis , proteins were electrotransferred from SDS-polyacrylamide gels to PVDF membrane ( Immobilon-P , Millipore ) and primed with affinity pure anti-PA5369 antibodies at 1:1 , 000 dilution . Affinity pure F ( ab ) 2 fragments of anti-rabbit IgG conjugated with horseradish peroxidase ( Jackson Immunological Res Lab ) at 1:5 , 000 dilution was used as secondary antibody . Detection was performed using SuperSignal West Dura Extended Duration Substrate ( Pierce ) . P . aeruginosa strains were grown on PIA plates for 2 days , bacteria were harvested , suspended in PBS containing protease inhibitor cocktail ( Roche ) to create a cell density of 5 . 0 ( OD 600 nm ) in a total volume of 500 μl , and centrifuged at 6 , 000 rpm , 5 min . The pellet was resuspended in 500 μl PBS containing protease inhibitor cocktail , and vigorously vortexed for 2 min . Cell surface associated proteins were separated by centrifuging for 5 min at 5 , 000 g . After centrifugation , the supernatants were diluted ( 1:5 ) with carbonate-bicarbonate buffer ( Sigma ) , and 200 μl/well was used for coating Maxisorp Loose Immuno-modules ( Nunc ) . Plates were incubated at 4°C , overnight , washed with PBS , and unbound sites were blocked with 3% bovine serum albumin in PBS for 30 min at room temperature . Rabbit polyclonal affinity purified anti-5369 antibody ( 1:1 , 000 ) followed by HRP-labeled affinity purified F ( ab ) 2 fragments of anti-Rb IgG ( Jackson Immunological Research Laboratories ) ( 1:5 , 000 ) , and o-phenylaminediamine ( Sigma ) were used to detect PA5369-like protein at 450 nm optical density . Polyclonal affinity purified rabbit anti-PA5369 antibody against 192–212 peptide of PA5369 were used to isolate Fab fragment by ImmunoPure Fab Preparation Kit ( Pierce ) accordingly to manufacturer protocol . Transmission electron microscopic analysis was performed as previously described [60] with minor modifications . Briefly , bacteria were grown for 48 hours on PIA media with/without Gm , 20 μg/ml . A drop of water was deposited on the edge of colony , and a Formvar-coated copper grid was immediately floated on the drop for 30 s , then rinsed with TE buffer ( 10 mM Tris-HCl , 1 mM EDTA , pH 8 . 0 ) and stained with a 1% aqueous solution of uranyl acetate . Samples were examined under 300 KV with a FEI Tecnai F30 electron microscope . For immuno-gold labeling , 200 mesh formvar-coated nickel grids were rinsed with TE buffer , rehydrated with PBS for 30 min and blocked with 1% BSA for 30 min followed by transferring to anti-PA5369 antibodies diluted as 1:100 in 1% BSA . Incubation was allowed at a humidified chamber for 3 . 5 hours , at room temperature , followed by extensive washing with PBS , blocking with 0 . 1% BSA for 25 min , and incubating in the humidified chamber for 1 hour with goat anti-rabbit IgG conjugated with 10 nm gold particles ( TED PELLA ) at 1:10 dilution in 0 . 1% BSA . Grids were washed with PBS , fixed with 1% glutaraldehyde in PBS for 10 min , washed with water , and stained briefly with uranyl acetate and lead citrate . Air dried grids were examined under 300KV with FEI Tecnai F30 . The pa5369 gene was amplified using PAO1 DNA and primers forward 5369F EcoRI 5' CCGGAATTCGATGAAACTCAAGCGTTTG 3'and reverse 5369R XbaI 5' GCTCTAGACAAGTCACTGGATTACAG 3' and cloned in E . coli-P . aeruginosa shuttle vector pUCP24 [61] using EcoRI and XbaI restriction sites to create pUCP24/5369 where pa5369 is expressed from Plac promoter . The plasmid pUCP24/5369 was electroporated in ΔPA5369 to create strain Δ5369/5369 . The DNA containing pa0686 and pa0687 was amplified using PAO1 DNA and primers forward PA0686-744301F-EcoRI 5' CCGGAATTCGCGCGGTACCGTTGG 3' and PA0687-746961R-XbaI 5' GCTCTAGACGGACTACTGGACCAGTTG 3'and cloned in E . coli-P . aeruginosa shuttle vector pUCP24 [61] using EcoRI and XbaI restriction sites to create pUCP24/0686–0687 under regulation of Plac promoter . The plasmid pUCP24/0686–0687 was electroporated in the MPAO1 mutant strain ΔHxcR ( ΔPA0686 ) to create strain ΔHxcR/0686–0687 . The forward primer 5369F EcoRI 5' CCGGAATTCGATGAAACTCAAGCGTTTG 3'and reverse primer 5369R XbaI 5' GCTCTAGACAAGTCACTGGATTACAG 3' were designed based on the sequence of P . aeruginosa PAO1 genome and used to amplify genes using genome DNA isolated from strains MDR 1 , MDR 13 , and MDR 25 . The gene analogous to pa14–55410 was amplified using genome DNA of strain MDR1 , and primers 55410F EcoRI 5'CCGGAATTCGATGTACAAGCGCTCTCTGAT 3' and 55410R XbaI 5' GCTCTAGACAAG TTAGAGCGGACGGCCGAT 3' designed based on the sequence of P . aeruginosa PA14 genome . Amplified DNAs were cloned directly into pCR2 . 1 ( Invitrogen ) , and the sequence was obtained using standard M13 Forward and M13 Reverse primers on an Applied Biosystems 3730XL genetic analyzer ( University of Chicago , Cancer Research Center , DNA Sequencing & Genotyping Facility ) . Strains MPAO1 , MDR25R , MDR25S , and MDR1 were grown overnight on PIA and PIA supplemented with 10 mM K-Ph buffer , pH 7 . 0 , and collected directly in the RNA protect buffer ( Qiagen ) , and RNA isolation , DNA degradation , and cDNA synthesis were performed as previously described [27] . Real time PCR was performed on the ABI 7900HT System using SYBR Green qPCR SuperMix-UDG ( Invitrogen ) , cDNA , and respective primers: for citrate synthase PA1580 gene gltA , PA1580–434 5' TCTACCACGACTCCCTGGAC 3' and PA1580–590 5' TTTTCCGCGTAGTTCAGGTC 3'; for PstS PA5369 gene pstS , PA5369–148 5' ACTCTGGCCAACCTGATGAC 3' and PA5369–335 5'CCGTACTTCTGCTCGAAAGC 3'; for phosphate uptake regulator PhoU PA5365 gene phoU , PA5365–523 5' CGCGAACTGGTCACCTACAT 3' and PA5365–711 5' CTCGACCTCTTCCTTCATGC 3'; for low phosphate response regulator PhoB PA5360 gene phoB , PA5360–7 5' GGCAAGACAATCCTCATCGT 3' and PA5360–164 5'CAGTCGAGCAGGATCAGGTC 3'; for PstS PA0688 gene pa0688 , PA0688–693 5'GGTGAACATCAACAGCAACG 3' and PA0688–872 5'TAACCGACGATGGAGTAGCC 3'; for PstS analogous to PA14–55410 ) gene dinG , S1-DING-427 5' CTCTGCCGTTCAACAAGTCA 3' and S1-DING-604 5'CGGGTGAACAGTTCGGTAGT 3' , for HxcX atypical pseudopilin PA0682 gene hxcX , PA0682–299 5' AAGACGAGCAGGGCAAGTT 3'and PA0682–454 5'GTGCATAGGAGGCGAGTACC 3' . 0 . 5 μg of RNA after DNAse treatment was converted to cDNA in 20 μl of reaction mixture ( High Capacity cDNA Reverse Transcription kit , Applied Biosystems ) . The cDNA and RNA ( -RT control ) were diluted either as 1:50 ( for cS , pstS 5369 , phoU , phoB ) or 1:10 ( for hxcX ) or 1:500 ( for pa0688 and dinG ) , and 5 μl of diluted mixture was used as a template added to 7 . 5 μl of master mix containing as manufactured ( Invitrogen ) SYBR green , ROX , and respective primers . The amplification was run in 384 well plates . Expression levels were calculated based on differences in Ct levels . All primers were confirmed for DNA amplification using genome DNAs from MPAO1 and MDR clinical isolates prior the Real Time experiments . All experiments were approved by the Animal Care and Use Committee at the University of Chicago ( Protocol IACUC 71744 ) . Six-seven week old male C57BL6 mice were ordered from Harlan Spraque Dawley Animal facility and allowed at least four days for housing acclimation prior to experiments . The mouse model of gut-derived sepsis was performed as previously described [15] with following modifications . Mice drank either water ( no phosphate supplementation ) , or 0 . 2x PBS ( 2 mM Pi ) , or 1x PBS ( 10 mM Pi ) for 36 hours prior to lethality experiments . Animals were anesthetized ( ketamine 100 mg/kg , xylazine 10 mg/kg ) intraperitoneally and a 30% hepatectomy was performed on the left lobe of the liver , and the bacterial suspension of P . aeruginosa clinical isolates MDR25 or MDR1 were injected directly into the distal ileum and cecum with a fine high gauze needle . The abdomen was closed in two layers with suture , and mice were allowed to drink either water or PBS but were given no food for 48 hours . Animals were followed for mortality and sacrificed when they appeared septic and moribund . Eight-week-old male C57BL6 mice were given only water for 36 hours prior to hepatectomy . P . aeruginosa MDR25S and MDR25R were grown overnight in Pseudomonas broth ( PB ) containing Gm , 50 μg/ml . Overnight cultures were diluted as 1:500–1:250 in water containing 10% glycerol , and 20–50 μl were plated on PIA , Gm , 50 μg/ml . After 24–36 hours of growth , cells were collected directly from plates using an Olympus SZX16 stereo microscope to insure proper collection of rough and smooth colonies . The cells were diluted in water containing 10% glycerol to OD600 nm 0 . 25 , and 200 μl of bacterial suspension was injected in the cecum of mice immediately after hepatectomy . The abdomen was closed , and mice were allowed drinking water . Mice were followed for mortality and sacrificed when they appeared septic and moribund . Data were analyzed using Kaplan-Meier surviving curves and SPSS software employing the Long-rank ( Mantel-Cox ) test for significance . Statistical analysis of the data was performed with Student t-test using Sigma plot software and Kaplan-Meier survival curves using SPSS software .
PstS PA5369 ( Pseudomonas aeruginosa PAO1 ) , NP_254056; PA0688 , probable binding protein component of ABC transporter ( Pseudomonas aeruginosa PAO1 ) , NP_249379; PA14_55410 , Hypothetical , unclassified , unknown ( Pseudomonas aeruginosa UCBPP-PA14 ) , complete genome NC_008463; PhoB , two-component response regulator ( Pseudomonas aeruginosa PAO1 ) , NP_254047; HxcX , atypical pseudopilin ( Pseudomonas aeruginosa PAO1 ) , NP_249373; PA0686 , probable type II secretion system protein ( Pseudomonas aeruginosa PAO1 ) , complete genome NC_002516; PhoU , phosphate uptake regulatory protein ( Pseudomonas aeruginosa PAO1 ) , NP_254052; citrate synthase ( Pseudomonas aeruginosa PAO1 ) , NP_250271; Pseudomonas aeruginosa PAO1 , complete genome , NC_002516; ExoU ( Pseudomonas aeruginosa ) , AAC16023; human plasma phosphate-binding protein ( HPBP ) , P85173 . In the present study: PstS ( Pseudomonas aeruginosa MDR1 ) , EF601157; PstS ( Pseudomonas aeruginosa MDR13 ) , EF601158; PstS ( Pseudomonas aeruginosa MDR25 ) , EF601159; DING ( Pseudomonas aeruginosa MDR1 ) , EF616488 . | The resistance of bacteria to multiple antibiotics is a major problem in critically ill patients who often become colonized by highly lethal pathogens such as Pseudomonas aeruginosa . During the course of critical illness , as many as 50% of patients' intestinal tracts become colonized with P . aeruginosa , with as many as 30% of strains being resistant to multiple antibiotics . Concomitantly , critical illness is characterized by acute depletion of phosphate , which itself has been shown to be an independent predictor of infection-related mortality . In the present study we determined that during low phosphate conditions , highly virulent multi-antibiotic–resistant strains of P . aeruginosa isolated from critically ill patients produce an abundance of the phosphate-binding protein , PstS , located on extracellular finger-like structures . These PstS rich appendages participate in the binding of P . aeruginosa to intestinal lining cells and may allow P . aeruginosa to acquire phosphate from its host while remaining at arm's length from the host immune system . This clever tactic may be one example by which successful opportunistic pathogens such as P . aeruginosa survive within complex ecological niches such as the intestinal tract and harm their hosts during the course of critical illness . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods",
"Supporting",
"Information"
] | [
"infectious",
"diseases",
"gastroenterology",
"and",
"hepatology",
"microbiology",
"animals",
"eubacteria"
] | 2008 | Structure–Function Aspects of PstS in Multi-Drug–Resistant Pseudomonas aeruginosa |
During meiosis , the maternal and paternal homologous chromosomes must align along their entire length and recombine to achieve faithful segregation in the gametes . Meiotic recombination is accomplished through the formation of DNA double-strand breaks , a subset of which can mature into crossovers to link the parental homologous chromosomes and promote their segregation . Breast and ovarian cancer susceptibility protein BRCA1 and its heterodimeric partner BARD1 play a pivotal role in DNA repair in mitotic cells; however , their functions in gametogenesis are less well understood . Here we show that localization of BRC-1 and BRD-1 ( Caenorhabditis elegans orthologues of BRCA1 and BARD1 ) is dynamic during meiotic prophase I; they ultimately becoming concentrated at regions surrounding the presumptive crossover sites , co-localizing with the pro-crossover factors COSA-1 , MSH-5 and ZHP-3 . The synaptonemal complex and PLK-2 activity are essential for recruitment of BRC-1 to chromosomes and its subsequent redistribution towards the short arm of the bivalent . BRC-1 and BRD-1 form in vivo complexes with the synaptonemal complex component SYP-3 and the crossover-promoting factor MSH-5 . Furthermore , BRC-1 is essential for efficient stage-specific recruitment/stabilization of the RAD-51 recombinase to DNA damage sites when synapsis is impaired and upon induction of exogenous damage . Taken together , our data provide new insights into the localization and meiotic function of the BRC-1–BRD-1 complex and highlight its essential role in DNA double-strand break repair during gametogenesis .
The genetic information encoded by DNA must be accurately copied and transmitted from one generation to the next . In somatic cells , DNA is duplicated and equally partitioned into daughter cells via mitosis , whereas in germ cells , which give rise to gametes , chromosome segregation relies on meiosis , a specialized cell division mechanism which produces haploid cells from diploid progenitors . Meiosis requires a unique programme of finely regulated events before cell division to accomplish faithful chromosome segregation . Cognate paternal and maternal chromosomes ( homologous chromosomes ) find each other ( homologous pairing ) and then fully align; the interaction is stabilized by formation of the synaptonemal complex ( SC ) . Ultimately , exchange of DNA ( recombination ) between the homologues chromosomes establishes physical connections , which are essential for faithful segregation [1 , 2] . The Caenorhabditis elegans gonad is a powerful system for studying chromosomes during both mitosis and meiosis because of the cytological accessibility and the spatio-temporal organization of nuclei into all prophase I stages [3] . Morphological changes to chromosomes mark the engagement of key steps in meiotic progression . At meiotic onset , chromatin adopts a clustered , “half-moon” shape , reflecting chromosome movement and reorganization [4–6] . This structure marks the transition zone ( corresponding to the leptotene–zygotene stages ) . Once homologues are aligned , a tripartite proteinaceous structure called synaptonemal complex ( SC ) is formed between each homologue pair to allow genetic exchange during CO-dependent DNA repair [1 , 2 , 7–10] . DNA recombination is initiated by the deliberate induction of DNA double-strand breaks ( DSBs ) by the topoisomerase II-like enzyme , SPO-11 [11 , 12] . In all species , the number of DSBs largely exceeds the final number of COs , suggesting that many DSBs are repaired via pathways such as inter-sister repair ( IS ) or synthesis-dependent strand annealing [13] . In C . elegans , most chromosomes receive one CO during meiosis [14] , and this depends on the function of many proteins , including the MSH-4/MSH-5 heterodimer ( orthologues of the yeast and mammalian MutSγ complex components , MSH4/MSH5 ) [15–18] , the cyclin COSA-1 ( orthologue of mammalian CNTD1 ) [19 , 20] and the E3 SUMO-ligase ZHP-3 ( orthologue of yeast Zip3 ) [21–23] . CO formation is abolished in absence of DSBs ( e . g . in spo-11 mutants ) or synapsis; however , unlike in other model systems , lack of DNA breaks does not prevent SC formation in C . elegans [7 , 11] . Meiotic DSB repair also relies on RAD-51-mediated repair in C . elegans [24 , 25]: the RAD-51 recombinase localizes to discrete chromatin-associated foci starting in the transition zone and peaking in early pachytene; RAD-51 disengages from DNA in mid-pachytene [7] . Markers of aberrant RAD-51 loading , such as increased foci number and/or extended accumulation , are bona fide indicators of defective DSB processing and recombination . CO induction triggers reorganization of the SC components into distinct domains on bivalents ( pairs of homologous chromosomes held together by a chiasma ) : the central elements are confined to the short arm ( containing the CO ) and the axial elements can be recruited at both the long and the short arm ( i . e . HTP-3 and HIM-3 ) or specifically targeted to the long arm ( HTP-1/-2 and LAB-1 ) [26–30] . This reorganization is particularly evident during diplotene , at which stage bivalents progressively condense and appear as six DAPI-stained bodies in diakinesis nuclei , which are a read-out for the successful execution of prophase I events ( aberrant structures include achiasmatic chromosomes ( univalents ) or fused/fragmented chromatin masses [11 , 16 , 31] ) . The breast and ovarian cancer susceptibility protein BRCA1 and its obligate heterodimeric partner BARD1 form an E3 ubiquitin ligase module ( the BCD complex ) , the functions of which have been extensively studied in mammalian mitotic cells [32] . In this system , it has been shown that the BRCA1–BARD1 heterodimer promotes homologous recombination ( HR ) during the S–G2 stages , by both favouring extended DNA break resection and preventing the non-homologous end joining ( NHEJ ) -promoting factor 53BP1 [33] from binding to the site of ongoing DNA repair . Moreover , the activity of the BCD complex also enhances BRCA2 and RAD51 loading at DNA damage sites to elicit accurate DNA repair [32] . BRCA1-null mutants are embryonic lethal in mammals , thus hindering the study of this factor in gametogenesis [34–44] . Mouse mutants containing hypomorphic and gain-of-function BRCA1 alleles show increased apoptotic cell death during spermatogenesis , as well as reduced loading of the pro-CO factor MSH4 and a severe delay in MLH1 focus formation during oogenesis [45] . C . elegans brc-1/BRCA1 and brd-1/BARD1 mutants are viable and fertile , albeit with increased DNA damage-dependent apoptosis during oogenesis and SPO-11-dependent accumulation of RAD-51 foci , suggesting a defect in processing meiotic recombination intermediates [46 , 47] . Importantly , blocking brc-1-brd-1 function in CO-defective mutants leads to the formation of aberrant chromatin bodies in diakinesis nuclei , underscoring the importance of BRC-1 in the IS repair pathway [47] . Here we report that in the C . elegans germline , unlike in mammalian systems [48 , 49] , BRC-1 and BRD-1 are abundantly expressed throughout meiotic prophase I and display a dynamic localization pattern in germ cells , switching from nucleoplasmic expression in early meiotic stages to SC association in pachynema , where they become progressively enriched at the short arm of the bivalent . We provide in vivo evidence that both BRC-1 and BRD-1 form complex ( es ) with both MSH-5 and the SC central element , SYP-3 . Localization of BRC-1 and BRD-1 in germ cells is differently regulated by synapsis and CO formation . Finally , we show that the BCD complex promotes stage-specific RAD-51 loading when SC formation is impaired and upon exogenous DNA damage induction . Similar findings are reported by Li and colleagues in the accompanying manuscript . Taken together , our data highlight multiple functions of the BRC-1–BRD-1 heterodimer during gametogenesis .
To gain insight into BRC-1 and BRD-1 function during gametogenesis , we analyzed their localization patterns during meiotic prophase I . To this end , we tagged the endogenous brc-1 locus with a 5´-GFP or a 3´-HA tag and the brd-1 locus with a 3´-HA tag using a CRISPR/Cas9 approach [50 , 51] . BRD-1 detection was performed by employing a previously characterized antibody as well [46 , 52] . Functionality of fusion proteins was assessed by exposing the tagged animals to ionizing radiation ( IR ) : as previously reported [46] , brc-1 and brd-1 mutants were sterile , whereas GFP::brc-1 , brc-1::HA and brd-1::HA worms responded to IR in a similar way as wild-type animals , thus proving that the tagged proteins are fully functional ( Fig 1A ) . Using a recently established method for isolating germline-enriched proteins [53] involving protein fractionation and western blot analysis , we showed that BRC-1::HA is enriched in the nucleus: most was in the soluble nuclear pool fraction and a smaller proportion was chromatin bound ( Fig 1B ) . Interestingly , unlike in whole-cell extracts , BRC-1::HA was detected as a doublet in fractionated samples , suggesting that a less abundant isoform becomes detectable after enrichment with this extraction method or perhaps the presence of a post-translational modification . Previous reports indicate that during mouse meiosis , BRCA1 localizes to nascent SC elements in leptotene/zygotene stages; in pachytene cells , it is exclusively located at asynapsed region of the XY-sex body during spermatogenesis or on asynapsed chromosomes during oocyte meiosis [48 , 49 , 54] . In contrast , in the C . elegans germline both BRC-1 and BRD-1 were expressed in all nuclei during meiotic prophase I ( Figs 1C and S1A ) and , as expected , largely co-localized ( Fig 1D ) . Next , we sought to investigate whether , as shown in mammals [55] , BRC-1 and BRD-1 display loading interdependency in nematodes . BRD-1 was neither detected in brd-1 ( gk297 ) , proving antibody specificity , nor in the brc-1 ( tm1145 ) mutants by immunofluorescence ( S1B Fig ) . However , western blot analysis surprisingly revealed complete lack of BRD-1 expression in brc-1 ( tm1145 ) mutant worms ( S1C Fig ) , which prompted us to investigate the brd-1 locus in brc-1 ( tm1145 ) mutant worms . Genotyping for the brd-1 ( dw1 ) allele [52] revealed the presence of this deletion in the brc-1 ( tm1145 ) genetic background ( S1C Fig ) , showing that the DW102 strain , broadly used in the community as the brc-1 ( tm1145 ) single mutant , contains the brd-1 ( dw1 ) deletion in addition to the brc-1 ( tm1145 ) allele , both closely linked on chromosome III . In order to circumvent this , we then generated the brc-1::HA brd-1 and the brd-1::HA brc-1 mutant backgrounds and analyzed HA staining . BRC-1::HA staining was not detectable in brd-1 ( gk297 ) mutants ( S1D Fig ) , showing that as in mammals , BRD-1 is essential for BRC-1 loading in the C . elegans germline . However , BRD-1::HA was normally recruited in brc-1 ( tm1145 ) mutant germlines ( S1D Fig ) . Irradiation experiments showed that brc-1 ( tm1145 ) brd-1::HA worms displayed a response similar to neither WT nor brd-1 null mutants ( Fig 1A ) , suggesting that this mutation either does not impair BRD-1 loading or that tm1145 might be a “separation of function” allele of brc-1 . To unambiguously clarify loading and functional dependencies between the two members of the BCD complex , we generated a full knock-out of brc-1 by CRISPR , in which we deleted the entire brc-1 locus: brc-1 ( KO ) animals displayed similar levels of embryonic lethality as brd-1 nulls upon IR ( Fig 1A ) and did not show detectable levels of BRD-1::HA or endogenous BRD-1 by immunofluorescence ( S1E Fig ) . Western blot analysis revealed presence of BRD-1::HA protein in brd-1::HA brc-1 ( KO ) extracts , confirming the loading impairment ( S1F Fig ) . Interestingly , both BRC-1::HA and BRD-1::HA displayed reduced levels in the reciprocal null mutant backgrounds compared to the relative control animals ( 36 . 3% BRC-1::HA in brd-1 mutants and 45 . 7% BRD-1::HA in brc-1 ( KO ) mutants ) , suggesting that stability of both BRC-1 and BRD-1 is reduced when the BCD complex cannot be assembled , which was similarly observed in mouse models [56] . In conclusion , our data indicate that loading of BRC-1 and BRD-1 is mutually dependent and that activity of the BCD complex relies on functional integrity of both its members . Intriguingly , at the transition between mid- and late- pachytene , BRC-1 and BRD-1 staining switched from a rather diffuse to a discrete linear pattern along the chromosomes; in late pachytene nuclei , BRC-1 and BRD-1 progressively retracted into six short “comet-like” structures ( Figs 1C , 1D , S1A and S1B ) , a specific pattern indicating localization to both CO sites and the short arm of bivalent [7 , 8 , 21 , 22 , 57 , 58] . To assess whether the BCD complex is indeed recruited to the short arm of the bivalent , we co-stained brc-1::HA germ lines with antibodies directed against the central element of the SC , SYP-1 [8] and the axial protein HTP-3 [59] . As shown in Fig 2A , BRC-1 co-localized with SYP-1 in late pachytene nuclei , confirming that the BCD complex becomes gradually concentrated at the short arm of the bivalents . Strikingly , BRC-1 enrichment at these regions occurred earlier than observed for SYP-1 ( Fig 2A ) , as six robust SYP-1 stretches were seen only at diplotene stage . At meiosis onset , the PLK-2 polo-like kinase is enriched at the nuclear envelope attachment sites of chromosome ends , where it promotes homologous pairing and synapsis [60 , 61] . In late pachytene , PLK-2 re-locates to discrete domains along the SC , marking local enrichment of recombination factors [62 , 63] . PLK-2 redistribution also occurs before SYP-1 redistribution to the short arm of the bivalent and influences the SC structure [62–64] . Given the similar localization kinetics of BRC-1 , we co-stained PLK-2 and BRC-1 ( Fig 2B ) and found that regions enriched for BRC-1 fully overlapped with the PLK-2 staining pattern in late pachytene and diplotene . Thus , we wondered whether BRC-1 localization dynamics required PLK-2 function . Analysis of BRC-1::HA staining in plk-2 null mutants revealed that BRC-1 association with the SC appeared drastically weakened , although a delayed formation of BRC-1 tracks occurred at the diplotene entry ( Fig 2C ) . Moreover , rather than a gradual re-localization at the short arm as observed in the controls , BRC-1::HA abruptly formed distinct foci and seemingly high levels of mis-localized protein were observed within the nucleus ( Fig 2C ) . We can conclude that the BCD complex is ubiquitously expressed during meiotic prophase I and co-localizes with PLK-2 in pachytene nuclei . Furthermore , PLK-2 promotes progressive enrichment of BRC-1 to the short arm of the bivalent prior to SYP-1 recruitment . In C . elegans , formation of inter-homologue COs depends on several proteins , such as the COSA-1/CNTD1 cyclin [20] , the MutSγ heterodimer , MSH4/MSH-5 [15 , 16] and the ZHP-3 E3 SUMO-ligase [22] . MSH-5 and ZHP-3 are detected at early meiotic stages , with the former accumulating in many foci ( these are probably recombination intermediates with both CO and non-CO ( NCO ) outcomes ) and the latter localizing along the SC [20–22] . COSA-1 is prominently detected at mid–late pachytene transition as six foci ( one CO for each homologue pair ) , which also contain MSH-5 and ZHP-3 [20] . Since we observed BRC-1 and BRD-1 recruitment to the short arm of bivalents ( chromosome subdomains caused by the formation of CO intermediates [26 , 27 , 29] ) , we wondered whether local enrichment of the BCD complex coincides with the regions labeled with pro-CO factors . Comparison of the localization dynamics of GFP::COSA-1 and BRC-1::HA showed that BRC-1 starts to become concentrated concomitantly with enhanced COSA-1 loading in mid-late pachytene nuclei and defines a discrete area which later also contains SYP-1 ( Fig 3A ) . We obtained the same localization pattern by monitoring BRD-1 loading ( S2 Fig ) . Furthermore , staining with anti-ZHP-3 antibody [21] also revealed full co-localization with BRC-1 ( Fig 3B ) . To evaluate BRC-1 co-localization with MSH-5 , we tagged the endogenous msh-5 locus with a 5′ GFP tag by CRISPR/Cas9 . The tagged line was fully functional , as it showed normal levels of fertility ( Fig 3C ) , suggesting that GFP::MSH-5 is competent in promoting CO formation . Similar to ZHP-3 and COSA-1 , BRC-1::HA labeled the entire chromosomal region bearing the GFP::MSH-5 foci ( Fig 3C ) . We performed structured illumination microscopy ( SIM ) to further analyze BRC-1 association with the CO site . For this , we added a 5′ OLLAS-tag to the endogenous cosa-1 locus [65 , 66] . This fully functional line ( Fig 3C ) was recombined into brc-1::HA worms and co-stained for OLLAS ( COSA-1 ) , BRC-1 and SYP-1 . This further confirmed BRC-1 enrichment around COSA-1-labeled CO sites ( Fig 3D ) and due to the higher resolution provided by the SIM microscopy , we could appreciate in these nuclei that BRC-1 decorates the region of the SC embracing the putative recombination site; thus , it appears to surround , rather than overlapping with , COSA-1 ( Fig 3E ) . Given their spatial association with both CO factors and the SC , we wondered whether BRC-1 and BRD-1 formed protein complexes with these factors in vivo . We performed pull-down experiments using the brc-1::HA; GFP::msh-5 strain ( Fig 3 ) and crossed brc-1::HA into worms expressing a single-copy insertion transgene encoding a largely functional GFP::SYP-3 protein [67] . The same was done to generate brd-1::HA; GFP::msh-5 and GFP::syp-3; brd-1::HA strains . Cytosolic , nuclear soluble and chromatin-bound protein fractions [53] were produced from all the above mentioned strains and both nuclear fractions were pooled for immunoprecipitation experiments . Western blot analysis using anti-HA antibodies on GFP pull-downs revealed co-immunoprecipitation of both BRC-1::HA and BRD-1::HA with GFP::MSH-5 and GFP::SYP-3 ( Fig 4A and 4B ) . Furthermore , we also generated i ) a GFP::brc-1; msh-5::2xHA strain to perform reciprocal IPs and assess whether co-IP of BRC-1 and MSH-5 was still occurring upon tag swapping , and ii ) a GFP; brc-1::HA strain in which the GFP was expressed , in order to rule out non-specific binding . MSH-5::2xHA co-immunoprecipitated also with GFP::BRC-1 ( Fig 4C ) , recapitulating , and further validating the result obtained with the brc-1::HA; GFP::msh-5 line ( Fig 4A ) . Importantly , no interaction was observed between BRC-1::HA and the GFP alone ( Fig 4C ) , confirming specificity of the interactions ( Fig 4A ) . Therefore , we can conclude that BRC-1 and BRD-1 form complex ( es ) with both MSH-5 and SYP-3 proteins in vivo . These results reveal a previously unknown physical interaction of the BCD complex with pro-CO factors , as well as SC components , highlighting a possible role for BRC-1–BRD-1 at the interface between synapsis and recombination . To assess whether BRC-1–BRD-1 redistribution depends on CO establishment , we generated a brc-1::HA; spo-11 mutant strain to monitor BRC-1::HA loading in absence of meiotic DSBs , which are essential for inducing CO formation . We found that BRC-1 and ZHP-3 retraction toward the CO site was largely impaired ( Fig 5A ) although sporadically their redistribution was observed in late pachytene nuclei ( S3 Fig ) , most likely due to formation of spontaneous or pre-meiotic DSBs which are proficient in triggering recruitment of CO-promoting factors and therefore elicit remodelling of the SC components and give rise to a chiasma [20] . Exogenous DSB induction is sufficient to temporarily restore COSA-1 loading and therefore chiasmata formation in spo-11 mutants [11 , 20 , 64] . Thus , we investigated whether γ-irradiation could rescue failure in BRC-1 redistribution . We exposed brc-1::HA; spo-11 mutant worms to 20 Gy and analyzed BRC-1 and ZHP-3 loading at 8 hours post-irradiation: at this time point , all late pachytene nuclei in spo-11 mutants display six COSA-1 foci , suggesting that CO designation is fully rescued [20] . In the irradiated samples , ZHP-3 was retracted towards the CO site and , consistently , BRC-1 also became concentrated on the short arm ( Fig 5B ) . Based on these data , we conclude that BRC-1 and BRD-1 localize to the short arms of bivalents and their reorganization in mid-pachytene nuclei is dependent on CO establishment . Given that CO establishment triggers BRC-1–BRD-1 redistribution , we sought to analyze their localization in mutants impaired at different steps of CO formation . As already mentioned , an absence of DSBs leads to a lack of recombination , which largely prevents BRC-1 and BRD-1 retraction to the short arms of bivalents . We therefore asked whether impaired DNA repair by HR , but not by DSB induction , influences BRC-1 and BRD-1 localization . To address this , we crossed brc-1::HA into the msh-5 mutant , where conversion of recombination intermediates into mature CO products is prevented [7 , 16] . In msh-5 mutants , BRC-1 accumulated along the SC but retraction was not observed ( Fig 5C ) , similar to the localization pattern observed in spo-11 ( Fig 5A ) . Then , we analyzed BRC-1::HA staining in rad-51 mutants , which have normal SC assembly but no homologous DNA repair due to lack of RAD-51-dependent strand displacement and invasion of the homologous chromosome [24 , 25] . Interestingly , BRC-1 had a rather punctate staining pattern but despite this , a strong association with SYP-1 in chromosome subdomains was observed in nuclei exiting the pachytene stage ( we also observed this in msh-5 mutants ) ( Fig 5C ) . We observed a similar pattern of BRD-1 localization in com-1 mutants ( S4 Fig ) : here , interfering with DSB resection impairs RAD-51 loading and therefore abolishes CO formation [68] . These results suggest that a lack of COs per se impairs redistribution of the BCD complex in late pachytene cells without perturbing loading along the SC . However , in mutants such as rad-51 that are defective in the early steps of recombination , BRC-1–BRD-1 association with the SC is also dramatically reduced . Next , we sought to analyze whether BRC-1 and BRD-1 loading is regulated by synapsis . We analyzed BRC-1::HA loading under complete and partial absence of SC , as well as in mutants in which synapsis occurs between non-homologous chromosomes . The central portion of the SC is formed by several proteins ( SYP-1–4 ) which are loaded in an interdependent manner; thus , all are necessary to establish synapsis [7 , 8 , 57 , 58] . In the syp-2 synapsis-null mutant [7] , BRC-1::HA had a rather punctate staining pattern throughout meiotic prophase I ( Fig 6A ) . Strikingly , unlike in the wild type , where BRC-1 starts to spread along the SC immediately after the disappearance of RAD-51 , in syp-2 mutants BRC-1 foci remained in close proximity to and co-localized with RAD-51 in mid- and late-pachytene nuclei ( Fig 6A and 6B ) . In C . elegans , a family of zinc-finger nuclear proteins connects chromosome-specific ends ( i . e . pairing centres ) to the nuclear envelope to promote chromosome pairing and synapsis [69 , 70] . ZIM-2 and HIM-8 bind to the ends of chromosomes V and X , respectively . Therefore , chromosome V is asynapsed in zim-2 mutants and chromosome X is asynapsed in him-8 mutants . We asked whether a partial deficiency in synapsis establishment ( affecting only one chromosome pair ) also changed BRC-1 loading dynamics . Analysis of BRC-1::HA expression in him-8 and zim-2 mutants revealed lack of BRC-1 on unsynapsed chromosomes pairs , despite normal loading along the SC and retraction towards the CO site in the remaining bivalents ( Fig 6C and 6D ) , suggesting that local synapsis defects do not impair global BRC-1 loading . Lastly , we analyzed BRD-1 loading in two mutants with aberrant SC assembly . HTP-1 is a HORMA-domain-containing protein essential to achieve normal levels of pairing and preventing SC assembly between non-homologous chromosomes , while PROM-1 is an F-box protein involved in promoting meiotic entry and homologous pairing . Both htp-1 and prom-1 mutants display extensive SYP-1 loading between non-homologous chromosomes as well as asynapsed chromosome regions; consequentially , chiasmata formation is severely impaired [26 , 71] . Remarkably , the degree of BRD-1 co-localization with SYP-1 was extremely reduced in both htp-1 and prom-1 mutants , with most BRD-1 detected as bright agglomerates within the nucleus ( S5A Fig ) . The same localization pattern was observed for BRC-1::HA in htp-1 mutants ( S5B Fig ) . Thus , we conclude that BRC-1 and BRD-1 redistribution during meiotic progression requires CO establishment , is tightly regulated by the SC and does not follow SC installation between non-homologous chromosomes . BRC-1 is dispensable for establishing synapsis and chiasmata; however , brc-1 mutant germlines have a higher number of and more persistent RAD-51-labeled recombination intermediates compared with the wild type [46 , 47] . Impaired BRC-1 localization , and probably also impaired function , in CO-defective mutants leads to the formation of abnormal chromosome structures in diakinesis nuclei , possibly due to deficient IS repair [47] . DSB repair during meiosis is channelled into both CO and NCO pathways . Since it has been suggested that BRC-1 might preferentially function in NCOs [47] , we investigated whether other factors involved in resolving the recombination intermediates required for both CO and NCO repair might also be affected . In somatic cells , the BTR complex , formed by BLM , RMI1 and TOP3A , mediates efficient resolution of recombination intermediates by promoting the dissolution of double Holliday junctions to yield non-CO products [72–74] . The C . elegans RMI1orthologue , RMH-1 , accumulates in many foci during meiotic prophase , possibly labelling all recombination intermediates . At late pachytene transition , the number of RMH-1 foci is reduced to roughly six per nucleus which co-localize with COSA-1 , MSH-5 and ZHP-3 . Lack of RMH-1 causes a drastic reduction in chiasmata formation due to impaired COSA-1 and MSH-5 loading . However , in CO-deficient backgrounds such as cosa-1 , msh-5 and zhp-3 mutants , RMH-1 is still recruited in early pachytene but is not retained until late pachytene . Therefore , it has been postulated that RMH-1 functions in both the CO and NCO pathways [75] . We scored COSA-1 , MSH-5 and RMH-1 nuclear localization in brc-1 mutants in nuclei in the transition zone to late pachytene stage . Interestingly , GFP::MSH-5 accumulation was mildly , although significantly , reduced in early and mid-pachytene , with a similar , albeit less prominent , behaviour for GFP::RMH-1 ( Fig 7A and 7B ) . By late pachytene however , both proteins had been recruited into six foci , together with COSA-1 , suggesting that BRC-1 might influence early processing of recombination intermediates , although formation of chiasmata was not affected . Given that BRC-1 and BRD-1 loading is regulated by synapsis and the establishment of COs , and that a lack of BRC-1 might affect the processing of NCOs rather than COs , we next assessed the effects of BRC-1 depletion in genetic backgrounds defective in chiasmata formation , which hence rely solely on NCOs to repair meiotic DSBs . We analyzed DAPI-stained bodies in diakinesis nuclei from cosa-1 brc-1 brd-1 , brc-1 brd-1; msh-5 and brc-1 brd-1; syp-2 mutants and observed the presence of aberrant chromatin structures ( Fig 8A ) , as previously reported [46 , 47] . As abnormalities in diakinesis nuclei can result from impaired RAD-51-dependent repair of meiotic DSBs [24 , 31 , 76] , we sought to analyze whether lack of function of the BCD complex altered RAD-51 dynamics . To this end , we quantified RAD-51 in the above-mentioned mutant backgrounds . Failure to convert recombination intermediates into mature CO products has been linked to increased RAD-51 levels and its delayed removal during meiotic prophase due to either excessive DSB induction or slower processing of recombination intermediates [5 , 6 , 15 , 16] , which are eventually channelled into alternative repair pathways ( e . g . IS repair ) [7] . In fact , cosa-1 , msh-5 and syp-2 mutants all accumulated high levels of RAD-51 , which disengaged from chromatin in mid- and late-pachytene ( Fig 8B ) [7 , 20] . Remarkably , removal of BRC-1 from cosa-1 and msh-5 mutants had different effects on RAD-51 dynamics compared to syp-2: in all these strains , there were fewer RAD-51 foci in early pachytene ( Fig 8B , zone 5 ) compared with cosa-1 , msh-5 and syp-2 single mutants; however , in cosa-1 brc-1 brd-1 and brc-1 brd-1; msh-5 mutants RAD-51 accumulation was dramatically prolonged until diplotene entry whereas in the brc-1 brd-1; syp-2 mutants RAD-51 staining was overall dramatically reduced ( Fig 8B , S1–S3 Tables ) . Aberrant chromosome structures occurred at a particularly high frequency in brc-1 brd-1; syp-2 mutants , consistent with the severe reduction in RAD-51 loading in pachytene nuclei ( Fig 8A ) . To be efficiently loaded to the single-stranded DNA ( ssDNA ) tails generated after resection , RAD-51 must be exchanged with RPA ( RPA-1 in worms ) , which coats ssDNA tails to stabilize them and prevent DNA from self-winding [77 , 78] . Given the altered dynamics of RAD-51 expression , we decided to analyze RPA-1 [79] to assess whether ssDNA was properly formed and processed . RPA-1 highly accumulated in brc-1 brd-1; syp-2RNAi , forming bright , discrete foci in both early and late pachytene cells ( S6A Fig ) . This indicates that blocking BRC-1 function in synapsis-deficient mutants prevents efficient RAD-51 loading at meiotic DSBs , causing accumulation of unrepaired ssDNA as evidenced by RPA-1 foci formation . The cosa-1 brc-1 brd-1 mutants did not show RPA-1 accumulation in early pachytene and late pachytene cells displayed occasional , weak foci , suggesting perhaps only a mild delay in the processing of recombination intermediates which eventually takes place faithfully , as shown by the formation of largely normal DAPI-bodies in diakinesis nuclei in the brc-1 brd-1; msh-5 mutants ( Figs S6B and 8A ) . Thus , in CO-defective mutants , BCD-dependent regulation of RAD-51 dynamics is altered by the presence of the SC . Exposure of brc-1 and brd-1 mutants to IR causes dose-dependent hypersensitivity which eventually culminates in full sterility , possibly due to the formation of highly unstructured chromatin bodies in diakinesis nuclei [46] . These structures resemble those formed upon BRC-2/BRCA2 depletion , which in worms is essential for RAD-51 loading [31 , 76] , and COM-1/Sae2 depletion , which promotes DSB resection [68 , 80] . Both mutants lack RAD-51 recruitment onto DNA during meiotic prophase I . We therefore sought to investigate whether the aberrant chromatin masses observed in irradiated brc-1 brd-1 mutants were caused by impaired RAD-51 recruitment . We analyzed RAD-51 and RPA-1 loading at two different time points post-irradiation . At 8h post-irradiation , we observed a dramatic reduction in RAD-51 focus formation specifically in mid- and late-pachytene nuclei of brc-1 brd-1 mutants , along with enhanced RPA-1 levels ( Fig 9B and 9C ) . At 24 hours post-irradiation , RAD-51 was still markedly reduced ( especially in mid-pachytene stage ) while RPA-1 levels in brc-1 brd-1; [rpa-1::YFP] mutants were comparable to the controls ( Fig 9B and 9C ) . Prompted by these results , we decided to analyze the loading dynamics of BRC-1::HA and RAD-51 after IR exposure to assess whether exogenous DSB formation affected the mutual spatio-temporal regulation of these proteins . Under homeostatic conditions , BRC-1 and RAD-51 localization did not overlap prior to BRC-1 enrichment in the SC , which occurs after RAD-51 disappearance ( S7A and S7B Fig ) . At 1 hour post-irradiation , BRC-1::HA started to form discrete chromatin-associated foci in pre-meiotic nuclei , often in close proximity to ( but not co-localizing with ) RAD-51 foci ( S7A and S7B Fig ) . Although abundant RAD-51 accumulation was triggered by IR exposure throughout the germline , BRC-1::HA levels were only modestly increased . However , western blot analysis revealed a shift in BRC-1::HA migration after IR which remained unchanged throughout the time course ( S7C Fig ) , suggesting that exogenous DNA damage might elicit post-translational modifications of BRC-1 . Western blot analysis also showed a slight increase in BRC-1::HA abundance , confirming our immunofluorescence data ( S7A Fig ) . Samples analyzed 8 hours after IR revealed robust BRC-1 and RAD-51 co-localization in nuclei residing in the mitotic tip; however , as at the earlier time point , no clear co-localization was observed in pachytene nuclei ( S7B Fig ) . At 24 hours post-irradiation , BRC-1::HA foci in the mitotic nuclei had largely disappeared and bright RAD-51 foci were observed only in enlarged , G2-arrested nuclei that were still undergoing repair; in contrast , bright RAD-51 foci co-localizing with BRC-1 were occasionally seen in non-arrested nuclei . Taken together , our observations revealed that BRC-1 accumulation in the germline is modulated by exogenous DNA damage and that the clear BRC-1 and RAD-51 co-localization observed only in mitotic nuclei was cell cycle dependent .
BRC-1 and BRD-1 display a highly dynamic localization pattern during meiotic prophase I progression , shifting from a rather diffuse accumulation at early stages to a robust association with the SC in late pachytene , which culminates in retention of the BCD complex at the short arm of the bivalent ( Figs 1–3 ) . Remarkably , accumulation of BRC-1–BRD-1 at specific chromosomal subdomains occurs prior to retraction of the SC central elements to those domains but is concomitant with recombination factor-dependent enrichment of PLK-2 at the SC ( Fig 2B ) [62 , 64] , suggesting that the BCD complex is actively targeted to the region surrounding the CO rather than passively recruited following SC remodelling . Importantly , recruitment of BRC-1–BRD-1 to the short arm of the bivalent depends on PLK-2 function , as chromatin association of BRC-1 is strongly impaired in plk-2 null mutants: enrichment of BRC-1::HA at the SC is extremely delayed and occurs only in a short region before diplotene entry in plk-2 mutants ( Fig 2C ) . Furthermore , rather than displaying a gradual retraction towards the short arm of the bivalent , BRC-1::HA abruptly formed discrete foci ( presumably at the CO sites which are still formed in absence of PLK-2 ) and was largely retained in the nucleoplasm ( Fig 2C ) . We envision a scenario where the localization of BRC-1 at the presumptive CO sites in plk-2 null mutants is accomplished via the physical interaction with the CO machinery ( Fig 4 ) . It has been shown that albeit reduced , COs still form in absence of PLK-2 , which are dependent on PLK-1 activity [60 , 61] . Therefore , we hypothesize that residual BRC-1 accumulation observed in plk-2 mutants might be dependent on PLK-1 . Our data favour a model in which the SC also exerts an essential role for the recruitment of the BCD complex onto the chromosomes and its later accumulation at the CO site , due to the local concentration of recombination factors . In fact , BRC-1 recruitment to the SC is not prevented in msh-5 or spo-11 mutants ( both of which are defective in CO formation but proficient in synapsis establishment ) . However , similar to ZHP-3 , BRC-1 fails to retract ( Fig 5A and 5C ) . Irradiation of spo-11 mutants restored BRC-1 and ZHP-3 redistribution to the short arms of bivalents ( Fig 5B ) , confirming that CO establishment per se is the key trigger of local BCD complex enrichment . Abrogation of synapsis dramatically changed the BRC-1 expression pattern: it remained punctate throughout meiotic prophase I and displayed extensive co-localization with RAD-51 specifically in late pachytene cells ( Fig 6A and 6B ) . However , in mutants in which only one chromosome pair was asynapsed , such as him-8 and zim-2 mutants , BRC-1 was not loaded onto the unsynapsed chromosomes but localization was normal on the remaining ones ( Fig 6C and 6D ) . Our data suggest that the SC alone is not sufficient for proper BRC-1 and BRD-1 loading but it rather cooperates with PLK-2 to regulate the function and localization dynamics of the BCD complex . In fact , plk-2 mutants still display extensive regions of synapsis throughout the germline , however recruitment of BRC-1 is aberrant . If successful loading of the BCD complex was solely dependent on the SC loading , then we would have expected to find unperturbed BRC-1 localization at the synapsed regions , which instead was not the case . This suggests that localization of BRC-1 and BRD-1 undergoes a complex and tightly controlled regulation . It was recently shown that PLK-2 plays a pivotal role in modulating the physical state of the SC in response to recombination and that absence of synapsis impairs PLK-2 redistribution from the nuclear envelope to chromosome subdomains [62–64] , which might explain the different BRC-1 localization patterns in syp-2 mutants . Different BRD-1 and BRC-1 localization patterns were as well observed in htp-1 and prom-1 mutants , both characterized by extensive non-homologous synapsis , in which the BCD complex accumulated in bright agglomerates in the nucleus ( S5A and S5B Fig ) . It is important to mention here , that htp-1 null mutants display an extremely reduced PLK-2 accumulation at the nuclear envelope at meiosis onset and absence of SC-associated PLK-2 in late pachytene nuclei [53] , suggesting , once again , that SYP loading per se is not sufficient to recruit BRC-1–BRD-1 onto the SC and that PLK-2 might exert further roles in regulating BRC-1 ( and by assumption , BRD-1 ) localization possibly through phosphorylation-dependent modifications . However , compared to plk-2 null mutants in which a residual accumulation of BRC-1 at the SC still occurs , in htp-1 mutants ( which also lack PLK-2 loading in late pachytene nuclei ) both BRC-1::HA and BRD-1 entirely fail to be recruited along the chromosomes , indicating that the requirements for proper localization of the BCD complex dwell in multiple layers of control and that ectopic polymerization of the SC between non-homologous chromosomes dramatically perturbs BRC-1 and BRD-1 association with the chromatin . Moreover , despite their clear enrichment at the putative CO sites , our data , as well as previous studies [46 , 47] , show that chiasmata formation occurs normally in absence of a functional BCD complex , indicating that recruitment of BRC-1-BRD-1 at CO-designation sites is not essential for eliciting COs but it might rather occur in response to their establishment . Intriguingly though , Li and colleagues show in the accompanying study that the recombination landscape is altered in brc-1 mutants , as recombination rate increases in the center of chromosomes at the expenses of the terminal regions , which instead bear the majority of recombination events in WT worms . Further , under compromised meiosis ( i . e . zim-1 mutants ) , BRC-1 can promote formation of extra-COs in the presence of chromosome pairs lacking a chiasma . This would indicate that BRC-1 and BRD-1 exert a regulatory activity on the recombination intermediates and can act as a switch in the choice of CO versus NCO pathway . Blocking BRC-1 function had opposing effects on the progression of recombination intermediates in cosa-1 and msh-5 , compared to syp-2 ( CO-defective ) mutants . RAD-51 accumulation was exacerbated in cosa-1 and msh-5 single mutants and largely suppressed in syp-2 mutants ( Fig 8 ) , leading to the formation of aberrant chromatin masses in diakinesis nuclei as previously reported [47] . Based on genetic data , BRC-1 function was previously postulated to be essential for IS repair of meiotic DSBs [47 , 81 , 82]; our data corroborate this model . In cosa-1 brc-1 brd-1 and brc-1 brd-1; msh-5 mutants , the presence of an intact SC might still impose a homologue-biased constraint for an inter-homologue , CO-independent pathway that relies on RAD-51-mediated repair but not on BRC-1 function . However , in the absence of synapsis , repair of recombination intermediates is probably channelled entirely through the IS repair pathway given that the sister chromatid is the only available repair template: in line with this , our data show that in fact SC depletion triggers association of BRC-1 with RAD-51 in late pachytene cells at presumptive repair sites , thereby likely promoting HR-mediated repair . We also observed fewer RAD-51 foci in brc-1 brd-1; syp-2 mutants during early pachytene , suggesting that BRC-1 is nonetheless required to ( directly or indirectly ) promote efficient RAD-51 loading at meiosis onset , although co-localization with RAD-51 in early stages might be very transient . We observed that lack of BRC-1 mildly , although significantly , impacts on the loading of recombination markers such as MSH-5 and RMH-1 in early pachytene , suggesting that even in the presence of the SC , BRC-1–BRD-1 function might be required to efficiently promote the processing of recombination intermediates . Moreover , in brc-1 brd-1 mutants exposed to exogenous DSB induction , RAD-51 is not efficiently retained in mid- and late-pachytene cells ( Fig 9 ) . This is not due to impaired resection , as shown by the abundant recruitment of RPA-1 , which stabilizes ssDNA . However , RAD-51 loading is comparable to controls in later stages , suggesting that stabilization , rather than loading per se , might require the action of the BCD complex . This is in line with the findings reported by Li et al . ( see accompanying manuscript ) . When we scored BRC-1 levels after exposure to IR , we detected a slight increase in abundance but a marked difference in protein migration on western blots ( S7A and S7C Fig ) , suggesting that exogenous DNA damage may promote post-translational modification of BRC-1 . Importantly , despite dramatically enhanced RAD-51 levels upon irradiation , we observed clear co-localization with BRC-1 only in mitotic cells and not during pachytene , once again confirming that these proteins co-localize only when the SC is indeed absent ( S7B Fig ) . Our findings suggest that the BCD complex responds to both synapsis and recombination and that the SC might act as a docking site for the BRC-1–BRD-1 complex to modulate its function in promoting meiotic DNA repair .
RNAi for syp-2 was performed employing the clone available in the Ahringer library . A single colony from a freshly struck glycerol stock on plates containing 100 mg/ml ampicillin and 12 . 5 mg/ml of tetracycline was inoculated in 20 ml of LB containing 100 mg/ml of ampicillin and grown overnight at 37°C . The following day , the bacteria were concentrated in 2 ml of LB containing 100 mg/ml ampicillin and 100μl of culture were seeded per plate , containing 100 mg/ml of ampicillin and 1mM IPTG . The same procedure was followed for the bacteria containing the pL4440 empty vector as a control . Plates were left at 37°C overnight to induce dsRNA and the following day , L4 [rpa-1::YFP] and brc-1; [rpa-1::YFP] worms were placed on the induced plates . F1 worms at L1 stage were picked and transferred onto freshly induced plates three days later . Worms were dissected 24h post L4 stage . The RNAi was performed at 20°C and only the germlines displaying 12 univalents in diakinesis nuclei , indicative of successful syp-2 depletion , were analyzed for YFP staining . For cytological analysis of whole-mount gonads , age-matched worms ( 20–24 hours post-L4 stage ) were dissected in 1× PBS on a Superfrost Plus charged slide and fixed with an equal volume of 2% PFA in 1× PBS for 5 min at room temperature . Slides were freeze-cracked in liquid nitrogen and then incubated in methanol -20°C for 5 min , followed by three washes in PBST ( 1× PBS , 0 . 1% Tween ) at room temperature . Slides were blocked for 1 hour at room temperature in PBST containing 1% BSA and then primary antibodies were added in PBST and incubated overnight at 4°C . Slides were then washed in PBST at room temperature and secondary antibodies were applied for 2 hours . After three washes 10 min each in PBST , 60μl of a 2 μg/ml stock solution of DAPI in water was added to each slide and stained for 1 min at room temperature . Samples were washed again for at least 20 min in PBST and then mounted with Vectashield . For detection of GFP::MSH-5 , worms were dissected and fixed in 1× EGG buffer containing 0 . 1% Tween instead of PBST . Detection of [RPA-1::YFP] was performed as previously described [83] . Primary antibodies used in this study were: mouse monoclonal anti-HA tag ( pre-absorbed on N2 worms to reduce non-specific binding; 1:100 dilution; Covance ) , rabbit anti-HA tag ( 1:250 dilution; Invitrogen ) , rabbit anti-BRD-1 ( pre-absorbed on brd-1 ( dw1 ) worms to reduce non-specific binding; 1:500 dilution ) [52] , chicken anti-SYP-1 ( 1:500 dilution ) [53] , guinea pig anti-HTP-3 ( 1:500 dilution ) [59] , mouse monoclonal anti-GFP ( 1:500 dilution; Roche ) , guinea pig anti-ZHP-3 ( 1:500 dilution ) [21] , rabbit anti-OLLAS tag ( pre-absorbed on N2 worms to reduce non-specific binding; 1:150 dilution; GenScript ) , rabbit anti-RAD-51 ( 1:10 , 000 dilution; SDIX ) and rabbit anti-PLK-2 ( 1:500 dilution ) [84] . Appropriate secondary antibodies were conjugated with Alexa Fluor 488 or 594 ( 1:500 dilution ) or with Alexa Fluor 647 ( 1:250 dilution ) . Images were collected as z-stacks ( 0 . 3 μm intervals ) using an UPlanSApo 100x NA 1 . 40 objective on a DeltaVision System equipped with a CoolSNAP HQ2 camera . Files were deconvolved with SoftWORx software and processed in Adobe Photoshop , where some false colouring was applied . Samples acquired by super-resolution microscopy ( Fig 3D and 3E ) were prepared as previously reported [62] without modifications and imaged with a DeltaVision OMX . For quantification of RPA-1::YFP and RAD-51 in Fig 9 , samples were acquired with same settings and identically adjusted in Fiji . Gonads were divided into four equal regions starting from transition zone and ending before diplotene entry . A circle with a fixed area was drawn in Fiji and intensity of fluorescence was scored in each nucleus for all regions . For whole-cell protein extraction , 200 age-matched animals ( 24 hours post-L4 stage ) were picked into 1× Tris-EDTA buffer ( 10 mM Tris pH 8 , 1 mM EDTA ) containing 1× protein inhibitor cocktail ( Roche ) and snap-frozen in liquid nitrogen . After thawing , an equal volume of 2× Laemmli buffer was added . Samples were boiled for 10 min , clarified and separated on pre-cast 4–20% gradient acrylamide gels ( Bio Rad ) . Fractionated protein extracts for western blotting and immunoprecipitation were prepared as previously reported [53] . Western blotting used 50 μg of protein samples from each fraction , whereas for immunoprecipitation assays at least 1 mg of extract from pooled soluble nuclear and chromatin-bound fractions was used . For the inputs , 5% of the amounts used for IPs was run . Immunoprecipitation of GFP-tagged proteins was performed with agarose GFP-traps ( Chromotek ) . For all immunoprecipitation experiments , pre-equilibrated beads in buffer D ( 20% glycerol , 0 . 2 mM EDTA pH 8 , 150 mM KCl , 20 mM Hepes-KOH pH 7 . 9 , 0 . 2% Triton X-100 , supplemented with protease inhibitor cocktail from Roche ) , were incubated with the extracts over night at 4°C in mild agitation . The following day , beads were separated from immuno-depleted extracts , washed extensively in buffer D , re-suspended in 40 μl of 2x Laemmli Buffer ( Sigma ) and boiled for 10 minutes to recover immunocomplexes . Beads were pelleted by centrifugation at maximum speed for 1 min and surnatants were run in 1× SDS-Tris-glycine buffer on a pre-cast 4%-20% TGX gels ( BioRad ) . Proteins were transferred onto nitrocellulose membrane for 1 hour at 4°C at 100V in 1× Tris-glycine buffer containing 20% methanol . Membranes were blocked for 1 hour in 1× TBS containing 0 . 1% Tween ( TBST ) and 5% milk; primary antibodies were added into the same buffer and incubated overnight at 4°C . Membranes were then washed in 1× TBST and then incubated with appropriate secondary antibodies in TBST containing 5% milk for 1 hour at room temperature . After washing , membranes were incubated with ECL ( Amersham ) and developed with a ChemiDoc system ( BioRad ) . To detect phosphorylated CHK-1S345 , TBST containing 5% BSA instead of milk was used for both blocking and antibody dilution . The following antibodies were used for western blotting: mouse monoclonal anti-HA tag ( 1:1000 dilution; Cell Signalling ) , rabbit anti-HA tag ( 1:500 dilution; Invitrogen ) , anti-BRD-1 [52] ( 1:1000 dilution ) , chicken anti-GFP ( 1:4000 dilution; Abcam ) , mouse anti-GAPDH ( 1:10 , 000 dilution; Ambion ) , goat anti-Actin ( 1:3000; Santa Cruz ) , rabbit anti-Histone H3 ( 1:100 , 000 dilution; Abcam ) ; rabbit anti-phospho-CHK-1S345 ( 1:1000 dilution; Cell Signalling ) , HRP-conjugated anti-mouse ( 1:2500 dilution ) and anti-rabbit ( 1:25 , 000 dilution; both Jackson ImmunoResearch ) , HRP-conjugated anti-chicken and anti-goat ( both 1:10 , 000 dilution; Santa Cruz ) . Age-matched worms ( 24 hours post-L4 stage ) were exposed to the indicated dose of IR with a Gammacell irradiator containing a 137Cs source . For viability screening , irradiated worms were allowed to lay eggs for 24 hours and then removed; hatched versus unhatched eggs were scored the following day . For cytological analysis , worms were dissected and immunostained at the indicated times . CRISPR-Cas9 tagging A C-terminal HA-tag was inserted at the endogenous locus encoding the brc-1 gene by using a CRISPR-Cas9 based approach as in [85] . Briefly , a 2 , 335 base pairs region ( 8541 to 10875 from the ATG ) of the brc-1 locus was amplified by PCR from genomic DNA and cloned in pCR2 . 1 vector ( TA cloning , Invitrogen ) . The plasmid obtained was used as a template to insert a 27 base pairs DNA fragment encoding the HA-tag before the STOP codon with the Gibson mutagenesis kit ( NEB ) . The full insert , now including the HA-tag , was amplified by PCR and cloned in pCFJ104 at the Bgl II site . N2 worms were injected with a mix containing 25 ng/μl of pCFJ90 ( Pmyo-2::mCherry; Addgene ) which was used as a co-injection marker , 200 ng/μl of the sgRNA vector ( pUC57 , in which the unc-119 sgRNA sequence was replaced with 5´-AAATGGAAAATTAATCCTGC-3´sequence ) , 175 ng/μl of the Peft-3Cas9-SV40 NLStbb-23´-UTR ) and 220 ng/μl of the donor vector . mCherry positive worms were individually picked and genotyped to identify insertion events . To generate the GFP::msh-5 and the GFP::brc-1 strains , the GFP was amplified with a pair of primers carrying 25 bases of homology to each side of the region flanking the ATG of the msh-5 or the brc-1 genes . The sequence 5´-TGGTTCAAATGTCCACTCGA-3´ was used as a crRNA ( Dharmacon ) for msh-5 and the 5´-AGATGGCAGATGTTGCACTG -3´ for brc-1 . The tagging strategy was the same used in [86] . The same experimental design was followed to tag the endogenous cosa-1 locus with a 5´- OLLAS-tag ( tag sequence: SGFANELGPRLMGK ) . A 200 base pairs DNA ultramer ( IDT ) carrying the OLLAS-tag immediately after the ATG was employed . The sequence 5´-AAGTGTCAATGTCAAGTTCT-3´ was used as a crRNA ( Dharmacon ) . Synthetic 200 base pairs DNA ultramers ( IDT ) were employed to generate the msh-5::2xHA and the brd-1::HA as well: the crRNAs used were 5´-CGAACGATCTATCGTCTCAT-3´ and 5´-ACGGAAAATGGTTAATGTGG-3´ respectively . To generate a full deletion of the brc-1 locus ( brc-1 ( KO ) ) , two sgRNAs were designed to target the beginning ( 5´-AGATGGCAGATGTTGCACTG-3´ ) and the end ( 5´-CGATTCGATAGGCTGCCTGC-3´ ) of the brc-1 gene . A repair template carrying the 5´-UTR directly in fusion with the STOP codon was synthesised ( IDT ) . The 9 . 713 base pairs deletion allele obtained was sequenced to assess deletion boundaries . The resulting sequence of the brc-1 locus in the KO allele is 5´-…atgaaatgttatttgtttaaaatttaatttCAGaggatTAAttttccatttcttcttcttctttctttgttc…-3´ , where “CAG” are the bases immediately preceding the ATG , and the “TAA” is the STOP-codon . All the strains generated by CRISPR were sequenced to ensure fidelity of the insertion and backcrossed to N2 worms at least twice prior usage . | Sexually reproducing species rely on meiosis to transmit their genetic information across generations . Parental chromosomes ( homologues ) undergo many distinctive processes in their complex journey from attachment to segregation . The physiological induction of DNA double strand breaks is crucial for promoting correct chromosome segregation: they are needed to activate the DNA repair machinery responsible for creating physical connections , or crossovers ( COs ) , between the homologues . In turn , crossovers promote the accurate segregation of the chromosomes in daughter cells . The BRCA1–BARD1 complex has a pivotal role during DNA repair in somatic cells and is exclusively located on unaligned chromosomal regions during mammalian meiosis . We show that in Caenorhabditis elegans , BRCA1 and BARD1 localize to chromosomes at all stages of meiotic prophase I and are enriched at presumptive crossover sites . We found that BRCA1 promotes DNA loading of the repair factor RAD-51 in specific mutant backgrounds and upon exogenous damage induction . Our data provide evidence for a direct physical association between BRCA1-BARD1 and pro-crossover factors ( including the synaptonemal complex ) and identify an important role for BRCA1 in stimulating meiotic DNA repair . Further studies are necessary to identify the substrates acted upon by the BRCA1–BARD1 complex to maintain genome stability in the gametes . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"meiosis",
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"reproductive",
"system",
"gonads",
"caenorhabditis",
"cell",
"cycle",
"and",
"cell",
"division",
"cell",
"processes",
"meiotic",
"prophase",
"animals",
"animal",
"models",
"dna",
"damage",
"caenorhabditis",
"elegans",
"model",
"organisms",
"experimental",
"organism",
"systems",
"dna",
"research",
"and",
"analysis",
"methods",
"synapsis",
"chromosome",
"biology",
"animal",
"studies",
"biochemistry",
"eukaryota",
"cell",
"biology",
"nucleic",
"acids",
"anatomy",
"genetics",
"nematoda",
"biology",
"and",
"life",
"sciences",
"dna",
"repair",
"dna",
"recombination",
"organisms",
"genital",
"anatomy"
] | 2018 | BRCA1-BARD1 associate with the synaptonemal complex and pro-crossover factors and influence RAD-51 dynamics during Caenorhabditis elegans meiosis |
Plasmodium falciparum , the causative agent of the most severe form of malaria in humans invades erythrocytes using multiple ligand-receptor interactions . The P . falciparum reticulocyte binding-like homologue proteins ( PfRh or PfRBL ) are important for entry of the invasive merozoite form of the parasite into red blood cells . We have analysed two members of this protein family , PfRh2a and PfRh2b , and show they undergo a complex series of proteolytic cleavage events before and during merozoite invasion . We show that PfRh2a undergoes a cleavage event in the transmembrane region during invasion consistent with activity of the membrane associated PfROM4 protease that would result in release of the ectodomain into the supernatant . We also show that PfRh2a and PfRh2b bind to red blood cells and have defined the erythrocyte-binding domain to a 15 kDa region at the N-terminus of each protein . Antibodies to this receptor-binding region block merozoite invasion demonstrating the important function of this domain . This region of PfRh2a and PfRh2b has potential in a combination vaccine with other erythrocyte binding ligands for induction of antibodies that would block a broad range of invasion pathways for P . falciparum into human erythrocytes .
Invasion of apicomplexan parasites into host cells is a complex process involving multiple ligands stored in apical organelles known as micronemes and rhoptries ( for review see [1] ) . The ligands are released from these compartments onto the invasive zoite form of the parasite during egress or invasion of the host cell where they are able to bind receptors . After initial contact involving low affinity interactions the parasite reorients so that the apical end is abutting the host cell membrane and a tight junction is formed with the invading parasite membrane . The tight junction involves specific parasite ligands and this structure is ultimately linked to the actomyosin motor that provides the force required for invasion ( see for review [2] ) . Entry into the host cell is mediated by movement of the tight junction across the surface to the posterior , where membrane fusion completes formation of a parasitophorous vacuole surrounding the internalised parasite . Whilst some apicomplexan parasites , such as Toxoplasma gondii , are able to invade many different host cells Plasmodium spp . merozoites have an exquisite preference for red blood cells and this is mediated by specific parasite ligand-host receptor interactions . In the case of P . falciparum , the causative agent of the most severe form of malaria in humans , this involves at least two protein families . Firstly , the erythrocyte binding-like ( EBL ) proteins have been shown to be important in merozoite invasion by inhibition with specific antibodies and also analysis of P . falciparum parasites in which the gene encoding them have been disrupted [3] , [4] , [5] , [6] , [7] . This family consists of EBA-175 ( MAL7P1 . 176 ) , EBA-181 ( JESEBL ) ( PFA0125c ) , EBL-1 ( GenBank: AAD33018 . 1 ) , EBA-165 ( PEBL ) ( PFD1155w ) and EBA-140 ( BAEBL ) ( MAL13P1 . 60 ) [4] , [8] , [9] . These proteins belong to a larger family of proteins in Plasmodium spp . that includes the Duffy binding proteins ( DBP ) in P . vivax and P . knowlesi [9] . EBA-175 and EBA-140 bind to glycophorin A and C respectively in a sialic acid-dependent manner and are responsible for specific invasion pathways through these receptors [9] , [10] , [11] , [12] . Two other ligand-receptor interactions requiring sialic acid are EBA-181 , which binds to an unknown receptor [13] and EBL-1 to glycophorin B , an interaction of lower significance since approximately 50% of P . falciparum strains analysed expressed a truncated protein [14] . EBA-165 appears to be a transcribed pseudogene as the protein has not been shown to be expressed in any P . falciparum parasites to date [15] . The second family of proteins important for invasion of merozoites is the reticulocyte binding-like ( RBP ) proteins of Plasmodium spp . that includes the Py235 family of P . yoelii and the P . vivax RBP 1 and 2 proteins [16] , [17] . These proteins have been implicated in mediating reticulocyte preference for P . yoelii and P . vivax . In P . falciparum , this family includes PfRh1 ( PFD0110w ) , PfRh2a ( PF13_0198 ) , PfRh2b ( MAL13P1 . 176 ) , PfRh3 ( PFL2520w ) , PfRh4 ( PFD1150c ) and PfRh5 ( PFD1145c ) [6] , [7] , [18] , [19] , [20] , [21] , [22] , [23] . The PfRh3 gene is a transcribed pseudogene in P . falciparum parasites whilst the other genes are differentially expressed and are localised to the neck of the rhoptries before merozoite invasion [6] , [24] . The PfRh1 , PfRh4 and PfRh5 proteins bind to specific receptors on the erythrocyte and the physical properties of these have been defined by analysis of binding and invasion into neuraminidase- , trypsin- and chymotrypsin-treated erythrocytes [7] , [18] , [19] , [20] , [22] , [25] , [26] , [27] , [28] , [29] . PfRh1 binds to a neuraminidase-sensitive receptor [18] , [19] whilst PfRh4 and PfRh5 bind different receptors in a sialic acid-independent manner ie . neuraminidase-resistant [25] , [26] , [28] . PfRh2a and PfRh2b have not been directly demonstrated to bind erythrocytes but analyses of P . falciparum gene knockout strains have shown that the latter protein functions in merozoite invasion [6] . In contrast , PfRh2a appears to be inactive in some strains of P . falciparum despite expression of the protein [6] , [30] . Recently , Complement receptor 1 ( CR1 ) has been identified as a receptor for P . falciparum invasion of erythrocytes [31] , [32] and PfRh4 binds directly to this receptor to mediate merozoite invasion [28] . P . falciparum strains are able to utilise variable patterns of alternate host receptors and this provides a mechanism of phenotypic variation to evade host immune responses and circumvent the polymorphic nature of receptors on the erythrocyte surface within the human population [6] . This is mediated by differential expression and function of both the EBL and PfRh proteins [6] , [25] , [33] , [34] . P . falciparum strains show a wide range of PfRh1 expression as a result of amplification of the gene and disruption of ligand expression by gene knockout results in increased function of other ligands from these families [7] . Both PfRh2a and PfRh2b are highly expressed in some P . falciparum lines; however , cannot be detected in others despite the presence of intact genes suggesting that they are differentially silenced [6] , [35] . It has not yet been demonstrated that these ‘silenced’ genes can be activated . In contrast , PfRh4 is also silenced in some lines and activated in a small proportion of the population and these can be selected either by disruption of the EBA-175 gene or alternatively growth of the parasites in neuraminidase-treated erythrocytes [25] , [36] . These selected parasites are more dependent on the PfRh4 ligand and CR1 receptor for merozoite invasion and this ability to switch ‘invasion pathways’ provides an additional mechanism of phenotypic variation for entry into the host cell [25] , [32] . Proteolytic cleavage of P . falciparum ligands plays an essential role in their function and fragments of the EBL and PfRh proteins are released into the supernatant during the invasion process [7] , [37] . It has been demonstrated that members of the EBL and PfRh family of proteins can be cleaved in the transmembrane region by the membrane associated protease rhomboid 4 ( PfROM4 ) ( PFE0340c ) which has implicated this protease in release of these proteins during invasion [29] , [37] , [38] . EBA-175 , PfRh1 and PfRh4 appear to be cleaved by PfROM4 as the expected products are produced during invasion [29] , [37] . PfRh1 and PfRh4 undergo a series of processing events during merozoite development resulting in a number of fragments and this appears to be critical for their function [29] . In this work , we show that PfRh2a and PfRh2b undergo a complex series of proteolytic cleavage events and that a defined region of the N-terminus directly binds erythrocytes . We show this region is critical to the function of these ligands as antibodies inhibit receptor binding and merozoite invasion .
PfRh2a and PfRh2b proteins are identical for most of the N-terminus and completely diverge in sequence 42 and 54 kDa respectively from the C-terminus ( Figure 1A ) [19] , [20] . To determine the role of proteolytic cleavage in these proteins a panel of six recombinant proteins were produced that spanned PfRh2a and PfRh2b and polyclonal and monoclonal antibodies raised ( Figure 1A ) . The six recombinant proteins were in addition to two other fusion proteins that had previously been used to generate the rabbit antibodies , R2A9 and R2A11 ( Figure 1A ) [20] . Polyclonal ( R1090 , PfRh2b-specific and R1088 , PfRh2a-specific ) and monoclonal antibodies ( 4B7 , PfRh2b specific and 8F9 , PfRh2a specific ) were made to the unique regions of PfRh2a and PfRh2b . Additionally , polyclonal ( R1070 , R1170 and R1171 ) and monoclonal antibodies ( 6F12 , 1F10 , 3A2 ) were derived to the common N-terminus of both proteins ( Figure 1A ) . The specificity of these antibodies was confirmed by immunoblot using parasite lines 3D7 and FCR3 , in which the former expresses PfRh2a and PfRh2b whilst the latter does not ( Figure 1B ) [7] . The monoclonal antibody 3A2 bound to protein bands corresponding to that expected for both PfRh2a and PfRh2b which were absent in FCR3 parasites . In contrast , the 4B7 monoclonal antibody bound to the protein bands corresponding to PfRh2b , which were absent in FCR3 . The 8F9 monoclonal antibody bound specifically to bands expected for PfRh2a expressed in 3D7 . The R1171 rabbit polyclonal bound to both major PfRh2a and b protein bands . The R1090 and R1088 antibodies were specific to PfRh2b and PfR2a respectively . Therefore the monoclonal antibody 4B7 and the polyclonal antibody R1090 are specific for PfRh2b whilst the monoclonal antibody 8F9 and polyclonal antibody R1088 reacts specifically with PfRh2a . In contrast , the monoclonal antibody 3A2 and polyclonal rabbit antibodies R1171 react with both PfRh2a and PfRh2b . Whilst the large PfRh2a and PfRh2b proteins were detected previously [6] , [7] , [18] , [19] , [20] , [21] , [22] , [23] , antibodies to the N-terminus have not been available ( Figure 1C ) . Immunoblots of 3D7 culture supernatants unexpectedly identified an 85 kDa protein using the four antibodies to different regions of this domain ( R1070 , R1170 and monoclonals 6F12 and 1F10 ) ( Figure 1A and 1C ) . The same sized 85 kDa protein was also detected in purified 3D7 schizonts , suggesting that this processing event occurred before rupture and release of merozoites for invasion . To confirm that R1070 , R1170 , 1F10 and 6F12 antibodies were detecting the same 85 kDa polypeptide from the PfRh2a and PfRh2b proteins we immunoprecipitated with each and probed immunoblots with the other antibodies ( Figure 1D ) . The results show that the four antibodies can specifically immunoprecipitate the 85 kDa processed PfRh2a/2b fragment and that this can be detected by immunoblot using the other antibodies . Therefore the full length PfRh2a ( 370 kDa ) and PfRh2b ( 382 kDa ) proteins are processed into two major peptides , during late schizont development , of approximately 85/285 kDa for PfRh2a and 85/297 kDa for PfRh2b . Previously , it has been shown that processing of PfRh2a and PfRh2b was brefeldin A ( BFA ) sensitive implying that a protease cleavage event occurred post-Golgi in the schizont [24] . To show that cleavage and release of the 85 kDa polypeptide derived from the full-length PfRh2a and PfRh2b proteins occurred after trafficking through the Golgi , we radiolabelled 3D7 schizonts in the presence or absence of BFA and immunoprecipitated with an antibody ( R1170 ) to the 85 kDa fragment ( Figure 2 A ) . BFA blocked appearance of the N-terminal 85 kDa polypeptide showing that this fragment was derived by processing from the full-length proteins , and also suggests this occurs at the schizont stage before merozoite egress ( Figure 2A ) . To confirm that processing occurs prior to merozoite egress , 3D7 parasites were radiolabelled in the presence or absence of the cysteine protease inhibitor E64 , an inhibitor of schizont rupture [39] , [40] . These experiments showed that E64 had no effect on cleavage of the PfRh2a/b proteins to the 85 kDa polypeptide consistent with this processing event occurring in schizonts prior to merozoite egress ( Figure 2A ) . To further confirm that the 85 kDa fragment was produced by processing of the full-length PfRh2a and PfRh2b proteins we used pulse chase experiments followed by immunoprecipitation with an N-terminal antibody ( Figure 2B ) . Synchronized 3D7 schizonts were radiolabelled with [35S]methionine/cysteine and chased with the same nonradioactive amino acids . Samples were taken following the commencement of cold chase and immunoprecipitated with R1170 , an antibody to the 85 kDa fragment . At the beginning of the chase there was little of the 85 kDa fragment , but this accumulated over the time points peaking at 15 minutes ( Figure 2B ) . A protein band of approximately 380 kDa , that co-migrated with the large unprocessed PfRh2a/b proteins ( Figure 2B ) , was very faintly visible that decreased in its accumulation concomitant with the increase in the 85 kDa band suggesting it is the precursor protein . However , this large molecular weight band was not heavily labelled consistent with the very low levels of the full-length PfRh2a/b proteins observed in immunoblots ( Figure 1B ) . The identity of the radiolabelled chased 85 kDa fragment was confirmed by Western blot with the 6F12 antibody ( Figure 2B ) . Taken together these data show that the full-length PfRh2a and PfRh2b proteins are proteolytically cleaved in the late schizont stage to produce an N-terminal 85 kDa fragment . To determine if the 85 kDa polypeptide cleaved from the PfRh2a and PfRh2b proteins in schizont stages co-localised with the C-terminal fragments of each protein we used synchronised parasites for immunofluorescence experiments . The PfRh2a specific monoclonal antibody 8F9 ( detects the C-terminal polypeptide ) and the R1170 rabbit antibody ( detects the N-terminal 85 kDa domain ) both showed an apical pattern with strong co-localisation ( Figure 3A ) . To determine if the 85 kDa fragment also co-localised with the 297 kDa C-terminal PfRh2b polypeptide parasites were dual stained with the monoclonal antibody 4B7 ( detects the C-terminal PfRh2b protein ) and R1170 ( detects the N-terminal 85 kDa of PfRh2a and PfRh2b ) rabbit antibody . Both antibodies showed a staining pattern of the apical end in merozoites with clear co-localisation ( Figure 3B ) . This suggested that the PfRh2a and PfRh2b cleaved polypeptides had the same subcellular co-localisation in schizont and merozoite stages . To confirm this parasites were dual stained with an antibody recognising the PfRh2a and PfRh2b C-terminal polypeptides and the 85 kDa N-terminal domain using R1170 rabbit antibodies . Both antibodies showed co-localisation in schizont and merozoite stages . ( Figure 3C ) . These results have confirmed that following cleavage of PfRh2a and PfRh2b the 85 kDa N-terminal fragment and the large C-terminal processed polypeptides co-localise in schizont and merozoite stages of P . falciparum consistent with them existing as a complex ( Figure 3C ) . To determine if the 85 kDa N-terminal region of PfRh2a and PfRh2b formed a complex with the corresponding C-terminal regions of each protein we used 3D7 schizont stage parasites in immunoprecipitation experiments with the R2A9 polyclonal antibodies ( recognises C-terminal PfRh2a and PfRh2b domain ) ( Figure 4 ) . The precipitated proteins were detected in immunoblots using the monoclonal 6F12 that recognised the N-terminal 85 kDa domain as well as the full-length PfRh2a and PfRh2b proteins ( Figure 4A ) . The same immunoprecipitate probed with the monoclonal 3A2 recognised the unprocessed and C-terminal processed forms of PfRh2a and PfRh2b . Similar results were obtained in immunoprecipitation experiments with culture supernatants ( Figure 4B ) . We did similar experiments with 3D7Δ2a and 3D7Δ2b , parasites that lack expression of PfRh2a or PfRh2b respectively , to determine if both proteins could form a complex after N-terminal processing [6] . This confirmed that PfRh2a and PfRh2b are both processed to yield the 85 kDa N-terminal fragment and that this can form a complex with either the PfRh2a or PfRh2b C-terminal domain and that these complexes are shed into the culture supernatant during merozoite invasion of red blood cells ( Figure 4C and D ) . Both the 85 kDa and 285/297 kDa C-terminal products of Rh2a/Rh2b have multiple cysteine residues and to determine if the complex formed between the N-terminal and C-terminal polypeptides are held together through disulphide bonds we used SDS-PAGE under non-reducing conditions and as a comparison also under reduced conditions ( Figure S1 ) . Proteins from culture supernatants were probed with antibodies that would detect the PfRh2a and PfRh2b 85 , 285 and 297 kDa polypeptides . These proteins migrated at an identical size compared to that observed under reducing conditions indicating the complex did not involve disulphide bonds ( Figure S1 ) . In order to follow the processing events before and after merozoite invasion we attempted to tag the 3′ end of the Pfrh2a and Pfrh2b genes with haemagglutinin ( HA ) epitopes by single crossover homologous recombination ( Figure 5A ) [41] . Unfortunately , we were unable to select parasites in which the plasmid had inserted into the Pfrh2b gene; however , this was successful for Pfrh2a by transfection of W2mef parasites to derive the parasite line W2mefRh2a-HA . Successful tagging was shown by immunoblots using anti-HA antibodies that detected two bands corresponding to PfRh2a in the transfected line but not the parental parasite ( Figure 5B , left panel ) . This was confirmed using the PfRh2a specific 8F9 monoclonal antibody that identified the protein in W2mef , and an approximately 5 kDa larger protein , corresponding to the epitope tag , in the W2mefRh2a-HA transfected line ( Figure 5B , right panel ) . Despite the small size of the tag , its very acidic characteristics ( pKi = 3 . 80 ) , may contribute to it causing a significant mobility difference on SDS-PAGE between the untagged and tagged versions of PfRh2a . The ligand EBA-175 is cleaved by the protease PfROM4 at the C-terminus in the transmembrane domain and we have demonstrated that PfRh1 and PfRh4 are similarly cleaved during merozoite invasion resulting in release of these processed proteins into the supernatant [29] , [37] . To test if PfRh2a was cleaved we purified ring stages less than 4 h post invasion to determine if the transmembrane and cytoplasmic tail of PfRh2a , potentially released by PfROM4 , could be detected using anti-HA antibodies ( Figure 5C ) . Immunoprecipitation of saponin or TNET solubilised proteins using mouse anti-HA antibodies followed by detection with rat anti-HA antibodies recognised a band of ∼17 kDa ( Figure 5C ) . While the calculated size of the protein stub following rhomboid cleavage within the transmembrane domain is 11 kDa , the highly acidic nature of this stub ( pKi = 3 . 77 ) may cause it to migrate aberrantly on SDS-PAGE . Hence this indicated that the tagged PfRh2a protein was C-terminally cleaved during merozoite invasion removing a stub migrating at ∼17 kDa , corresponding in size to the transmembrane , cytoplasmic tail and HA3 epitope tag , consistent with the function of a PfROM protease to release the ectodomain into the supernatant . In addition , a small proportion ( <5% ) of unprocessed PfRh2a , migrating at over 250 kDa , was observed indicating that proteolytic cleavage of the cytoplasmic tail was highly efficient during merozoite invasion ( Figure 5C ) . To confirm that the HA-tagged PfRh2a C-terminal stub was carried into the ring stage of the parasite , synchronous PfRh2aHA parasites containing late schizont and early ring forms were dual stained with anti-HA and PfRh2a/b common region antibodies ( Figure 5D ) . As expected the developing merozoites in schizont stages showed apical staining with both anti-HA and anti-PfRh2a/b ectodomain antibodies . Following invasion the ring stage parasites reacted with anti-HA antibodies indicating that the HA-tagged stub was carried into ring stages , while the PfRh2a/b ectodomain was not , but instead was presumably released into the culture supernatant . This was consistent with PfROM4 cleavage of PfRh2a , and by inference PfRh2b , during merozoite invasion that would result in shedding of the 285 kDa PfRh2a and 297 kDa PfRh2b ectodomains and the 85 kDa N-terminal region into the culture supernatant with the remaining cleaved stub carried into the ring stage after invasion of the red blood cell ( Figure 1B ) . This result is consistent with previous data showing that the PfRh2a and b putative recognition sequences for PfROM4 can be cleaved by this protease when expressed in mammalian cells [29] , [37] , [38] . We have previously localised PfRh2a and PfRh2b to the neck of the rhoptries in merozoites by immunoelectron microscopy [6] . However , it has not been technically possible to purify viable P . falciparum merozoites in sufficient quantities to allow the localisation of proteins during invasion of erythrocytes by immunoelectron microscopy . Recently , a new method has been developed for purification of invasive merozoites and we used this to superinfect erythrocytes to trap P . falciparum parasites during invasion for the localisation of PfRh2a and PfRh2b by immuno-electron microscopy with antibodies to the conserved common domain of both proteins . As we have described previously , PfRh2a/b was predominantly retained within the rhoptry neck before active invasion ( Figure 6C , inset ) [6] . After attachment , the results suggested movement of PfRh2a/b from within the rhoptry neck to the merozoite surface in some parasites , predominantly anterior to the tight junction within the invasion pit . In other instances , evidence for immunogold labelling at the tight junction was observed ( Figure 6C ) , supporting recent immunofluorescence data for another PfRh family member , PfRh1 [6] . For a comparison we also labelled free merozoites with EBA-175 antibodies as the PfRh and EBL protein families appear to have overlapping functions in erythrocyte invasion [[25] , [42] . The antibodies to EBA-175 displayed a micronemal localisation ( Figure 6D ) as has been shown previously [43] . From attachment through to invasion localisation was difficult to follow , but suggested that surface-released EBA-175 capped the apical tip , with labelling seen in the invasion pit of some parasites caught mid invasion ( Figure 6D ) . A proportion of EBA-175 labelling , however , remained within the apex of the merozoite , even after initiation of invasion . The overlap in function between EBL and PfRh proteins during invasion shown previously , is consistent with the images shown here suggesting that these proteins play an important role at or close to the tight junction during merozoite invasion [25] , [42] . It has been suggested previously that the large PfRh2a and PfRh2b proteins shed into culture supernatants did not bind red blood cells and additional experiments using new antibody reagents , generated in this study , was consistent with these results ( Figure S2 ) [6] . However , PfRh2b was functional in 3D7 parasites as it was possible to inhibit invasion with antibodies to this protein suggesting it interacted either directly or indirectly with a trypsin- and neuraminidase-resistant erythrocyte receptor . Identification of an 85 kDa fragment from both PfRh2a and PfRh2b raised the possibility that this contained the erythrocyte binding domain . In order to test this possibility we performed erythrocyte-binding assays using culture supernatant from 3D7 and W2mef parasites ( Figure 7A ) . Using the anti-PfRh2a/b monoclonal 6F12 we detected the 85 kDa protein from both 3D7 and W2mef after it was eluted from red blood cells . This binding was trypsin-resistant and partially chymotrypsin- and neuraminidase-sensitive which is similar but not identical to the physical properties of the host receptor determined previously by analysing PfRh2b parasites in which the gene had been disrupted as well as inhibition of function with specific antibodies for invasion into enzyme-treated erythrocytes [6] . Taken together these results show that the 85 kDa PfRh2a and PfRh2b polypeptide contains the binding site for human erythrocytes . To confirm that the 85 kDa PfRh2a and 2b protein was able to bind erythrocytes , we tested antibodies to this region ( Figure 1 ) for their ability to block binding of the 85 kDa protein to erythrocytes . Rabbit antibodies ( R1070 and R1170 ) and both the 6F12 and 1F10 monoclonals were used at concentrations from 0 . 1 to 1 . 0 ug/ul and only R1170 antibody efficiently blocked binding of native PfRh2a/b to erythrocytes ( Figure 7B ) . Normal rabbit serum , polyclonal antibodies to other regions of the 85 kDa domain ( R1070 , Figure 7B ) and the monoclonal antibody 1F10 ( Figure 7B ) and 6F12 ( data not shown ) did not block binding . This suggested that the binding region was located towards the C-terminus of the 85 kDa polypeptide ( Figure 1 ) . In order to define the receptor binding region within the 85 kDa PfRh2a and b proteins we made a recombinant hexa-His tagged protein of 15 kDa corresponding to amino acids 446 to 557 ( rRh215 ) . This protein fragment was located close to similar regions of PfRh1 and PfRh4 that have been shown to bind erythrocytes ( Figure S3 and S4 ) [26] , [27] , [28] , [32] . Rabbit antibodies ( R1170 ) made to rRh215 that block binding of native PfRh2a and b to erythrocytes have already been described above ( Figure 7B ) . The rRh215 fusion protein ( Figure S5 ) bound to erythrocytes whereas the 2b1 protein ( Figure S5 ) from the C-terminal region of PfRh2b showed no detectable binding ( Figure 8A , 8B and 8C ) . Additionally , we tested binding of a number of other recombinant proteins corresponding to different regions of the PfRh2a , PfRh2b and PfRh1 proteins and none of these regions showed binding to red blood cells ( Figure S6 ) . The specificity of binding of the rRh215 fragment was shown by heat denaturing the fusion protein and binding to erythrocytes ( Figure 8B ) . Only the native fragment bound erythrocytes . The rRh215 erythrocyte binding was resistant to trypsin treatment but partially sensitive to chymotrypsin and neuraminidase treatment , a pattern of binding we observed for the P . falciparum expressed 85 kDa protein from culture supernatants . The enzyme-treated erythrocytes were also used in binding assays with parasite culture supernatants and the eluted proteins probed with EBA-175 antibodies to ensure that the enzyme treatments had been successful . To show that binding of rRh215 to erythrocytes was specific we showed that Protein G-purified R1170 antibodies blocked binding ( Figure 8D ) . The R1170 antibodies inhibited binding of rRh215 in a dose-dependent manner with almost full blocking at an antibody concentration of 0 . 26 mg/ml . ( Figure 8D ) . Therefore the erythrocyte-binding domain of PfRh2a and b is located within the region defined by the 15 kDa rRh215 recombinant protein ( Figure 8 ) . Previously , it has been shown that antibodies to the C-terminus of the PfRh2a and 2b common region inhibit merozoite invasion and this confirmed that PfRh2b plays a direct role in this process [6] . To determine if antibodies to rRh215 ( R1170 ) inhibit invasion we tested them in growth inhibition assays with normal and trypsin-treated erythrocytes . The anti-rRh215 antibodies showed approximately 18% inhibition of 3D7 invasion into normal erythrocytes compared to no inhibition for antibodies ( R1070 ) to a second fusion protein away from the receptor binding site and this inhibition was increased for trypsin-treated cells to 38% ( Figure 9A ) . The enhancement of inhibition occurred as a result of removal of trypsin-sensitive receptors from erythrocytes thus limiting those available . The PfRh2a/b erythrocyte receptor is trypsin-resistant and removal of other receptors by this treatment increased the potency of these inhibitory antibodies [6] . To show that the inhibitory effect was specific and also to determine if it was acting on the function of both PfRh2a and PfRh2b we used the P . falciparum lines in which each gene had been specifically disrupted ( 3D7Δ2a and 3D7Δ2b ) or lacked expression of these proteins ( FCR3 ) [6] ( Figure 9B ) . For untreated erythrocytes anti-rRh215 antibodies inhibited growth of 3D7Δ2a ( which lacks expression of PfRh2a ) at about the same level as for , the 3D7 parent ( Figure 9A and B ) and this was enhanced for both parasites when using trypsin-treated erythrocytes . In contrast , the P . falciparum lines 3D7Δ2b ( lacks expression of PfRh2b ) and FCR3 ( lacks expression of PfRh2a and may express very low levels of PfRh2b ) were not inhibited ( Figure 9B ) . Therefore the anti-rRh215 antibodies to the receptor-binding site directly inhibit only PfRh2b function but not PfRh2a function , confirming that PfRh2a was not functional in 3D7 [6] , [7] , [18] , [19] , [20] , [21] , [22] , [23] .
Invasion of host cells by Plasmodium spp . requires specific ligand-receptor interactions to identify the appropriate target followed by activation of the entry process ( see for review [1] ) . Different P . falciparum strains utilise alternative host receptors for invasion of red blood cells by merozoites that provides a mechanism of phenotypic variation for evasion of host immune responses and a means to circumvent the polymorphic nature of the erythrocyte in the human population [6] . The PfRh family of proteins are key players in this process and some have been shown to bind to red blood cells and function directly in merozoite invasion [6] , [7] , [18] , [19] , [20] , [21] , [22] , [23] . PfRh2a and PfRh2b are important members and they undergo a complex series of protease cleavage events that is critical for their function in invasion . We have shown that both proteins bind directly to erythrocytes and have defined the region involved in receptor binding . Antibodies to the receptor-binding region can directly inhibit merozoite invasion by blocking interaction with the host receptor demonstrating the important function of this domain . The PfRh protein family plays an important role in merozoite invasion and proteolytic processing appears to be critical to their function [29] . In mature schizonts both PfRh2a and PfRh2b are processed into an 85 kDa fragment that exists as a complex with the C-terminal region of the corresponding protein . This processing event occurs post-Golgi as it is sensitive to BFA , an inhibitor of SEC7 , resulting in retrograde protein transport and accumulation in the endoplasmic reticulum [24] . However , the protease cleavage of PfRh2a and b must occur before release of the merozoites from the host red blood cell as it was observed in the presence of an inhibitor of parasite egress ( Figure 10 ) . PfRh1 undergoes a similar processing event in schizonts in which the 360 kDa protein is processed to 120 kDa and 240 kDa with the larger fragment containing the erythrocyte binding region . Previously , it was not possible to show that the 120 and 240 kDa PfRh1 fragments formed a complex in schizonts , in contrast to PfRh2a and PfRh2b , and it was hypothesised that the soluble 240 kDa fragment may complex with another protein [29] . Whilst PfRh2a and PfRh2b processed complexes could be detected in both schizonts and culture supernatant there appears to be a pool of free proteins similar to the situation with PfRh1 . In red cell binding experiments the 85 kDa region of PfRh2a and PfRh2b binds but there was no detectable C-terminal region suggesting either that the complex in culture supernatants may be unstable and difficult to detect or that the 85 kDa region undergoes conformational change that releases the associated C-terminal region upon binding of the receptor on the red blood cell . Our inability to detect these complexes with PfRh2a and PfRh2b antibodies other than the R2A9 polyclonal would be consistent with this possibility . PfRh2a was processed during merozoite invasion at the transmembrane region suggesting cleavage by PfROM4 protease ( Figure 10 ) . PfROM4 has been shown to cleave the transmembrane of EBA-175 , and PfRh1 and PfRh4 are also similarly cleaved resulting in the ectodomains being released into the supernatant during merozoite invasion [29] , [37] . These proteins are substrates for PfROM4 when expressed in a heterologous system suggesting they are all processed by this protease during invasion [38] . It was not possible to show that PfRh2b was processed as we were unable to tag the C-terminus; however , the evidence supporting such an event for PfRh1 , PfRh2a and PfRh4 suggests that it was also cleaved by PfROM4 resulting in shedding of the ectodomain from the merozoite surface [29] ( Figure 10 ) . This processing event at the end of invasion would allow disengagement of this protein family so that the merozoite can be completely encased within the parasitophorous vacuolar membrane in the early ring stage ( Figure 10 ) . PfRh2b has been shown previously to mediate an invasion pathway into erythrocytes using a chymotrypsin-sensitive and trypsin/neuraminidase-resistant receptor Z [6] . This was defined by invasion of P . falciparum into enzyme-treated erythrocytes and inhibition with antibodies that specifically inhibit PfRh2b function . We have not been able to demonstrate binding of either the PfRh2a or PfRh2b C-terminal processed fragment suggesting that it could function as a scaffold on which the 85 kDa N-terminus complexes and this would allow potential signalling through the transmembrane region as has been suggested by previous work [6] , [16] , [44] . The 85 kDa domain of PfRh2a and b bound to erythrocytes via a trypsin-resistant and chymotrypsin/neuraminidase-sensitive receptor . The difference in neuraminidase sensitivity of the putative receptor to that observed previously may be because we are directly observing binding in comparison to previous indirect methods [6] . It is also possible that PfRh2b functions with a co-receptor and as a result the physical properties of the receptor ( s ) results in an average of the receptor and co-receptor attributes . The use of co-receptors for entry of viruses into host cells is very common and for example CCR5 and CXCR4 are utilised for entry of HIV-1 with different isolates varying greatly in their ability to use both [45] . It is possible that P . falciparum ligands are also able to mediate invasion through co-receptors although none have as yet been identified . Previous data has shown that the PfRh2a ligand does not function in the P . falciparum line 3D7 despite it being expressed and capable of binding directly to erythrocytes [6] . This was shown by disruption of PfRh2b ( but not PfRh2a ) expression in 3D7 , that resulted in abrogation of the inhibitory effect of anti-PfRh2a/b antibodies demonstrating that only PfRh2b functioned in merozoite invasion in these parasite lines . The only parasite line in which PfRh2a has been shown to function is W2mef where it appears to act to increase efficiency of invasion into neuraminidase-treated erythrocytes [30] , [46] . Antibodies to the PfRh2a and b binding domains inhibit invasion of 3D7 only in the parental parasites expressing both PfRh2a and PfRh2b and also 3D7Δ2a that only expresses PfRh2b . This is in agreement with previous results demonstrating that PfRh2a in 3D7 parasites was not functional most likely due to the inactive cytoplasmic domain of this protein [6] . Once the PfRh proteins bind to the erythrocyte it has been hypothesised that they transmit a signal through the cytoplasmic tail to activate subsequent events in merozoite invasion and it is possible that this region of PfRh2a is incapable of performing this function thus inactivating this specific invasion pathway [6] , [16] , [44] . The receptor binding regions of PfRh1 and PfRh4 have been localised to small regions at the N-terminus of each protein [26] , [27] , [28] . Alignment of the binding-regions of PfRh1 and PfRh4 with other PfRh proteins has defined a semi-conserved domain and this overlaps directly with the 15 kDa region of PfRh2a and b that we have defined as the receptor binding domain [47] . This suggests that the binding of each PfRh protein is contained within the N-terminal semi-conserved region of each protein and that the rest of these large proteins act as a scaffold for display and access to this region . Previously a domain within the paralogous Py235 protein family has been identified as a ATP-binding domain [48] , [49] and this plays a role in altering affinity of binding of these proteins to the receptor . Whilst it has not been shown that ATP binds to PfRh2a/b if it did in a similar way to the Py235 protein the binding site would not be located in the 85 kDa receptor binding region as it is within the C-terminal processed domain . The PfRh proteins are located at the neck of the merozoite rhoptry [6] and must move to the apical surface before invasion to allow binding to their respective receptor on the surface of the erythrocyte [29] . Current evidence is consistent with an overlapping and cooperative function for the EBL and PfRh family suggesting they would be present in similar locations during merozoite invasion [25] , [42] . Interestingly , the EBL family member EBA-175 is located in micronemes [50] and presumably it and the PfRh proteins are released at a similar time onto the apical end of the merozoite . Once at the apical end of the merozoite these proteins would be free to bind directly with specific receptors on the red blood cell and they appear to play an important role not only in the apical interaction of the merozoite but also commitment to irreversible attachment that triggers downstream events of invasion [44] , [51] . Subcellular localisation of PfRh2a/b and EBA-175 during merozoite invasion of erythrocytes is consistent with an overlapping function and has suggested these proteins are located at or near the tight junction as it moves over the surface of the invading parasite . This is in agreement with previous results for the subcellular localisation of PfRh1 on invading merozoites and suggests that once these proteins bind to their receptors on the host cell they move with the tight junction during invasion and the ectodomain would be subsequently released into the culture supernatant by ROM4 cleavage . This would infer that the PfRh and EBL protein families could play an additional role ( s ) after apical interaction during active invasion of the host cell . Whilst this manuscript was under review a study was published that also identified the 85 kDa processed N-terminus of PfRh2a and PfRh2b and that this fragment bound to red blood cells in agreement with our data [52] . They suggested that the C-terminal region also showed some binding; however , we have not been able to categorically show binding of this region to red blood cells in this study and our previous work [6] . The PfRh family of proteins has been shown to be a target of invasion inhibitory antibodies by the human immune system and there is evidence that this is an important component of acquired protective immunity that would act by blocking merozoite invasion [53] . Previously , it has been shown that specific antibodies to PfRh2a and b can inhibit merozoite invasion; however , these were raised to a C-terminal region of PfRh2a and b [6] . Antibodies to the receptor-binding region of PfRh2a and b also block invasion presumably by directly interfering with binding to the erythrocyte surface . It has been shown that strains of P . falciparum can utilise different combinations of PfRh proteins for merozoite invasion and this mechanism of phenotypic variation is mediated by differential expression and the ability of some parasites to activate specific PfRh genes [6] , [25] . The ability of P . falciparum to use different invasion pathways suggests that any PfRh protein by itself would not make an effective vaccine and this is evident from the low level of growth inhibition observed for antibodies to the receptor-binding region [25] , [42] . Additionally , the proteins in the PfRh family are large suggesting that specific functional regions need to be identified to which antibodies can be raised that block merozoite invasion . Indeed , recently it has been shown that immunisation of rabbits with a combination of domains from EBA-175 , PfRh2a/b and PfRh4 raised antibodies that were very potent inhibiting merozoite invasion up to 90% [25] , [42] . Identification of the receptor-binding site of PfRh2a and b that has been defined within a 15 kDa domain of these large proteins suggests this region would be an excellent candidate for inclusion in such a combination vaccine .
Antibodies were raised in mice and rabbits under the guidelines of the National Health and Medical Research Committee and the PHS Policy on Humane Care and Use of Laboratory Animals . The specific details of our protocol were approved by the Royal Melbourne Hospital Animal Welfare Committee . P . falciparum parasites were maintained in human O+ erythrocytes . 3D7 is a cloned line derived from NF54 obtained from David Walliker at Edinburgh University . W2mef is a cloned line derived from the Indochina III/CDC strain and FCR3 is a cloned line . 3D7 parasites containing either a disrupted Pfrh2a gene ( 3D7Δ2a ) or a disrupted Pfrh2b gene ( 3D7Δ2b ) have been described previously [7] . Parasite cultures were synchronised with 5% sorbitol [54] . The pHA3 plasmid enabled tagging of the PfRh2a gene with three HA epitopes ( pHA3 ) and has been described previously [29] . Genomic DNA from P . falciparum was used as template to amplify a ∼920 bp fragment from the C-terminal end of the PfRh2a or PfRh2b genes by polymerase chain reaction ( PCR ) using the oligonucleotide pairs [5′AGCTagatctAAGACAAGAACAAGAACGACT]/[5′AGCTctgcagCAATTGTACTATCATTATCATTAAAAG] for Pfrh2a and [5′AGCTagatctGTATGATCATGTTGTTTCAGA]/[5′AGCTctgcagCAAAATATTTTTCTTCATTTTCATC] for PfRh2b [55] . The Bgl II and Pst I sites were used to clone the PCR fragments into the pHA3 plasmid . P . falciparum was transfected with 80 µg of purified plasmid DNA ( Qiagen ) and selection for stable transfectants by single recombination crossover was carried out as described [56] . To generate recombinant proteins , DNA fragments were either amplified from a codon-optimised ( GeneArt , Germany ) synthetic gene for N-terminal proteins ( Figure S7 ) or from 3D7 genomic DNA by PCR and cloned into pET45 . Bacterial cultures expressing the corresponding recombinant proteins were grown at 37°C to A600 = 0 . 5 then induced with IPTG at 30°C for 5 h . Cell pellets were lysed in PBS/0 . 1% TX100/0 . 5 mg/ml lysozyme/containing EDTA-free COMPLETE protease inhibitors ( Roche ) . DNAse I was added ( 10 µg/ml ) before sonication and centrifuged at 8000 g for 30 min at 4°C . For insoluble recombinant proteins the inclusion bodies were resuspended in PBS/0 . 1% TX100/0 . 5 mg/ml lysozyme , sonicated and washed with PBS/0 . 1% TX100 before solubilisation and further purification as below . Recombinant proteins that were soluble were further purified as outlined below . The recombinant protein used to raise the antibody R1070 , was generated using oligonucleotide pairs [5′AGCTggatccCAGCAACAGCGTGCTGGAT]/[5′AGCTctcgagTTAACGGTTCAGATACAGATC] and amplified from a synthetic gene ( Figure S7 ) . The insoluble recombinant protein as inclusion bodies was solubilised in 6 M guanidine/100 mM phosphate pH 8/250 mM NaCl , purified by Ni-NTA chromatography ( Qiagen ) and eluted with 6 M guanidine/100 mM phosphate pH 4 . 5/250 mM NaCl . The recombinant protein was refolded by diluting 100-fold in 2 M urea/20 mM Tris pH 8/20 mM NaCl/1 mM reduced glutathione/1 mM oxidised glutathione before further purification by ion-exchange on a HiTRAP Q ( GE Healthcare ) column run in 2 M urea/20 mM Tris pH 8 . Elution was effected with a linear salt gradient from 20 mM to 1 M NaCl . The purified protein was dialysed against 0 . 5 M Urea/20 mM Tris pH 8/20 mM NaCl before injection into animals . The rRh215 recombinant protein ( used to raise antibodies R1170 and 1F10 ) was generated using oligonucleotide pairs [5′AGCTggatccCAAAAAAAAATACGAAACCTATG]/[5′AGCTctcgagTTAATCGGTTTTTTCGATGTAGTTG] whilst the recombinant protein used to raise antibody R1088 and monoclonal antibody 8F9 was derived using oligonucleotide pairs [5′AGCTggatccCCACATAAAAAGTAAACTAGAATC]/[5′AGCTctcgagTTATGATCGAGAAAAATTTCTATC] . The soluble recombinant proteins were purified over Ni-NTA resin by washing with a buffer containing 50 mM phosphate pH 8/300 mM NaCl/10 mM imidazole and elution with 50 mM phosphate pH 8/300 mM NaCl/300 mM imidazole . Further purification was achieved by size exclusion chromatography on a Superdex 75 column ( GE Healthcare ) . The recombinant protein used to raise monoclonal antibody 6F12 was obtained by using the oligonucleotide pair [5′AGCTggatccCGAAAGCTATGTGATGAAC]/[5′AGCTctcgagTTAGCTGGTGTTCAGAATGG] and amplified from a synthetic gene ( Figure S7 ) . Washed inclusion bodies were solubilised in 6 M guanidine/100 mM phosphate pH8/250 mM NaCl , purified over Ni-NTA resin before elution of fusion protein with 6 M guanidine/100 mM phosphate pH 4 . 5/250 mM NaCl . The fusion protein was refolded by dilution 100-fold in 2 M urea/20 mM Tris pH 8/20 mM NaCl . The partially purified protein was concentrated by TCA precipitation before running in a SDS-PAGE gel and elution of the protein from the acrylamide for injection into animals . The recombinant protein used to raise antibody R1171 was generated using the oligonucleotide pair [5′AGCTggatccCTATGTAGATGTGGACGTTTCC]/[5′AGCTctcgagTTAAATGTCCTTATTTTTTTCATCC] . Washed inclusion bodies were solubilised in 6 M guanidine/100 mM phosphate pH 8/250 mM NaCl , purified over Ni-NTA resin before elution of fusion protein with 6 M guanidine/100 mM phosphate pH 4 . 5/250 mM NaCl . The fusion protein was refolded by dilution 100-fold in 25 mM Tris pH7 . 4/20 mM NaCl , before further purification by ion-exchange on a HiTRAP Q column run in 20 mM Tris pH 7 . 4/20 mM NaCl . Elution was effected with a linear salt gradient from 20 mM to 1 M NaCl . The purified fusion protein was dialysed against 20 mM Tris pH 7 . 4/20 mM NaCl before injection into animals . The 2b1 recombinant protein ( used to raise antibody R1090 and monoclonal 4B7 ) was generated using oligonucleotide pairs [5′AGCTggatccCAAGATAGATGAAAGTATAACTAC]/5′AGCTctcgagTTAGTTTGAATACCTTTCATTATTG] . The soluble recombinant protein was purified over Ni-NTA resin by washing with a buffer containing 50 mM phosphate pH8/300 mM NaCl/10 mM imidazole before elution with 50 mM phosphate pH 8/300 mM NaCl/300 mM imidazole . Further purification was achieved by size exclusion chromatography on a Superdex 75 column ( GE Healthcare ) . Final purification was achieved by ion-exchange on a HiTRAP Q column run in 20 mM Tris pH 8 . Elution was effected with a linear salt gradient . Purified fusion proteins were used to immunise rabbits and mice . Rabbit immunoglobulins were purified on Protein G-Sepharose and dialysed against PBS . Sera from mice immunised with fusion proteins were screened by ELISA against the fusion protein and by Western blot against late-stage parasite proteins . Splenic fusions from positive mice resulted in monoclonal antibodies ( WEHI Monoclonal Antibody facility ) . The R2A9 and R2A11 antibodies have been described previously [20] . The monoclonal antibody 3A2 was raised using the GST fusion protein 2A9 previously described and used to generate R2A9 antibodies . The anti-His antibodies used were a pentaHis mouse monoclonal ( Qiagen ) and the anti-HA antibodies used were either the mouse monoclonal 12CA5 or the rat monoclonal 3F10 ( Roche Applied Science ) . Culture supernatants enriched in proteins released during merozoite invasion were obtained by synchronisation of parasite cultures followed by treatment with trypsin and neuraminidase in order to prevent reinvasion of erythrocytes following schizont rupture , as described [7] . In some cases , culture supernatants were made by allowing schizont rupture of synchronised late stage cultures at high parasitaemia , without trypsin and neuraminidase treatment . Total proteins from schizont stage parasites were obtained by synchronisation and further cultured until mature schizonts were present . Parasite proteins were then obtained by saponin lysis of erythrocytes . All parasite preparations were made in the presence of COMPLETE protease inhibitors ( Roche ) . Proteins were separated on either 3–8% Tris Acetate ( for large proteins ) or 4–12% Bis Tris with MES buffer ( for small proteins ) SDS-PAGE gels ( Invitrogen ) . Western blotting onto nitrocellulose ( 0 . 45 µm , Schleicher and Schuell ) was performed according to standard protocols and blots were processed with a chemiluminescence system ( ECL , Amersham ) . Mature 3D7 schizonts from highly synchronous cultures were metabolically radiolabelled with 200 uCi/ml [35S]methionine/cysteine as described previously [57] . To analyse the effect of brefeldin A ( BFA , Sigma ) on PfRh2a/b processing , BFA at 5 ug/ml or an equal volume of methanol was added to cultures for 1 h . Cultures were then spun down and resuspended in one fifth volume of methionine/cysteine-free medium , containing methanol or BFA as before and 200 µCi/ml [35S]methionine/cysteine for 1 h at 37°C . Labelled parasites were washed twice with PBS , uninfected erythrocytes lysed with 0 . 1% saponin and labelled schizonts re-washed with PBS . To analyse the effect of E64 ( trans-epoxysuccinyl-L-leucylamido- ( 4-guanidino ) butane , Sigma ) on PfRh2a/b processing , E64 at 10 µM or an equal volume of water was added to cultures under identical conditions to those described for BFA treatment . Mature 3D7 schizonts from highly synchronous cultures were labelled for 10 min then chased as previously described [57] . Chase samples were taken at t = 0 , 5 , 15 , 30 and 60 min and frozen directly at −70°C . Radiolabelled saponin-lysed parasites from both BFA and E64 experiments were lysed in 10 volumes TNET ( 1% TX100 , 150 mM NaCl , 10 mM EDTA , 50 mM Tris pH7 . 4 ) containing COMPLETE ( Roche ) protease inhibitors . Lysates were sonicated and clarified by centrifugation . Radiolabelled pulse-chase samples were instead solubilised in five volumes of Denaturing buffer ( 1% SDS , 50 mM Tris pH8 . 0 , 5 mM EDTA , COMPLETE protease inhibitors ) . Samples were boiled , clarified by centrifugation , then the clarified supernatants diluted 10-fold in TNET + COMPLETE inhibitors . All immunoprecipitations of parasite proteins were carried out in a 2-step procedure by first adding antibodies , then capturing antigen:antibody complexes with ProteinG-Sepharose . Parasite proteins from schizont stages were obtained by first saponin-lysing synchronised cultures followed by schizont solubilisation with TNET . Immunoprecipitations from culture supernatants directly or from cold or radiolabelled TNET lysates were all carried out by the same 2-step procedure . Washed immunoprecipitates were separated by SDS-PAGE , blotted to nitrocellulose , then probed with antibodies of a different species to the immunoprecipitating antibodies , to avoid reactivity with immunoglobulin heavy and light chains . For experiments in Figure 5C , young Rings less than 4 h old were obtained by sorbitol synchronising a late schizont/Ring stage culture to remove schizonts . The Rings were then lysed either with 0 . 1% saponin or TNET , spun down , and both supernatant and pellet fractions collected . HA-containing proteins from both supernatant fractions were immunoprecipitated with anti-HA Abs . Immunoprecipitated proteins together with proteins from the saponin and TNET pellets were separated by SDS-PAGE , blotted , then probed with HRP-labelled anti-HA Abs . Light microscopy was performed with synchronised parasites at schizont , merozoite and ring stages . Ring stages were fixed in solution with 4% paraformaldehyde and 0 . 0075% glutaraldehyde ( ProSciTech ) as described [58] . Merozoites were obtained by air-drying smears of late-stage parasites , then fixing either in 100% methanol at −20°C . Schizonts were either fixed with paraformaldehyde/glutaraldehyde or in methanol . For dual-colour fluorescence , fixed parasites in solution or on slides were blocked in 3% BSA in PBS , before both primary antibodies were added . Parasites were washed , secondary Alexa Fluor 488/594 Abs added , then mounted with VectaShield ( Vector Laboratories ) containing 1 µg/ml DAPI . Images were captured with a Zeiss LSM5 Live microscope with an AxioCam camera and Axiovision 4 . 7 software . For paraformaldehyde/glutaraldehyde-fixed parasites , single z-stacks shown were processed with AxioVision 4 . 7 deconvolution software . Images from methanol-fixed parasites were not deconvolved . For immunoelectron microscopy , free or invading merozoites were fixed in 1% glutaraldehyde ( ProSciTech , Australia ) on ice for 30 min . Samples were pelleted in low-melt agarose before being transferred into water , dehydrated in ethanol and embedded in LR Gold Resin ( ProSciTech , Australia ) . Following polymerization by benzoyl peroxide ( SPI-Chem , USA ) samples were sectioned on a Leica Ultracut R ultramicrotome ( Wetzlar ) and then prepared for imaging with 2% aqueous uranyl-acetate followed by 5% triple lead and observed at 120 kV on a Philips CM120 BioTWIN Transmission Electron Microscope . PfRh2 proteins were stained with the 2A9 antibody ( Figure 1 ) while EBA-175 was stained with an antibody to the C-terminal cysteine-rich region . Erythrocyte binding assays using enriched culture supernatant from parasites were performed as described previously [20] . For binding assays to enzyme treated erythrocytes , erythrocytes were incubated with neuraminidase ( 66 . 7 mU/ml ) , high trypsin ( 1 . 0 mg/ml ) , Low trypsin ( 0 . 067 mg/ml ) and chymotrypsin ( 1 . 0 mg/ml ) for 1 hr at 37°C , then washed prior to the addition of soybean trypsin inhibitor at 0 . 5 mg/ml for 10 min at 37°C . Treated erythrocytes were subsequently washed three times and added to the binding assay as described above . Bound proteins eluted with 1 . 5 M NaCl were separated on SDS-PAGE gels , Western blotted , then probed with anti-PfRh2 or anti-EBA-175 antibodies . Growth inhibition assays were performed with modifications of a described method [59] . Briefly , twice-synchronised trophozoite stage parasites were added to either untreated target erythrocytes ( at a final parasitemia of 0 . 5% ) or to trypsin-treated erythrocytes ( at a parasitemia of 1% ) and haematocrit of 2% in 45 µl of culture medium in 96 well round bottom microtiter plates ( Becton Dickinson , U . S . A . ) . 5 µl of purified antibody was added to a final concentration of 2 mg/ml . After incubation with antibodies for 1 cycle of parasite growth ( ∼48 h ) , parasites were stained with ethidium bromide ( 10 µg/ml final ) , and the parasitaemia of each well determined by flow cytometry using a FACSCalibur with a plate reader ( Becton Dickinson , U . S . A . ) . At least two independent assays in triplicate were performed . Data was analysed using FlowJo software ( Tree Star Inc ) . Growth was expressed as a percentage of the parasitaemia in the presence of the Protein-G purified antibodies of the prebleed serum of the same rabbit also at a final concentration of 2 mg/ml . Trypsin treatment of erythrocytes was carried out as described previously [7] . eba-175 ( MAL7P1 . 176 ) ; eba-181 ( JESEBL ) ( PFA0125c ) ; ebl-1 ( GenBank: AAD33018 . 1 ) ; eba-165 ( PEBL ) ( PFD1155w ) ; eba-140 ( BAEBL ) ( MAL13P1 . 60 ) ; Pfrh1 ( PFD0110w ) ; Pfrh2a ( PF13_0198 ) ; Pfrh2b ( MAL13P1 . 176 ) ; Pfrh3 ( PFL2520w ) ; Pfrh4 ( PFD1150c ) ; Pfrh5 ( PFD1145c ) ; Pfrom4 ( PFE0340c ) . | The causative agent of the most severe form of malaria in humans is the protozoan parasite Plasmodium falciparum . These parasites are carried by a mosquito that infects humans during feeding resulting in injection of sporozoite forms that infect and develop in the liver into the merozoite stage . The merozoites are released into the blood stream where they invade erythrocytes in which they can grow and divide . Invasion of the red blood cell by P . falciparum merozoites involves a cascade of protein-protein interactions . The P . falciparum reticulocyte binding-like homologue proteins ( PfRh or PfRBL ) are an important protein family involved in binding to specific receptors on the red blood cell . We have analysed two members of this protein family , PfRh2a and PfRh2b , and show that they undergo a complex series of cleavage events before and during merozoite invasion . We have defined the region of these ligands that bind red blood cells and show that antibodies to this receptor-binding region block merozoite invasion demonstrating the important function of this domain . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"biology"
] | 2011 | Plasmodium falciparum Merozoite Invasion Is Inhibited by Antibodies that Target the PfRh2a and b Binding Domains |
Experiments that study neural encoding of stimuli at the level of individual neurons typically choose a small set of features present in the world—contrast and luminance for vision , pitch and intensity for sound—and assemble a stimulus set that systematically varies along these dimensions . Subsequent analysis of neural responses to these stimuli typically focuses on regression models , with experimenter-controlled features as predictors and spike counts or firing rates as responses . Unfortunately , this approach requires knowledge in advance about the relevant features coded by a given population of neurons . For domains as complex as social interaction or natural movement , however , the relevant feature space is poorly understood , and an arbitrary a priori choice of features may give rise to confirmation bias . Here , we present a Bayesian model for exploratory data analysis that is capable of automatically identifying the features present in unstructured stimuli based solely on neuronal responses . Our approach is unique within the class of latent state space models of neural activity in that it assumes that firing rates of neurons are sensitive to multiple discrete time-varying features tied to the stimulus , each of which has Markov ( or semi-Markov ) dynamics . That is , we are modeling neural activity as driven by multiple simultaneous stimulus features rather than intrinsic neural dynamics . We derive a fast variational Bayesian inference algorithm and show that it correctly recovers hidden features in synthetic data , as well as ground-truth stimulus features in a prototypical neural dataset . To demonstrate the utility of the algorithm , we also apply it to cluster neural responses and demonstrate successful recovery of features corresponding to monkeys and faces in the image set .
The question of how the brain encodes information from the natural world forms one of the primary areas of study within neuroscience . For many sensory systems , particularly vision and audition , the discovery that single neurons modulate their firing of action potentials in response to particular stimulus features has proven foundational for theories of sensory function . Indeed , neuronal responses to contrast , edges , and motion direction appear to form fundamental primitives on which higher-level visual abstractions are built . Nevertheless , many of these higher-level abstractions do not exist in a stimulus space with obvious axes . As a result , experimenters must choose a priori features of interest in constructing their stimulus sets , with the result that cells may appear weakly tuned due to misalignment of stimulus and neural axes . For example , in vision , methods like reverse correlation have proven successful in elucidating response properties of some cell types , but such techniques rely on a well-behaved stimulus space and a highly constrained encoding model in order to achieve sufficient statistical power to perform inference [1–3] . However , natural stimuli are known to violate both criteria , generating patterns of neural activity that differ markedly from those observed in controlled experiments with limited stimulus complexity [3–5] . Information-based approaches have gone some way in addressing this challenge [4] , but this approach assumes a metric structure on stimuli in order to perform optimization , and was recently shown to be strongly related to standard Poisson regression models [6] . More recently , Gallant and collaborators have tackled this problem in the context of fMRI , demonstrating that information present in the blood oxygen level-dependent ( BOLD ) signal is sufficient to classify and map the representation of natural movie stimuli across the brain [7–9] . These studies have used a number of modeling frameworks , from Latent Dirichlet Allocation for categorizing scene contents [9] to regularized linear regression [8] to sparse nonparametric models [7] in characterizing brain encoding of stimuli , but in each case , models were built on pre-labeled training data . Clearly , a method that could infer stimulus structure directly from neural data themselves could extend such work to less easily characterized stimulus sets like those depicting social interactions . A rich body of previous work has addressed the problem of identifying low-dimensional latent dynamics underlying neural firing . Typically , these models assume a continuous latent state governed by a linear dynamical system [10–17] . Using generalized linear models and latent linear dynamical systems as building blocks , these models have proven able to infer ( functional ) connectivity [10] , estimate spike times from a calcium images [11] , and identify subgroups of neurons that share response dynamics [13 , 16 , 17] . Inference in these models is generally performed via expectation maximization , though [14–19] also used a variational Bayesian approach . In each case , the focus has typically been on inferring the dynamics of intrinsic neural activity , perhaps conditioned on known covariates xt . Our work is distinct , however , in focusing on inferring features within stimuli that drive repeatable patterns of firing across time and trials . Our model sits at the intersection of these regression and latent variable approaches . We utilize a Poisson observation model that shares many of the same features as the commonly used generalized linear models for Poisson regression . We also assume that the latent features modulating neural activity are time-varying and Markov . However , we make 3 additional unique assumptions: First , we assume that the activity of each neuron is modulated by a combination of multiple independent latent features governed by Markov dynamics . ( This can be extended to the semi-Markov case; see Supplementary Information ) . This allows for latents to evolve over multiple timescales with non-trivial duration distributions , much like the hand-labeled features in social interaction data sets . Second , we assume that these latents are tied to stimulus presentation . That is , when identical stimuli are presented , the same latents are also present . This allows us to selectively model the dynamics of latent features of the stimulus that drive neural activity , rather than intrinsic neural dynamics ( e . g . , variation within and across trials ) . Finally , we enforce a sparse hierarchical prior on modulation strength that effectively limits the number of latent features to which the population of neurons is selective . This allows for a parsimonious explanation of the firing rates of single units in terms of a small set of stimulus features . Finally , we perform full variational Bayesian inference on all model parameters and take advantage of conditional conjugacy to generate coordinate ascent update rules , nearly all of which are explicit . Combined with forward-backward inference for latent states , our algorithm is exceptionally fast , automatically implements Occam’s razor , and facilitates proper model comparisons using the variational lower bound . However , as noted above , we are not the first to employ variational Bayesian methods to the problem of inferring latent firing rate states . Moreover , several other models have made use of the idea of discrete latent states and Markov models as explanations of neural dynamics [19 , 20] . Both of those methods used a Hidden Markov Model ( HMM ) to capture variability in neural firing in time and identify discrete modes or states of spiking that could be driven by both spike history and external covariates . In [19] , this state space was assumed to be organized according to a binary tree , dramatically reducing model complexity . Our model differs from both of these in assuming that the states that govern firing are deterministic functions of stimuli , and that these states are a collection of discrete , independent stimulus features , not a single HMM . Thus , while previous models serve well to capture transitions between discrete states of neural activity , our model discovers statistically reliable patterns of activity that are consistent across repeated presentations of a given stimulus . By directly associating latent factors that drive firing with stimulus features , we thus achieve a means of ( multiply ) coding a given stimulus . That is , we focus on binary latent states as a means of labeling a finite number of overlapping stimulus features . Most importantly , as we will show , the stimulus features found by our model are often interpretable . The choice to assign multiple discrete , independent tags to each stimulus results in a combinatorial code , with capacity exponential in the number of tags . This can , in principle , accommodate a hierarchical structure ( as in [19] ) , but need not . Yet the ultimate goal of latent state models such as ours is to provide a low-dimensional description of neural responses , not simply a compression of them . In practice , experimentalists may perform an initial screening experiment by exposing an organism to a broad range of stimuli , with few fixed hypotheses about responsiveness . A given population of neurons may respond to only a few stimulus features , and features so inferred do not necessarily generalize to new brain structures , nor to stimuli outside the initial set . The value of our model , as with topic models and other latent space models , comes in identifying stimulus features that are readily interpretable: we expect our method will be most useful when the latent tags it identifies group stimuli into useful categories that generate hypotheses for future experiments . In the sections below , we outline the mathematics behind our model , discuss the process of approximate Bayesian inference we use to infer stimulus features , and perform a series of validation experiments on both synthetic data and real data sets of spiking responses . In the latter , we have chosen datasets where the features that drive spiking are reasonably well understood . We train our model without using this information and then compare the inferred and experimenter-labeled features as a means of illuminating strengths and weaknesses of our model . We conclude by discussing possible extensions and applications to other domains .
Consider a population of U spiking neurons or units exposed to a series of stimuli indexed by a discrete time index t ∈ {1 … T} . We assume that this time index is unique across all stimuli , such that a particular t represents a unique moment in a particular stimulus . In order to model repeated presentations of the same stimulus to the same neuron , we further assume that each neuron is exposed to a stimulus Mtu times , though we do not assume any relationship among Mtu . That is , we need not assume either that all neurons see each stimulus the same number of times , nor that each stimulus is seen by all neurons . It is thus typical , but not required , that Mtu be sparse , containing many 0s , as shown in Fig 1 . Each unique observation m in our data set consists of a spike count Nm for a particular ( time , unit ) pair ( t ( m ) , u ( m ) ) . We model these spike counts as arising from a Poisson distribution with rate Λtu and observation-specific multiplicative overdispersion θm: N m∼ Pois ( Λ t ( m ) , u ( m ) θ m ) where θ m∼ Gamma ( s u ( m ) , s u ( m ) ) ( 1 ) That is , for a given stimulus presentation , the spiking response is governed by the firing rate Λ ( we set Δt = 1 for convenience ) , specific to the stimulus and unit , along with a moment-by-moment noise in the unit’s gain , θm . We restrict these θm to follow a Gamma distribution with the same shape and rate parameters , since this results in an expected noise gain of 1 . In practice , we model this noise as independent across observations , though it is possible to weaken this assumption , allowing for θm to be autocorrelated in time ( see Supplementary Information ) . Note that both the unit and time are functions of the observation index m , and that the distribution of the overdispersion for each observation may be specific to the unit observed . At each stimulus time t , we assume the existence of K binary latent states ztk and R observed covariates xtr . The binary latent states can be thought of as time-varying “tags” of each stimulus—for example , content labels for movie frames—and are modeled as Markov chains with initial state probabilities πk and transition matrices Ak . The observed covariates , by contrast , are known to the experimenter and may include contrast , motion energy , or any other a priori variable of interest . We further assume that each unit’s firing rate at a particular point in time can be modeled as arising from the product of three effects: ( 1 ) a baseline firing rate specific to each unit ( λ0 ) , ( 2 ) a product of responses to each latent state ( λz ) , and ( 3 ) a product of responses to each observed covariate ( λx ) : Λ t u = λ 0 u ∏ k = 1 K ( λ z u k ) z t k ∏ r = 1 R ( λ x u r ) x t r ( 2 ) Note that this is conceptually similar to the generalized linear model for firing rates ( in which we model log Λ ) with the identification β = log λ . However , by modeling the firing rate as a product and placing Gamma priors on the individual effects , we will be able to take advantage of closed-form variational updates resulting from conjugacy that avoid explicit optimization ( see below ) . Note also , that because we assume the ztk are binary , the second term in the product above simply represents the cumulative product of the gain effects for those features present in the stimulus at a given moment in time . In addition , to enforce parsimony in our feature inference , we place sparse hierarchical priors with hyperparameters γ = ( c , d ) on the λz terms: λ z u k∼ Gamma ( c z k , c z k d z k ) c z k∼ Gamma ( a c k , b c k ) d z k∼ Gamma ( a d k , b d k ) ( 3 ) That is , the population distribution for the responses to latent features is a gamma distribution , with parameters that are themselves gamma-distributed random variables . As a result , E [ λ u ] = d - 1 and var[λu] = ( cd2 ) −1 , so in the special case of c large and d ∼ O ( 1 ) , the prior for firing rate response to each latent feature will be strongly concentrated around gain 1 ( no effect ) . As we show below , this particular choice results in a model that only infers features for which the data present strong evidence , controlling for spurious feature detection . In addition , this particular choice of priors leads to closed-form updates in our variational approximation . For the baseline terms , λ0u , we use a non-sparse version of the same model; for the covariate responses , λxu , we model the unit effects non-hierarchically , using independent Gamma priors for each unit . Putting all this together , we then arrive at the full generative model: p ( N , Λ , θ ) = p ( N | Λ , θ ) p ( Λ | λ , z ) p ( λ | γ ) p ( γ ) p ( z | A , π ) p ( A ) p ( π ) p ( θ | s ) p ( s ) ( 4 ) where p ( λ | γ ) = ∏ u p ( λ 0 u | c 0 , d 0 ) ∏ k r p ( λ z u k | c z k , d z k ) p ( λ x u r ) ( 5 ) and p ( γ ) = p ( c 0 ) p ( d 0 ) ∏ k p ( c z k ) p ( d z k ) ( 6 ) in conjunction with the definitions of p ( N|Λ , θ ) and Λ ( λ , z , x ) in Eqs ( 1 ) and ( 2 ) . The generative model for spike counts is illustrated in Fig 2 . Given a sequence of stimulus presentations ( t ( m ) , u ( m ) ) and observed spike counts Nm , we want to infer both the model parameters Θ = ( λ0 , λz , λx , A , π , c0 , d0 , cz , dz , s ) and latent variables Z = ( zkt , θm ) . That is , we wish to calculate the joint posterior density: p ( Θ , Z | N ) ∝ p ( N | Z , Θ ) p ( Z ) p ( Θ ) ( 7 ) In general , calculating the normalization constant for this posterior is computationally intractable . Instead , we will use a variational approach , approximating p ( Θ , Z|N ) by a variational posterior q ( Z , Θ ) = qZ ( Z ) qΘ ( Θ ) that factorizes over parameters and latents but is nonetheless close to p as measured by the Kullback-Leibler divergence [21 , 22] . Equivalently , we wish to maximize the variational objective L ≡ E q [ log p ( Θ , Z | N ) q ( Θ , Z ) ] = E q [ log p ( Θ , Z | N ) ] + H [ q Θ ( Θ ) ] + H [ q Z ( Z ) ] ( 8 ) with H the entropy . We adopt the factorial HMM trick of [23] , making the reasonable assumption that the posterior factorizes over each latent time series z⋅k and the overdispersion factor θm , as well as the rate parameters λ⋅u associated with each Markov process . This factorization results in a variational posterior of the form: q ( Θ , Z ) = q ( c 0 ) q ( d 0 ) ∏ m q ( θ m ) ∏ u q ( s u ) q ( λ 0 u ) ∏ r q ( λ x u r ) × ∏ k q ( c k ) q ( d k ) q ( λ z u k ) q ( c z k ) q ( d z k ) q ( z k ) q ( π k ) q ( A k ) ( 9 ) With this ansatz , the variational objective decomposes in a natural way , and choices are available for nearly all of the qs that lead to closed-form updates . From Eqs ( 1 ) and ( 2 ) above , we can write the probability of the observed data N as p ( N , z|x , Θ ) =∏m[ ( θmΛt ( m ) u ( m ) ) Nme−θmΛt ( m ) u ( m ) Nm ! ]∏mk ( Ak ) zt ( m ) +1 , k , zt ( m ) , k∏k ( πk ) z0k ( 10 ) where again , m indexes observations of ( t ( m ) , u ( m ) ) pairs , the portion in brackets is the Poisson likelihood for each bin count and the last two nontrivial terms represent the probability of the Markov sequence given by ztk . From this , we can expand the log likelihood: log p ( N , z | x , Θ ) = ∑ m k r [ N m ( log θ m + log λ 0 u ( m ) + z t ( m ) k log λ z u ( m ) k + x t ( m ) r log λ x u ( m ) r ) ] - ∑ m θ m Λ t ( m ) u ( m ) + ∑ m k log ( A k ) z t ( m ) + 1 , k , z t ( m ) , k + ∑ k log ( π k ) z 0 k + constant , ( 11 ) Given that Eq ( 11 ) is of an exponential family form for θ and λ when conditioned on all other variables , free-form variational arguments [21] suggest variational posteriors: λ 0 u∼ Gamma ( α 0 u , β 0 u ) ( 12 ) λ z u k∼ Gamma ( α z u k , β z u k ) ( 13 ) λ x u r∼ Gamma ( α x u r , β x u r ) ( 14 ) For the first of these two , updates in terms of sufficient statistics involving expectations of γ = ( c , d ) are straightforward ( see Supplementary Information ) . However , this relies on the fact that zt ∈ {0 , 1} . The observed covariates xt follow no such restriction , which results in a transcendental equation for the βx updates . In our implementation of the model , we solve this using an explicit BFGS optimization on each iteration . Moreover , we place non-hierarchical Gamma priors on these effects: λxur ∼ Gamma ( axur , bxur ) . As stated above , for the latent states and baselines , we assume hierarchical priors . This allows us to model each neuron’s firing rate response to a particular stimulus as being drawn from a population response to that same stimulus . We also assume that the moment-to-moment noise in firing rates , θm , follows a neuron-specific distribution . As a result of the form of this hierarchy given in Eq ( 3 ) , the first piece in Eq ( 8 ) contains multiple terms of the form E q [ ∑ u log p ( λ u | c , d ) ] = ∑ u E q [ ( c - 1 ) log λ u - c d λ u + c log c d - log Γ ( c ) ] ( 15 ) In order to calculate the expectation , we make use of the following inequality [24] 2 π ≤ z ! z z + 1 2 e - z ≤ e ( 16 ) to lower bound the negative gamma function and approximate the above as log p ( λ ) ≥ ∑ u [ ( c - 1 ) ( log λ u + 1 ) - c d λ u + c log d + 1 2 log c ] ( 17 ) Clearly , the conditional probabilities for c and d are gamma in form , so that if we use priors c ∼ Gamma ( ac , bc ) and d ∼ Gamma ( ad , bd ) the posteriors have the form c∼ Gamma ( a c + U 2 , b c + ∑ u E q [ d λ u - log λ u - log d - 1 ] ) ( 18 ) d∼ Gamma ( a d + U E q [ c ] , b d + ∑ u E q [ c λ u ] ) ( 19 ) This basic form , with appropriate indices added , gives the update rules for the hyperparameter posteriors for λ0 and λz . For θ , we simply set c = su and d = 1 . For each latent variable z , the Markov Chain parameters πk and Ak , together with the observation model Eq ( 11 ) determine a Hidden Markov Model , for which inference can be performed efficiently via conjugate updates and the well-known forward-backward algorithm [25] . More explicitly , given π , A , and the emission probabilities for the observations , log p ( N|z ) , the forward-backward algorithm returns the probabilities p ( zt = s ) ( posterior marginal ) , p ( zt+1 = s′ , zt = s ) ( two-slice marginal ) and log Z ( normalizing constant ) . Our final algorithm is presented in Algorithm 1 . Equation numbers reference posterior definitions in the text . Exact updates for the sufficient statistics are presented in Table 2 of S1 Text . Algorithm 1 Iterative update for variational inference 1: procedure Iterate 2: Update baselines λ0 ▷ conjugate Gamma Eq ( 12 ) 3: Update baseline hyperparameters γ0 ▷ conjugate Gamma ( Eqs 18 and 19 ) 4: for k = 1 … K do 5: Update firing rate effects λzk ▷ conjugate Gamma Eq ( 13 ) 6: Update firing rate hyperparameters γzk ▷ conjugate Gamma ( Eqs 18 and 19 ) 7: Calculate expected log evidence ηk ▷ ( S13 ) 8: Update Markov chain parameters A ˜ k , π ˜ k ▷ ( S11 , S12 ) 9: ξk , Ξk , log Zk← FORWARD-BACKWARD ( η k , A ˜ k , π ˜ k ) ▷ [26 , 27] 10: if semi-Markov then 11: Update duration distribution pk ( d|j ) ▷ BFGS optimization ( S25 ) 12: end if 13: Update cached F ▷ ( S8 ) 14: end for 15: Update covariate firing effects λx ▷ BFGS optimization ( Eq 14 , S54 , S55 ) 16: Update cached G ▷ ( S9 ) 17: Update overdispersion θ ▷ conjugate Gamma ( Eqs 18 and 19 ) 18: end procedure
We generated synthetic data from the model for U = 100 neurons for T = 10 , 000 time bins of dt = 0 . 0333s ( ≈ 6min of movies at 30 frames per second ) . Assumed firing rates and effect sizes were realistic for cortical neurons , with mean baseline rates of 10 spikes/s and firing rate effects given by a Gamma ( 1 , 1 ) distribution for Kdata = 3 latent features . In addition , we included R = 3 known covariates generated according to Markov dynamics . For this experiment , we assumed that each unit was presented only once with the stimulus time series , so that Mtu = 1 . That is , we tested a case in which inference was driven primarily by variability in population responses across stimuli rather than pooling of data across repetitions of the same stimulus . Moreover , to test the model’s ability to parsimoniously infer features , we set K = 5 . That is , we asked the model to recover more features than were present in the data . Finally , we placed hierarchical priors on neurons’ baseline firing rates and sparse hierarchical priors on firing rate effects of latent states . We used 10 random restarts and iterated over parameter updates until the fractional change in L dropped below 10−4 . As seen in Fig 3 , the model correctly recovers only the features present in the original data . We quantified this by calculating the normalized mutual information I ^ ≡ I ( X , Y ) / H ( X ) H ( Y ) , between the actual states and the inferred states , with H ( Z ) and I estimated by averaging the variational posteriors ( both absolute and conditioned on observed states ) across time . Note that superfluous features in the model have high posterior uncertainty for zk and high posterior confidence for λzk around 1 ( no effect ) . We applied our model to a well-studied neural data set comprising single neuron recordings from macaque area LIP collected during the performance of a perceptual discrimination task [28 , 29] . In the experiment , stimuli consisted of randomly moving dots , some percentage of which moved coherently in either the preferred or anti-preferred direction of motion for each neuron . The animal’s task was to report the direction of motion . Thus , in addition to 5 coherence levels , each trial also varied based on whether the motion direction corresponded to the target in or out of the response field as depicted in Fig 4 . ( In the case of 0% coherence , the direction of motion was inherently ambiguous and coded according to the monkey’s eventual choice . ) For our experiment , we only analyzed correct trials , on which the animal’s choice ( target IN or OUT of response field ) was synonymous with the direction of dot motion . We fit a model with K = 10 features and U = 27 units to neural responses from the 1-second stimulus presentation period of the task . Spike counts corresponded to bins of dt = 20ms . For this experiment , units were individually recorded , so each unit experienced a different number of presentations of each stimulus condition , implying a ragged observation matrix . As a result , this dataset tests the model’s ability to leverage shared task structure across multiple sessions of recording , demonstrating that simultaneously recorded units are not required for inference of latent states . Fig 4 shows the experimental labels from the concatenated stimulus periods , along with labels inferred by our model . Once again , the model has left some features unused , but correctly discerned differences between stimuli in the unlabeled data . Even more importantly , though given the opportunity to infer ten distinct stimulus classes , the model has made use of only five . Moreover , the discovered features clearly recapitulate the factorial design of the experiment , with the two most prominent features , Z1 and Z2 , capturing complementary values of the variable with the largest effect in the experiment: whether or not the relevant target was inside our outside the receptive field of the recorded neuron . This difference can be observed in both the averaged experimental data and the predicted data from the model ( see Fig 4C ) , where the largest differences are between the dotted and solid lines . Finally , we can ask whether the reconstructed firing rates are in quantitative agreement with the data estimates by calculating an RMS error for each curve in Fig 4C . That is , we calculate E [ ( f i - f a ) 2 ] E [ f i ] E [ f a ] for each unit , where fi is the inferred firing rate from the model , fa is the mean firing rate estimated from data , and expectations are taken across time bins . For our model , these values range from 4% to 12% across coherence levels . But the model also reproduces less obvious features: it correctly discriminates between two identical stimulus conditions ( 0% coherence ) based on the monkey’s eventual decision ( In vs Out ) . In addition , the model correctly captures the initial 200ms “dead time” during the stimulus period , in which firing rates remain at pre-stimulus baseline . ( Note that the timing is locked to the stimulus and consistent across trials , not idiosyncratic to each trial as in [30] . ) Finally , the model resists detection of features with little support in the experimental data . For instance , while feature Z4 captures the large difference between 50% coherence and other stimuli , the model does not infer a difference between intermediate coherence levels that are indistinguishable in this particular dataset . That is , mismatches between ground truth labels and model-inferred features here reflect underlying ambiguities in the neural data , while the model’s inferred features correctly pick out those combinations of variables most responsible for differences in spiking across conditions . As a second test of our model , we applied our algorithm to a designed structured stimuli dataset comprising U = 56 neurons from macaque inferotemporal cortex [31] . These neurons were repeatedly presented with 96 stimuli comprising 8 categories ( M = 1483 total trials , with each stimulus exposed between 12 to 19 times to each unit ) comprising monkey faces , monkey bodies , whole monkeys , natural scenes , food , manmade objects , and patterns ( Fig 5A ) . Data consisted of spike time series , which we binned into a 300ms pre-stimulus baseline , a 300ms stimulus presentation period , and a 300ms post-stimulus period . Three trials were excluded because of the abnormal stimulus presentation period . To maximize interpretability of the results , we placed strong priors on the πk to formalize the assumption that all features were off during the baseline period . We also modeled overdispersion with extremely weak priors to encourage the model to attribute fluctuations in firing to noise in preference to feature detection . We again fit K = 10 features with sparse hierarchical priors on population responses . The inferred categories based on binned population responses are shown in Fig 5B . For clarity , in Fig 5 , we only show population mean effects with a > 5% gain modulation sorted from the highest to the lowest , though the full set of inferred states can be found in Fig 6 . Out of the original categories , our model successfully recovers three features clearly corresponding to categories involving monkeys ( Features 0–2 ) . These can be viewed additively , with Feature 0 exclusive to monkey face close-ups , Feature 1 any photo containing a monkey face , either near or far; and Feature 2 any image containing a monkey body part ( including faces ) ; but as we will argue , given the nature of the model , it may be better to view these as a “combinatorial” code , with monkey close-ups encoded as 0&1&2 ( ∼ 59 . 46% increase in firing ) , whole monkeys as 1&2 ( ∼ 32 . 47% increase ) , and monkey body parts as 2 ( ∼ 7 . 62% increase ) . Of course , this is consistent with what was found in [31] , though our model used no labels on the images . And our interpretation that these neurons are sensitive to close-ups and faraway face and body parts is consistent with findings by another study using different experimental settings [32] . Again , as noted above , our results in Fig 5A and 5B indicate predicted population responses , derived from the hierarchical prior . As evidenced in Fig 5C and 5D , individual neuron effects could be much larger . These panels show data for two example units , along with the model’s prediction . Clearly , the model recapitulates the largest distinctions between images in the data , though the assumption that firing rates should be the same for all images with similar features fails to capture some variability in the results . Here , RMS errors range from 16% to 238% across units , with most units showing at least qualitative agreement from only a handful of presentations of each stimulus . Even so , uncertainties in the predicted firing rates are also in line with uncertainties from those of observed rates , indicating that our model is correctly accounting for trial-to-trial noise . Finally , even the weaker , sparser features inferred by our model captured intriguing additional information . As shown in Fig 6 , Feature 4 , a feature only weakly present in the population as a whole ( and thus ignored in Fig 6A ) , when combined with the stronger Features 0 , 1 , and 2 , successfully distinguishes between the monkey close-ups with direct and averted gaze . ( Stimulus 5 , with averted gaze , is additionally tagged with Feature 5 , which we view as an imperfect match . ) Thus , despite the fact that Feature 4 is barely a 3 . 4% gain change over the population , it suggests a link between neural firing and gaze direction , one for which there happens to be ample evidence [33 , 34] . Similarly , Feature 5 , barely a 1 . 1% effect , correctly tags three of the four close-ups with rightward gaze ( with one false positive ) . Clearly , neither of these results is dispositive in this particular dataset , but in the absence of hypotheses about the effect of head orientation and gaze on neuronal firing , these minor features might suggest hypotheses for future experiments . An additional feature of our approach is that the generated labels provide a concise and fairly complete summary of the stimulus-related activity of all neural recordings , which can be observed by comparing the categorization performance of decoded neural activity to the categorization performance of the decoded features . Although our model is not a data compression method , it nonetheless preserves most of the information about image category contained in the N = 56 dimensional spike counts via a 10-dimensional binary code . That is , using a sparse logistic regression on two-bit and three-bit combinations of our features to predict stimulus category ties and outperforms , respectively a multinomial logistic regression on the raw spike counts ( see Supplementary Information ) .
Here , we have proposed and implemented a method for learning features in stimuli via the responses of populations of spiking neurons . This work addresses a growing trend in systems neuroscience—the increasing use of rich and unstructured or structured stimulus sets—without requiring either expert labeling or a metric on the stimulus space . As such , we expect it to be of particular use in disciplines like social neuroscience , olfaction , and other areas in which the real world is complex and strong hypotheses about the forms of the neural code are lacking . By learning features of interest to neural populations directly from neural data , we stand to generate unexpected , more accurate ( less biased ) hypotheses regarding the neural representation of the external world . Here , we have validated this method using structured , labeled stimuli more typical of neuroscience experiments , showing that our model is capable of parsimoniously and correctly inferring features in the low signal-to-noise regime of cortical activity , even in the case of independently recorded neurons . Furthermore , by employing a fully variational , Bayesian approach to inference , we gain three key advantages: First , we gain the advantages of Bayesianism in general: estimates of confidence in inferences , parsimony and regularization via priors , and the ability to do principled model comparison . Second , variational methods scale well to large datasets and can be easily parallelized when combining data from multiple recording sessions . Finally , variational methods are fast , in that they typically converge within only a few tens of iterations and in many case ( such as ours ) can be implemented using explicit coordinate update rules , eliminating the need to tune a learning rate . Finally , even small features in our model recapitulated known physiological results regarding face encoding in single neurons . And while these features alone might not provide proof positive of , e . g . , viewpoint tuning , similar findings would be valuable in generating hypotheses in cases where the stimulus space and its neural correlates remain poorly understood . Thus our model facilitates an iterative experimental process: subjects are first be exposed to large , heterogeneous data; stimuli are then tagged based on neural responses; and finally , features with the largest effects are used to refine the set until it most accurately represents those stimuli with the largest neural correlates . Combined with the modularity of this and similar approaches , such models provide a promising opportunity to “build out” additional features that will meet the challenges of the next generation of experimental data . | Many neuroscience experiments begin with a set of reduced stimuli designed to vary only along a small set of variables . Yet many phenomena of interest—natural movies , objects—are not easily parameterized by a small number of dimensions . Here , we develop a novel Bayesian model for clustering stimuli based solely on neural responses , allowing us to discover which latent features of complex stimuli actually drive neural activity . We demonstrate that this model allows us to recover key features of neural responses in a pair of well-studied paradigms . | [
"Abstract",
"Introduction",
"Model",
"Results",
"Discussion"
] | [
"action",
"potentials",
"medicine",
"and",
"health",
"sciences",
"markov",
"models",
"membrane",
"potential",
"vertebrates",
"electrophysiology",
"neuroscience",
"animals",
"mammals",
"primates",
"optimization",
"systems",
"science",
"mathematics",
"old",
"world",
"monkeys",
"computer",
"and",
"information",
"sciences",
"monkeys",
"animal",
"cells",
"hidden",
"markov",
"models",
"dynamical",
"systems",
"probability",
"theory",
"macaque",
"cellular",
"neuroscience",
"cell",
"biology",
"physiology",
"neurons",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"physical",
"sciences",
"amniotes",
"neurophysiology",
"organisms"
] | 2017 | Neuron’s eye view: Inferring features of complex stimuli from neural responses |
In nature , arthropod-borne viruses ( arboviruses ) perpetuate through alternating replication in vertebrate and invertebrate hosts . The trade-off hypothesis proposes that these viruses maintain adequate replicative fitness in two disparate hosts in exchange for superior fitness in one host . Releasing the virus from the constraints of a two-host cycle should thus facilitate adaptation to a single host . This theory has been addressed in a variety of systems , but remains poorly understood . We sought to determine the fitness implications of alternating host replication for West Nile virus ( WNV ) using an in vivo model system . Previously , WNV was serially or alternately passed 20 times in vivo in chicks or mosquitoes and resulting viruses were characterized genetically . In this study , these test viruses were competed in vivo in fitness assays against an unpassed marked reference virus . Fitness was assayed in chicks and in two important WNV vectors , Culex pipiens and Culex quinquefasciatus . Chick-specialized virus displayed clear fitness gains in chicks and in Cx . pipiens but not in Cx . quinquefasciatus . Cx . pipiens-specialized virus experienced reduced fitness in chicks and little change in either mosquito species . These data suggest that when fitness is measured in birds the trade-off hypothesis is supported; but in mosquitoes it is not . Overall , these results suggest that WNV evolution is driven by alternate cycles of genetic expansion in mosquitoes , where purifying selection is weak and genetic diversity generated , and restriction in birds , where purifying selection is strong .
West Nile virus ( WNV , family Flaviviridae: Flavivirus ) is an arthropod-borne virus ( arbovirus ) that has demonstrated remarkable success since being introduced to North America in 1999 . Within three years after its introduction the virus had adapted to local mosquito vectors and within 8 years had become fully endemic [1] , [2] , [3] . Viruses with RNA genomes , like WNV , have higher mutation rates than those of most DNA viruses due to error-prone replication [4] . However , arboviruses seem to evolve more slowly compared to single-host RNA viruses [5] . The trade-off hypothesis is a commonly postulated theory suggesting that this slower rate derives from the biological requirement for alternating replication in two taxonomically divergent hosts ( vertebrates and arthropods ) . Under the trade-off hypothesis , virus fitness in both hosts is reduced in comparison to single host viruses , which can “specialize” on a single host environment [Recently reviewed by Ciota and Kramer [6]] . Several studies have reported that releasing arboviruses from host alternation and allowing sustained replication in a single host results in rapid adaptation to the specialized host , often with a corresponding fitness loss in the bypassed host , providing support for the trade-off hypothesis [7] , [8] , [9] , [10] . Nonetheless , considerable ambiguity exists in the literature concerning the impact of host alternation on arbovirus adaptation and fitness . Importantly , neither the receptor-ligand interactions most important for virus entry and tropism nor the intracellular resources that might form the basis for host specialization in a putative fitness “trade-off” are well understood . Most studies of the trade-off hypothesis have involved either flaviviruses or alphaviruses ( family Togaviridae ) , two of largest , most medically relevant families of arboviruses . Results of these studies are inconsistent and seem to differ between virus families ( alphaviruses vs . flaviviruses ) and experimental systems ( cell-culture vs . animals ) . Among flaviviruses , the trade-off hypothesis is often only partially supported . Host specialization frequently results in fitness increases; however , that these increases carry a fitness cost in the bypassed host is less well supported . Work in cell-culture with dengue virus ( genus flavivirus ) has shown that single-host-specialized virus replicated faster and reached higher titers in the specialized cell-line but reciprocal fitness losses were less extreme and inconsistent [10] , [11] . Another cell-culture study found that mosquito cell-specialized WNV and St . Louis encephalitis virus ( SLEV , flavivirus ) displayed improved fitness and more rapid replication in mosquito cells with only modest and inconsistent fitness losses in chicken cells [12] . In vivo studies with flaviviruses have also been difficult to reconcile with the trade-off hypothesis . For example , chick-specialized SLEV showed increased infectivity in chicks but was unchanged in mosquitoes , while mosquito-specialized virus was unchanged in both systems [13] . Conversely , serial passage of WNV in mosquitoes resulted in faster replication and higher peak titers in mosquitoes with no significant cost to replication in live chicks [14] . The impact of extensive in vivo serial passage on fitness of WNV within biologically relevant hosts has been difficult to resolve because the determinants of virus fitness in either host ( mosquitoes or birds ) have been poorly understood . Recent advances have provided a more complete mechanistic understanding of in vivo fitness determinants that may shed light on phenomena previously attributed to “trade-offs” . For example , genetic diversification of WNV is driven by , and can circumvent , mosquito immune mechanisms [15] . Additionally , the avian environment applies purifying selection to virus populations , but mosquitoes do not [16] , [17] . Observed fitness “trade-offs” may thus be partially attributable to diversity-permissive and -restrictive environments in mosquitoes and birds , respectively . In light of this , we re-examined the trade-off hypothesis by determining the impact of host specialization on WNV fitness , here defined as the capacity for successful genome replication . In particular , ( a ) bird-specialized WNV , ( b ) mosquito-specialized WNV , ( c ) alternately passed WNV and ( d ) unpassed WNV , were competed against genetically marked WNV in vivo in mosquitoes and chickens . In previous studies we passed WNV exclusively in chicks or mosquitoes 20 times , or passed the virus alternately between mosquitoes and chicks a total of 20 times ( figure 1 ) [17] , [18] . In this study , the resulting WNV was competed against unpassed marked reference virus ( WNV-REF ) , derived from the same clone used for passage initiation in order to determine whether host specialization leads to fitness gains and/or losses in the WNV system . Our studies , through the use of in vivo model animals for both passage and fitness determination , as well as triplicate performance of each treatment and the use of a higher passage number than typically used , provide a more representative model of the effect of host specialization on WNV fitness and the trade-off hypothesis , than has been possible with cell-culture models .
When competitions were conducted in chickens , serially and alternately passed WNV demonstrated clear fitness changes . Serial passage in chicks resulted in fitness increases compared to unpassed virus in the homologous host ( table 1 , figure 2 , unpaired t-test , P = 0 . 0102 ) . Conversely , after serial passage in mosquitoes WNV displayed significantly decreased replicative fitness in chicks ( P = 0 . 0056 ) . WNV from the alternating passage series also experienced significant fitness changes . Alternating passage that concluded in chickens demonstrated fitness gains ( P = 0 . 0061 ) but alternating passage that concluded in mosquitoes did not ( P>0 . 05 ) . Unpassed WNV-WT ( wild-type , unmarked WNV derived from the same clone used to generate the marked reference virus ) had fitness similar to WNV-REF ( table 1 ) . Sera and brain tissue from all chick cohorts were also collected on day 5 post-inoculation and showed no significant difference from day 2 sera ( data not shown ) . When competitions were conducted in Cx . pipiens , fitness changes also were observed ( figure 2 ) . Chick-passed WNV displayed significant replicative fitness increases over the unpassed marked reference virus ( P = 0 . 0056 ) . Additionally , after 20 serial passes in mosquitoes replicative fitness increases were observed ( P = 0 . 0291 ) . Neither the viruses from the alternate passage series nor the control WNV-WT exhibited significant changes in replicative fitness ( table 1 ) . In contrast to the results from the competition experiments in chicks and Cx . pipiens there were no significant replicative fitness changes for any of the serially or alternately passed viruses when competed against the unpassed marked reference virus in Cx . quinquefasciatus mosquitoes ( table 1 , figure 2 ) . To determine whether intrahost WNV genetic diversity was associated with fitness in the three hosts in which measurements were conducted , fitness was plotted as a function of viral genetic diversity ( figure 3 ) and dN/dS ( data not shown ) . Fitness was computed as the difference in proportion of test virus after competition compared to its proportion at input . Sequence diversity and dN/dS were computed as described previously [17] , [18] . In mosquitoes , intrahost genetic diversity was not correlated with fitness ( for Cx . pipiens , Spearman r = −0 . 8117 , P>0 . 05 , for Cx . quinquefasciatus , Spearman r = −0 . 05798 , P>0 . 05 ) . In chickens , diversity was significantly negatively associated with fitness ( r = −0 . 9856 , P = 0 . 0028 ) . Similar results were obtained when the relationship between dN/dS and fitness was analyzed , with increasing dN/dS associated with lower fitness in chickens , but not mosquitoes ( not shown ) .
Because arboviruses replicate in both arthropod and vertebrate hosts and seem to evolve more slowly than single-host RNA viruses , it is often proposed that they “trade” optimal fitness in either host in exchange for adequate fitness in both . Tests of this “trade-off” hypothesis most often consist of releasing a virus from host alternation and allowing it to specialize on one host or the other , then comparing fitness or genetic sequence data to the unpassed or alternately passed virus [9] , [10] , [13] , [18] , [19] . Due to the complexity of arbovirus transmission cycles , and in many cases the lack of appropriate in vivo models ( for dengue virus , for example ) these studies have largely been conducted in vitro in tissue culture , with inconsistent results . This lack of consistency appears to be related to differences in virus families , host species , passage regimes and approaches to measuring virus fitness . In most cases , the mechanistic basis for observed trade-offs have not been identified . Moreover , the diversity of experimental systems has made it difficult to identify the merits and defects of the trade-off hypothesis . Here , we used a completely in vivo approach to test whether or not WNV host alternation supports the trade-off hypothesis . By conducting the passage series and the competitions in relevant hosts in vivo , we sought to circumvent several of the caveats required in interpreting many previous studies . In sum , our data support the growing body of evidence that the trade-off hypothesis does not accurately predict WNV population dynamics [14] , [20] . Interestingly , our findings are somewhat at odds with a similar literature developing in the field of alphavirus-host interactions , which tend to support the trade-off hypothesis [7] , [8] , [9] , [19] . The reasons for this are not entirely clear , but may be related to differences in virus replication in vivo ( i . e . differences in host factors required for replication or host-cell receptor utilization ) . Differences in replication and/or mutation rates could also impact genetic diversity or population composition resulting in fitness changes . Additional comparative studies are required to develop a complete understanding of the underlying differences between fitness trade-offs in flaviviruses compared to alphaviruses . Serial passage in chickens resulted in fitness gains in both chickens and Cx . pipiens mosquitoes , but fitness in a related mosquito species ( Cx . quinquefasciatus ) was unchanged . These results are at odds with the trade-off hypothesis because although the observed fitness increases in chickens would have been predicted , expected losses in the bypassed host ( mosquitoes ) were not observed . Notably , the WNV strains that had undergone sequential passage exclusively in chickens had patterns of nucleotide substitution suggesting that they were subject to strong purifying selection during replication in chickens [17] . In addition , intrahost genetic diversity in general was very low after passage in chickens . Collectively these observations suggest that during WNV replication in chickens , high overall fitness is maintained because deleterious mutations are rapidly removed by selection . Serial passage of WNV in mosquitoes resulted in slight fitness gains in one species of mosquito ( Cx . pipiens , the host in which the virus was sequentially passed ) , no change in a related species ( Cx . quinquefasciatus ) , and extreme fitness losses in chickens . These findings seem to support the trade-off hypothesis . Purifying selection is relaxed in mosquitoes leading to high genetic diversity [16] , [17] . It therefore seems likely that much of the genetic diversity generated during mosquito infection consists of mutations that are selectively neutral in mosquitoes but are slightly or strongly deleterious in the chick environment , leading to chick-specific fitness declines . This observation is supported by our analysis of the relationship between virus fitness and the genetic diversity within the WNV test population ( figure 3 ) . Moreover , our results suggest that the mechanistic basis for the observed fitness trade-off following mosquito passage is likely related to intrahost genetic diversity and different selective environments in each host type . Alternating passage of WNV generally produced negligible fitness changes , largely in accordance with the trade-off hypothesis , although minor non-significant changes are apparent upon visual inspection of the data in figure 2 . Possible explanations for adaptation in the absence of genetic coding change may include post-transcriptional modification or codon usage differences between the two environments ( avian and mosquito ) that may influence replication efficiency in the subsequent heterologous host . Concluding alternating passage in both mosquitoes and chicks also permitted us to examine the impact of two serial passes in each type of host . Interestingly , fitness gains were observed in chicks when they were the final host for alternating passage . These cohorts essentially represent two serial passes in chicks – the final pass of this series was in chicks and the subsequent competition was in chicks . After 20 alternate passes , two serial passes produced fitness gains comparable in magnitude to those observed after 20 serial passages . This finding underscores our understanding that purifying selection is extremely strong in these hosts . Importantly , our passage regimen of 20 passes may not be robust enough to allow establishment of equilibrium for the virus populations being examined . Passage was terminated after 20 rounds because the logistics of serial passage become more restrictive when working with an in vivo model compared to an in vitro model . It is possible that with additional passage , the predictions of the trade-off hypothesis may be better satisfied in the invertebrate host . However , most data thus far indicates that mosquitoes are a diversity-permissive environment for WNV [17] , [21] . Therefore , it is not clear how many additional passages would be required to achieve equilibrium in these hosts . The trade-off hypothesis is only partially supported , and in a host-dependent manner , by our findings . When competed in chickens both single-host-specialized virus cohorts conformed to the predictions of the trade-off hypothesis; chick-specialized virus showed increased fitness and Cx . pipiens -specialized virus showed decreased fitness . Most of the data from this study , however , do not conform to the predictions of the trade-off hypothesis . When competed in mosquitoes , all chick-specialized as well as Cx . pipiens-specialized viruses displayed significant fitness gains . Whereas all passage series resulted in at least moderately improved fitness in Cx . pipiens , no passage series resulted in significant fitness changes in the Cx . quinquefasciatus environment . A similar study looking only at mosquito-specialized WNV also reports replicative fitness increases in mosquitoes without a corresponding cost in chicks [14] . Overall , these data suggest that the trade-off hypothesis , as conventionally stated , does not accurately predict WNV transmission dynamics because it fails to incorporate the mechanistic basis underlying fitness differences . Specifically , high mutational diversity of WNV increases fitness in mosquitoes by facilitating escape from their dominant RNAi-based antivirus response [15] . This fitness advantage in mosquitoes carries a selective cost in chickens because putative mosquito RNAi-escape mutations likely negatively impact virus replication . Higher fitness gains for all four passage regimes were observed in Cx . pipiens compared to Cx . quinquefasciatus . These sibling species are primary vectors of WNV in the northern and southern United States , respectively [22] . Despite their close taxonomic relationship , differences have been noted previously in vector competence between the two species following feeding on WNV [1] , [23] , [24] . Mosquito passage of WNV was conducted in Cx . pipiens , which may account for the larger fitness increases observed during competitions in Cx . pipiens . However , this does not explain the puzzling extreme fitness gains for chick-specialized virus when competed in Cx . pipiens . We think it likely that after undergoing continuous purifying selection in the bird environment the virus replication is very efficient when it is then placed in the relatively permissive Cx . pipiens environment . It is possible that RNAi responses of differing magnitudes may contribute to the disparity observed for chick-specialized virus in the two mosquito species examined . It has been shown that the RNAi pathway in Cx . quinquefasciatus promotes genetic diversification [15] , however no data for Cx . pipiens are available and the relative magnitudes of the RNAi have not yet been examined . Overall , our divergent results in Cx . pipiens and Cx . quinquefasciatus suggest that fitness determinants may be mosquito species-dependent . In our studies , mosquito infection for passage and competition was achieved through intrathoracic ( IT ) -inoculation , which bypasses the mosquito midgut . This method was chosen because achieving adequately high virus titers for oral blood-feeding would require further passage of the virus in cell-culture and would potentially confound any effects of serial passage . Midgut infection and/or escape is considered a major bottleneck to vector infection by arboviruses [25] . This restriction is likely specific to infection in general and not one that necessarily affects the genetic composition of the virus population achieving infection . Recently we have shown that genetic bottlenecks within Cx . quinquefasciatus do not significantly reduce WNV population diversity during horizontal transmission [26] . However , midgut infection and escape barriers cannot be entirely ruled out as influencing virus population genetics through , for example , selective constraints ( i . e . as opposed to stocahstic effects ) . Because our infections were done by IT-inoculation , any selective constraints such as those imposed by transmission “barriers” in the natural transmission cycle were overcome . Nonetheless , the methods and approaches used to accomplish the passages described here allowed us to examine the impact of replication in divergent hosts in the absence of several factors ( such as barriers and co-infections , for example ) that would likely be present in nature . The complete genome sequences for the endpoint viruses from all serially or alternately passed WNV lineages used in the current study have been previously published [17] , [18] . Numerous synonymous and non-synonymous mutations were found in both structural and non-structural coding regions , but no signature mutations were found to be associated with any passage series and there was no data to suggest that adaptation had occurred . Importantly , individual virus isolates are known to comprise a mutant swarm that may contain minority genotypes not detectable in the consensus sequence that may exhibit dominant phenotypes [27] . Recent developments in deep sequencing technology will greatly facilitate future efforts at understanding the contributions of individual quasispecies to the overall fitness of arboviruses [28] . In conclusion , when released from the obligate cycling between avian and mosquito hosts , WNV experienced symmetrical fitness gains in specialized hosts but fitness losses in bypassed hosts were asymmetrical . In the avian environment fitness trade-offs are apparent and robust; however , in the mosquito environment no obvious fitness trade-offs were observed . These data are consistent with previously published work showing that the mosquito environment permits a much higher level of viral genetic diversity than the avian environment [16] , [17] . Our results add to a growing amount of evidence that arboviruses in general do not fall into an intuitive pattern represented by the host trade-off assumption . WNV adaptation and evolution therefore seem likely to be driven by alternating between diversity-permissive and diversity-restrictive environments in the invertebrate and vertebrate hosts . Mosquito infection enables the development of genetic diversity and novel variants of WNV , while infection of birds applies purifying selection that maintains high replicative fitness .
Experiments involving animals were conducted in accordance with protocols approved by the University of New Mexico Institutional Animal Care and Use Committee in strict adherence to recommendations set forth in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health ( Assurance No . A3350-01 ) , and approved by the University of New Mexico IACUC ( protocol # 100450 ) . The wild-type virus ( WNV-WT ) that was used to initiate all passage series was derived from a WNV infectious clone as previously described [29] . The genetically marked reference virus ( WNV-REF ) was also derived from a WNV infectious clone and has been previously characterized [21] . This reference virus is identical to WNV-WT except for five sequential non-coding changes in the NS5 region of the genome from nucleotide positions 8313–8317 . These changes were engineered using site-directed mutagenesis as described previously [30] and changed the parental sequence CTC TCA CGG to CTa agc aGG without altering the amino acid sequence or the replication kinetics and infectivity of the virus [21] . Viral RNA for WNV-WT and WNV-REF was electroporated into baby hamster kidney ( BHK ) cells and progeny virus was harvested directly and used without further cell-culture passage . Serial and alternate passage of WNV-WT in chicks and mosquitoes has been previously described [17] , [18] ( figure 1 ) . Briefly , 1–3 day old chicks [Charles River Specific Pathogen Free Avian Services ( Franklin , CT ) or Sunrise Farms ( Catskill , NY ) ] were inoculated with 100 times the ID50 in three replicate concurrent lineages . At day 2 post-inoculation , serum was harvested , titrated to determine the correct dilution for re-inoculation and used to inoculate 100 times the ID50 into the next round of 1–3 day old chicks . After 20 such serial passes , the end-point sera for each of the three replicate concurrent lineages were harvested , titered and stored at −80°C until competition experiments were conducted . Adult , female Cx . pipiens mosquitoes ( colony derived from larvae collected in Pennsylvania and maintained at the Wadsworth Center Arbovirus Laboratories since 2002 ) were IT-inoculated with 100 times the ID50 . Three concurrent replicate lineages were maintained , with approximately ten individual mosquitoes per replicate inoculated . At day 7 post-inoculation , individual mosquitoes were triturated and homogenates were clarified by centrifugation , titrated to determine the correct dilution for re-inoculation , and used to inoculate 100 times the ID50 into the next cohort of mosquitoes . A single mosquito with the median viral load was selected for further passage . After 20 such serial passes , the end-point homogenates were titrated and stored at −80°C until competition experiments were conducted . Alternating passage was also conducted in three concurrent replicate lineages and was begun with Cx . pipiens IT-inoculation as described above . After trituration , clarified homogenates were used to inoculate 100 times the ID50 into chicks as described above . Day 2 chick serum was then used to inoculate 100 times the ID50 into the next cohort of Cx . pipiens and alternate passage continued for 10 complete cycles or 20 total virus passes . Alternating passage was concluded in each host type to evaluate the possibility that a single round of replication in one or the other host might influence virus fitness . After passage , the resulting viruses were characterized with respect to both complete genome sequence and population diversity . Results of these studies are described in detail in previous publications [17] , [18] . Briefly , complete genome sequences were unremarkable , with no consistent changes noted at the genome level . However , purifying selection was associated with exclusive passage in chickens or alternating passage , with mosquito-passed WNV lacking evidence of purifying selection . Intrahost genetic diversity was related to passage history , with higher diversity associated with exclusive passage in mosquitoes , or alternating passage , but not with exclusive passage in chickens . These genetically characterized , passed viruses constitute the “test” viruses for competition studies . The inocula for competitions were created by mixing equal number of plaque forming units ( pfu ) of WNV-REF and passed “test” WNV . In most cases the inocula were prepared in advance , aliquoted and stored at −80°C with a fresh aliquot used for each competition . In 5 of the 13 Cx . pipiens competitions the inocula were re-created and are comparable to those used in the corresponding chick and Cx . quinquefasciatus competitions . Chicks were reared and competitions were performed in the University of New Mexico's animal biosafety level-3 ( ABSL-3 ) laboratory . Specific-pathogen-free eggs were incubated and chickens hatched and maintained as described above and elsewhere [17] , [18] , [21] . At approximately 24 h post hatching , chicks were subcutaneously inoculated in the cervical region with 100 µL ( 2 . 5×102–2 . 5×105 total pfu ) of mixed 1∶1 test∶REF WNV in animal inoculation medium ( endotoxin- and cation-free phosphate buffered saline with 1% FBS ) then returned to their brooders . At 48 h post-inoculation approximately 50 µL blood was collected in heparinized capillary tubes after brachial venipuncture . Serum was separated and used for viral RNA isolation . Experiments involving animals were conducted in accordance with protocols approved by the University of New Mexico Institutional Animal Care and Use Committee in strict adherence to recommendations set for in the Guide for the Care and Use of laboratory Animals of the National Institutes of Health ( Assurance No . A3350-01 ) , and approved by the University of New Mexico IACUC ( protocol # 100450 ) . Mosquitoes were reared and competitions were performed in the University of New Mexico's bio-safety level-3 ( BSL-3 ) insectary . Culex mosquitoes were obtained from colonies at the University of New Mexico and the Wadsworth Center , New York State Department of Health . Mosquitoes were maintained at 27°C with a 16∶8 L∶D photoperiod and were used in competitions at 3–7 days post-emergence . Mosquitoes were anesthetized with CO2 and were IT-inoculated with 70–840 nl ( 2–18 total pfu ) of mixed 1∶1 test∶REF WNV in animal inoculation medium using a Nanoject II ( Drummond Scientific Company , Boomall PA ) . Inoculated mosquitoes were incubated in quart-sized cardboard containers with water and 10% sucrose provided ad libitum . At 7-days post inoculation whole individual mosquitoes were triturated using a TissueLyser ( Qiagen Inc . , Valencia , CA ) and homogenates were clarified by centrifugation then used for viral RNA isolation . Total viral RNA was isolated from chick sera or mosquito homogenates using the RNeasy RNA Purification Kit ( Qiagen Inc . , Valencia , CA ) . Reverse-transcriptase polymerase chain reaction ( RT-PCR ) was then performed as previously described [21] . Briefly , one-step RT-PCR was performed using the SuperScriptIII system ( Invitrogen Corporation , Carlsbad , CA ) and primers designed to amplify an 853 base-pair region containing the 5-nucleotide genetic marker . Amplicon DNA was then purified using the QIAquick PCR Purification Kit ( Qiagen Inc . , Valencia , CA ) and sequenced by the Sanger sequencing method ( Genewiz Inc . , South Plainfield , NJ ) . Sequence chromatograms were analyzed using the polySNP program ( http://staging . nybg . org/polySNP . html ) as described elsewhere [21] , [31] . The proportion of each genotype was then computed for each of the 5 nucleotide positions and the mean proportion of all 5 is reported as the overall proportion of each genotype in the DNA sample . Comparison of relative proportions for input ( inocula ) and output ( chick sera or mosquito homogenates ) was done by an unpaired t-test performed in the GraphPad software package . A t-test P-value of ≤0 . 05 was interpreted as statistically significant . The measure of competitive fitness as determine by quantitative sequencing was limited by the range of detection which was previously determined to be 10–90% [21] . Thus the virus that “loses” in the competition may still be present in biologically significant proportions despite being undetectable by the quantitative sequencing assay . Before and after inoculation into chicks or mosquitoes , virus preparations were titered to determine the amount of total WNV ( test-WNV plus WNV-REF ) . Inocula were serially diluted 10-fold in cell-culture medium [Eagle's minimum essential medium ( MEM ) containing 10% heat-inactivated fetal bovine serum ( FBS ) , penicillin/streptomycin ( 100 units/mL ) , gentamicin ( 50 µg/mL ) , L-glutamine ( 1× ) and fungizone ( 125 ng/mL ) ] , then adsorbed onto confluent monolayers of Vero cells at room temperature for 60 min with continual rocking . Cells were then overlaid with cell-culture medium containing 0 . 4% agarose and maintained at 37°C with 5% CO2 . After 48 h a second overlay of cell-culture medium containing 0 . 4% agarose and neutral red ( 66 µg/mL ) was applied and incubated for an additional 24 h before plaques became apparent and were counted for virus titer calculation . | Arthropod-borne viruses , including West Nile virus ( WNV ) have long been considered to be subject to a fitness “trade-off” because they must replicate in at least two taxonomically divergent hosts in order to perpetuate . Results of studies testing this trade-off hypothesis have been inconsistent and largely dependent on which virus family is studied and which experimental system used . Therefore , considerable ambiguity exists in the literature regarding how host alternation influences virus population biology . Accordingly , we allowed WNV to adapt to each of its main hosts ( mosquitoes and birds ) for 20 sequential passes in order to determine whether host alternation in nature imparts a fitness compromise , as predicted by the trade-off hypothesis . After host-specialization , passed viruses were competed in vivo in mosquitoes and chicks against an unpassed marked reference virus to determine whether replicative fitness gains or losses had occurred . Our results demonstrate that the trade-off hypothesis accurately predicts WNV adaptation in the avian environment but not in mosquitoes . Overall , our results suggest that WNV adaptation is controlled by alternating cycles of genetic expansion in a permissive mosquito environment and restriction in avians , where purifying selection is dominant . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"infectious",
"diseases",
"public",
"health",
"and",
"epidemiology",
"veterinary",
"diseases",
"zoology",
"epidemiology",
"biology",
"microbiology",
"evolutionary",
"biology",
"population",
"biology",
"public",
"health",
"veterinary",
"science"
] | 2011 | West Nile Virus Experimental Evolution in vivo and the Trade-off Hypothesis |
Craniofacial development requires signals from epithelia to pattern skeletogenic neural crest ( NC ) cells , such as the subdivision of each pharyngeal arch into distinct dorsal ( D ) and ventral ( V ) elements . Wnt signaling has been implicated in many aspects of NC and craniofacial development , but its roles in D-V arch patterning remain unclear . To address this we blocked Wnt signaling in zebrafish embryos in a temporally-controlled manner , using transgenics to overexpress a dominant negative Tcf3 , ( dntcf3 ) , ( Tg ( hsp70I:tcf3-GFP ) , or the canonical Wnt inhibitor dickkopf1 ( dkk1 ) , ( Tg ( hsp70i:dkk1-GFP ) after NC migration . In dntcf3 transgenics , NC cells in the ventral arches of heat-shocked embryos show reduced proliferation , expression of ventral patterning genes ( hand2 , dlx3b , dlx5a , msxe ) , and ventral cartilage differentiation ( e . g . lower jaws ) . These D-V patterning defects resemble the phenotypes of zebrafish embryos lacking Bmp or Edn1 signaling , and overexpression of dntcf3 dramatically reduces expression of a subset of Bmp receptors in the arches . Addition of ectopic BMP ( or EDN1 ) protein partially rescues ventral development and expression of dlx3b , dlx5a , and msxe in Wnt signaling-deficient embryos , but surprisingly does not rescue hand2 expression . Thus Wnt signaling provides ventralizing patterning cues to arch NC cells , in part through regulation of Bmp and Edn1 signaling , but independently regulates hand2 . Similarly , heat-shocked dkk1+ embryos exhibit ventral arch reductions , but also have mandibular clefts at the ventral midline not seen in dntcf3+ embryos . Dkk1 is expressed in pharyngeal endoderm , and cell transplantation experiments reveal that dntcf3 must be overexpressed in pharyngeal endoderm to disrupt D-V arch patterning , suggesting that distinct endodermal roles for Wnts and Wnt antagonists pattern the developing skeleton .
A fundamental question in skeletal biology is how cartilages and bones with distinct shapes arise from skeletogenic precursor cells . Much of the craniofacial skeleton derives from neural crest ( NC ) cells that migrate in streams into the pharyngeal arches and contain anterior-posterior ( A-P ) patterning information obtained prior to migration [1]–[3] . However , these NC cells also become intimately associated with epithelia , including surface ectoderm and pharyngeal endoderm , which produce signals important for skeletal patterning . For example , Fgf8 from the facial ectoderm regulates A-P polarity of the mandibular arch as well as NC proliferation/survival [4]–[6] . Surgical disruption of the pharyngeal endoderm in chick [7] , or mutations that disrupt endoderm in zebrafish , lead to severe cartilage malformations [8]–[10] . Endoderm-derived Fgf3 induces cartilage formation [9] and sphingosine phosphate-1 from endoderm modulates Shh signaling to promote mandibular growth and patterning [11]–[13] . Ectoderm-derived Shh induces upper jaw and neurocranial structures [14] , [15] . Thus , craniofacial skeletal shapes reflect interplay between epithelial signals and intrinsic properties of mesenchyme , but the mechanisms underlying these interactions remain unclear . One well-studied example of such epithelial-mesenchymal interactions in the pharyngeal skeleton is the induction of ventral skeletal fates along the dorsal-ventral ( D-V ) axis of the mandibular arch by ectodermal Endothelin 1 ( Edn1 ) and Bone morphogenetic proteins ( Bmps ) [16]–[22] . Conditional loss of Bmp4 in the facial ectoderm in mice inhibits ventral mandibular growth and patterning [23] . Loss of Edn1 and/or any of several components of its signal transduction pathway leads to severe jaw truncations , both in mice and in humans , and in some cases transformations of ventral tissues ( the lower jaw ) to a more dorsal identity [22] , [24]–[27] . Initially both Edn1 and Bmps induce similar subsets of ventral/intermediate genes as well as restricting Jag1 signaling to the dorsal domain [16] , [19] , [20] , [28] . But once arch primordia are established , effects of Bmps become more ventrally restricted to domains that no longer depend on Edn1 , particularly the transcription factor Hand2 . Uniform application of Bmp or Edn1 proteins can restore many aspects of D-V patterning in Bmp- and Edn1-deficient zebrafish embryos , suggesting that other ventralizing signals must interact to control D-V patterning [22] . Wnts are good candidates for additional signals involved in D-V patterning based on their localized expression and known requirements in craniofacial development . Several Wnt ligands show restricted expression in facial epithelia ( ectoderm , endoderm ) in zebrafish , chick and mouse embryos [29]–[32] and Frizzled ( Fzd ) receptors are expressed throughout arch NC cells and endoderm [31] , [33] , [34] . In addition , expression of Lef1 , Tcf1 , β-catenin ( βcat ) , [35] , and transgenic Wnt signaling reporter lines ( Lef/Tcf promoters driving β-gal or LacZ expression ) in mice all are ventrally ( distally ) restricted in the mandibular as well as distal maxillary prominences [36] , [37] . Like Bmps , Wnt signaling is necessary for early NC cell induction [38] , [39] and also plays later roles in NC migration , fate specification , and proliferation [40] . In zebrafish , conditional overexpression of a dominant negative Tcf3 ( dntcf3 ) during NC cell specification dramatically reduces NC cell numbers [41] , similar to depletion of Fzd3 receptors in Xenopus [42] . Wnt1−/−/Wnt3−/− double mutant mice show reduced proliferation of pre-migratory NC cells [43] . Conditional loss of βcat in the pharyngeal ectoderm impairs growth of the facial prominences [44] , while conditional loss of βcat in cranial NC cells leads to apoptosis and a nearly complete loss of NC-derived craniofacial structures [45] . Finally , loss of Tcf4/Lef1 function or overexpression of the Wnt inhibitor Dickkopff-1 ( Dkk1 ) results in smaller facial structures and clefting between the frontonasal and maxillary prominences [36] . Similarly , Wnt signaling is important for facial midline development in humans as incidences of cleft lip and palate have been mapped to genetic variations in Wnt ligands [46] . In this study , we examine temporal requirements for Wnt signaling in zebrafish D-V craniofacial development . We utilize two transgenic lines , Tg ( hsp701:dkk1-GFP ) ( dkk1+ ) and Tg ( hsp70I:tcf3-GFP ) ( dntcf3+ ) , to interfere with Wnt signaling conditionally , in a stage-specific manner . Tg ( hsp701:dkk1-GFP ) embryos overexpress dkk1 , a secreted negative regulator , while Tg ( hsp70I:tcf3-GFP ) embryos overexpress a dominant negative form of the Tcf3 transcription factor . Both methods of inhibiting Wnt signaling after NC cell migration result in proliferation and ventral patterning defects in the mandibular and hyoid arches . Interestingly , dkk1+ embryos also show unique clefting of the lower jaw . Defects in ventral-intermediate specific gene expression and expansion of the dorsal specific jag1b resemble loss of Bmp and Edn1 signaling [19] , [20] . We show that Wnt signaling promotes Bmp signaling through regulation of expression of two specific Bmp receptors , bmpr1ab and bmpr1ba , in the pharyngeal arches . Ectopic Bmp protein can rescue msxe , dlx3b , dlx5a but not hand2 in the absence of Wnt signaling , demonstrating that Wnts participate in a regulatory network with Bmp and Edn1 signaling , but separately in hand2 regulation , to control D-V pharyngeal patterning . Chimeric analyses reveal that dntcf3 acts cell autonomously in pharyngeal endoderm , which also expresses dkk1 . This suggests that Wnts regulate patterning in the endoderm , which through some as yet unknown signal imparts D-V patterning upon neighboring skeletal progenitors in the NC .
Numerous Wnt ligands ( Wnt2 , Wnt4 , Wnt5a/b , Wnt 6 , and Wnt7a/b ) and receptors ( Fzd1 , Fzd3 , Fzd4 , Fzd6 , Fzd7 , Fzd8 , and Fzd10 ) are expressed broadly in the pharyngeal ectoderm , endoderm , neural crest ( NC ) , and mesoderm [29] , [32] , [34] , [35] . To determine which regions of the pharyngeal arches respond directly to Wnt signaling we used in situ hybridization ( ISH ) to examine expression of the direct downstream Wnt target mycn ( Fig . 1A , B ) , an oncogene with roles in regulating Wnt-dependent morphogenesis and proliferation [47] , [48] . mycn mRNA was detected throughout the arches but at higher levels in the ventral domain , primarily within the NC mesenchyme ( arrowheads in Fig . 1A , B ) . To further address which pharyngeal tissues respond directly , we examined expression of a transgenic Wnt reporter zebrafish Tg ( 7xTCF-Xla . Siam:GFP ) ia4 ( 7xTCF:GFP ) [49] , which contains seven TCF response elements driving expression of GFP , thus acting as a live reporter in cells where stabilized β-catenin ( βcat ) interacts with Tcf transcription factors . ISH for GFP mRNA at 28 hours postfertilization ( hpf ) revealed regions of 7xTCF:GFP expression in the ventral first and second arches ( Fig . 1C ) , which in transverse sections appeared localized both to arch NC cells and pharyngeal endoderm , but not pharyngeal ectoderm ( Fig . 1D ) . To bypass earlier requirements for Wnts in embryogenesis we took a conditional loss-of-function approach using heat shock-inducible transgenic zebrafish lines to inhibit Wnt signaling in a temporally-controlled manner . Tg ( hsp70I:tcf3-GFP ) ( hs-dntcf3 ) embryos overexpress a truncated form of the transcription factor tcf3 with GFP replacing the βcat-interacting domain , under control of heat shock promoter 70 [41] . With a similar hsp70 promoter , Tg ( hsp701:dkk1-GFP ) ( hs-dkk1 ) embryos overexpress full length dkk1b tagged with a GFP [50] , which prevents Fzd-Lrp co-receptor binding [51] , [52] . To verify that Wnt-βcat signaling was affected in hs-dkk1+ and hs-dntcf3+ embryos we used ISH to examine the expression of mycn and axin2 , both direct Wnt targets , after heat shocking during stages of craniofacial patterning . At 4 hours post heat shock ( hphs ) hs-dntcf3+ embryos heat shocked at 22 hpf showed severe reductions in expression of mycn in the arches , eyes , and brain and axin2 ( which shows only very weak or no expression in pharyngeal arches ) , in the eyes and brain ( Fig . 1E–H ) . Similarly , compared with controls ( Fig . S1A , C ) at 2 hphs hs-dkk1+ embryos heat shocked at 24 hpf showed reduced mycn in the arches , eye , and brain and axin2 expression in the brain but to a lesser extent than in hs-dntcf3+ embryos ( Fig . S1B , D ) . To determine stage-specific defects caused by disrupting Wnt signaling in hs-dntcf3+ and hs-dkk1+ embryos we performed quantitative real time PCR ( qPCR ) analysis of direct Wnt targets . In hs-dntcf3+ embryos , axin2 and mycn expression was reduced significantly from 2–8 hphs ( Fig . 1I ) , but recovered and slightly increased by 24 hphs ( Fig . 1I ) . Lef1 , a transcriptional cofactor in the Wnt pathway [53] , [54] , was also significantly reduced in hs-dntcf3+ embryos from 2–8 hphs ( Fig . 1I ) . In hs-dkk1+ embryos , axin2 expression was severely reduced from 1–8 hphs , while Lef1 downregulation started later , from 4–8 hphs ( Fig . S1E ) . Thus both hs-dntcf3 and hs-dkk1 lines deplete Wnt signaling almost immediately after heat shock and repress Wnt signaling for up to 8 hphs . Alcian blue staining of cartilage in larvae at 96 hpf revealed that , compared with controls , hs-dkk1+/− heterozygotes heat shocked for 30 min at 16–28 hpf developed mandibular clefting and reduced Meckel's cartilages ( Mc ) , as well as mild reductions in other craniofacial cartilages ( Fig . 2A , B , D ) . Homozygous hs-dkk1+/+ larvae displayed dramatic shortening of Mc ventrally in the first arch , as well as the symplectic ( Sy ) , a more intermediate/dorsal element of the second arch ( Fig . 2C , E; [19] , [20] ) . Thus , hereafter “dkk1+” refers to homozygous hs-dkk1 embryos/larvae heat shocked for 30 min between 20–22 hpf . Heterozygous hs-dntcf3+ larvae heat shocked slightly later ( 22–24 hpf ) also showed mild reductions in Mc , but in this case Mc was fused to the more dorsal palatoquadrate ( Pq ) in arch 1 ( Fig . 2F , H ) . The ceratohyal ( Ch ) in ventral arch 2 was also variably reduced , but more posterior cartilages appeared largely unaffected . Cartilage defects in homozygous hs-dntcf3 larvae heat shocked similarly were much more severe ( Fig . 2E , G ) – 41% showed reduced Mc and Ch reduction/loss , while 66% showed joint fusion between Mc and Pq ( Fig . 2H ) . Thus , hereafter “dntcf3+” refers to heterozygous hs-dntcf3+/− embryos/larvae heat shocked for 12 min between 22–24 hpf . To determine tissue-specific requirements for Wnt signaling in the pharyngeal arches , we transplanted dntcf3+ cells at gastrula stages either into the fate map position that gives rise to NC or co-injected with Taram-A mRNA to drive them to an endodermal fate , into non-transgenic WT hosts [9] . While dntcf3+ NC cells in chimeras that virtually filled the entire mandibular arch caused no discernable cartilage defects ( not shown ) , large grafts of dntcf3+ endodermal cells into the pharyngeal region induced D-V patterning defects that resembled dntcf+ embryos , including reduced Mc and fused Mc-Pq ( Fig . S2A ) . These results suggest that the critical direct response to Wnt occurs in the endoderm ( which expresses 7XTCF:GFP ) and is indirectly relayed to surrounding NC cells . To verify that cartilage defects in dkk1+ and dntcf3+ larvae reflect specific requirements for Wnt signaling we attempted to rescue them using the compound 6-bromoindirubin-3′-oxime ( BIO ) , which stabilizes Wnt signaling by inhibiting GSK-3 [55] . BIO treatments of 7xTCF:GFP embryos at 24 hpf caused ectopic gfp expression and direct Wnt targets were upregulated in a dose-dependent manner as determined by ISH and qPCR analysis at 30 hpf ( Fig . S3 ) . Treatment of wild type embryos with BIO resulted in an overall reduction of cartilages in a dose-dependent manner , which correlated with reduced proliferating cell nuclear antigen ( pcna ) expression ( which marks cells in mitosis [56] ) in the arches at 30 hpf , indicating reduced proliferation of cartilage precursors ( Fig . S4J–L ) . Despite their smaller sizes , dorsal cartilages acquired more rod-like morphologies similar to ventral Mc and Ch , suggesting partial ventralization ( Fig . S4A–C ) . BIO treatments of both heat shocked dntcf3+ and dkk1+ transgenics partially rescued cartilage defects , including Mc-Pq joint fusions and Ch was consistently restored in dntcf3+ larvae ( Fig . S4D–I , M ) . BIO also rescued Mc clefting in dkk1+ larvae at higher concentrations ( Fig . S4G–I ) . Therefore loss of canonical Wnt signaling in dntcf3+ and dkk1+ embryos accounts for the majority of craniofacial defects . Cartilages in dntcf3+ and dkk1+ larvae were 30–50% smaller than controls ( Fig . 2 ) . This reduced cartilage size was not due to increased cell death as we could detect no differences in the number of acridine orange stained cells in the arches between dntcf3+ , dkk1+ and control embryos at 6 hphs ( Fig . S5 ) . To examine proliferation in the arches we performed ISH for pcna . Pcna mRNA was detected throughout the pharyngeal arches from 3–22 hphs ( 25–44 hpf ) in controls ( Fig . 3A–D ) , but somewhat reduced at 3 hphs ( 25 hpf ) in both dkk1+ and dntcf3+ embryos ( Fig . 3A , E , I , M ) . By 6 hphs ( 28 hpf ) pcna expression was nearly undetectable in the arches in both dkk1+ ( 66% , n = 6 ) and dntcf3+ ( 60% , n = 10 ) ( Fig . 3B , F , J , M ) . By 8 hphs ( 30 hpf ) , pcna expression had recovered slightly in dntcf3+ embryos ( 36% , n = 11 ) ( Fig . 3K , M ) but not in dkk1+ embryos ( 80% , n = 5 ) ( Fig . 3G , M ) . Both recovered completely by 22–28 hphs ( 44 hpf ) ( Fig . 3D , H , L , M ) . To confirm these apparent defects in proliferation we used an antibody that recognizes phosphoHistone3 ( pH 3 ) , a protein involved in chromosome condensation in mitotic cells [57] , which marks a subset of pcna+ dividing cells . In dkk1+ and dntcf3+ embryos pH 3 staining was reduced throughout the eye and brain ( Fig . S6A–B , D–E ) . At 4 hphs , dkk1+ embryos had a 75% reduction in pH 3+ cells in the arches compared to controls ( Fig . S6A′ , B′ , C ) . Similarly , at 3 hphs dntcf3+ embryos had approximately 50% fewer pH 3+ cells in the arches than controls ( Fig . S6D′ , E′ , F ) . Thus depleting Wnt signaling in embryos between 24–30 hpf severely impairs proliferation in the arches , which correlates with reductions in cartilage and in Wnt target gene expression ( Figs . 1 , 2 ) . To investigate roles for Wnt signaling in D-V patterning within the arches , we examined expression of genes that mark distinct ventral , intermediate and dorsal regions of the arch primordia in dntcf3+ and dkk1+ embryos with ISH [19] . hand2 expression in the ventral-most domains of each arch was severely reduced in both dkk1+ ( 53% , n = 15 ) and dntcf3+ ( 59% , n = 17 ) embryos ( Fig . 4A–C , Q ) , with a small domain of expression remaining at the arch 1–2 boundary in dntcf3+ embryos ( Fig . 4C ) . Similarly , expression of dlx3b and dlx6a in the intermediate domains of each arch were mildly reduced in dkk1+ ( 44% , n = 18 , 23% , n = 21 ) and severely reduced in dntcf3+ embryos ( 83% , n = 12; 90% , n = 21 ) ( Fig . 4D–I , Q ) . Finally , expression of the Notch ligand , jag1b in the dorsal-most domains of each arch [28] , was variably expanded ventrally in dkk1+ ( 10 . 5% , n = 57 ) embryos and consistently expanded in dntcf3+ embryos ( 55 . 5% , n = 9 ) as well as chimeras in which dntcf3+ cells were transplanted into the pharyngeal endoderm ( Fig . 4J–L , Q; Fig . S2B–C ) . These gene expression changes were not simply due to an overall loss of arches or NC cells , since dlx2a expression ( Fig . 4N , P ) as well as sox10:lynTdtom expression throughout the D-V extent of the arch NC were unaffected in the arches of both dkk1+ and dntcf3+ embryos ( Fig . S7 ) . Additionally , BIO treatments of wild type embryos caused dorsal expansion of expression of the ventral-intermediate gene msxe , mild expansion of dlx3b and hand2 expression , and reduced jag1b expression in the dorsal domain ( Fig . S8 ) . Therefore , Wnt signaling promotes ventral and intermediate-cell fates in the arches . Dntcf3+ embryos in particular , with residual hand2 expression at the arch 1–2 boundary , closely resemble mutants in Bmp and Edn1 signaling [19] , [20] . Because dntcf3+ embryos showed D-V defects in cartilage morphology and gene expression that more closely resembled Bmp- and Edn1-deficient embryos than dkk1+ we focused on dntcf3+ . To examine interactions between Wnt and Bmp signaling in the arches we used an antibody that recognizes phosphorylated Smad1/5/8 ( pSmad1/5/8 ) in dntcf3+ embryos . In controls pSmad1/5/8 localized to ventral arches 1 and 2 where levels of Bmp signaling have been shown to be highest at 24 hpf ( Fig . 5A–C; [19] ) . Anti-pSmad1/5/8 staining was slightly reduced in the first arch at 2 hphs ( 24–26 hpf ) in dntcf3+ embryos ( Fig . 5D ) , in both arches by 4 hphs ( Fig . 5E ) , and virtually lost altogether at 6 hphs ( Fig . 5F ) . Western blots confirmed that pSmad1/5/8 levels were much lower than controls at 6 hphs ( Fig . 5G , H ) . To examine potential interactions between Wnt and Edn1 signaling in the arches we performed qPCR for Edn1 in dntcf3+ embryos at 6 hphs . Edn1 expression was significantly reduced relative to control ( Fig . 5I ) . These results reveal an indirect role for Wnts in D-V patterning through regulation of both Bmp and Edn1 signaling . Bmps act together with Edn1 to promote ventral-intermediate cell fates in the arches [16] , [17] , [21] , [22] , [58] . Therefore we examined the ability of Bmp and Edn1 to restore ventral-intermediate gene expression in Wnt signaling-deficient embryos . Beads coated in human recombinant BMP4/7 heterodimers effectively induce Bmp target genes in zebrafish pharyngeal arches [19] . Similarly , microinjection of a 25 ng/nl BMP4/7 solution extracellularly on one side of the head induced Bmp signaling , as measured by expression of the transgenic Bmp-response element reporter ( Bre:Gfp; [19] ) at 8 hours post injection ( hpi ) ( Fig . 6A , B ) . Unilateral injections of BMP4/7 protein into dntcf3+ embryos at 4 hphs partially rescued cartilage defects on the injected side ( Fig . 6C–J ) . Typically this restored Mc length and Ch , but not the Mc-Pq joint , and rescue was dose-dependent ( Fig . 6C , D ) . These results suggest that Wnt signaling acts upstream of , or possibly in parallel to , Bmp signaling to promote ventral cartilage cell fates in the arches . EDN1 protein injections have previously been shown to rescue an Edn1 mutant phenotype and partially rescue a Bmp loss of function phenotype [16] , [19] . EDN1 injections into dntcf3+ embryos also partially rescued Mc length , but notably were more proficient at rescuing Ch and joint development ( Fig . 6C–D , L ) . While both Bmp and Edn1 signaling induce many of the same genes that specify ventral-intermediate NC cell fates in the early arches , by later stages Bmps become much stronger inducers of hand2 ( ventral ) and msxe ( ventral-intermediate ) [19] , [20] . Strikingly , neither BMP4/7 nor EDN1 protein injections at 4 hphs were sufficient to rescue hand2 expression in dntcf3+ embryos ( Fig . 7A–D , Q ) . BMP4/7 but not EDN1 restored msxe expression ( Fig . 7E–H ) , particularly in the mandibular arch near the injection site ( Fig . 7G ) . In contrast , both BMP4/7 and EDN1 injections restored dlx3b and dlx5a expression in the intermediate domain ( Fig . 7I–P , Q ) . These results suggest that hand2 expression absolutely requires Wnt signaling to respond to Bmps , while other signals can partially substitute for Wnts in induction of more intermediate-dorsal NC cell fates . To further investigate how Wnts might regulate the ability of NC cells to respond to Bmp signaling , we examined whether or not dntcf3+ embryos show any changes in expression of Bmp receptors . Zebrafish have four type 1 receptors ( Bmpr1aa , ab , ba , bb ) and two type II receptors ( Bmpr2a and b ) . Whole mount ISH for all six receptors revealed that only bmpr1ab , bmpr1ba , and bmpr1bb are expressed strongly in the arches at 24 hpf ( Fig . S9C–F , I–J , M–R ) . bmpr1aa , bmpr2a , and bmpr2b were detected much more broadly throughout the embryo at this stage ( Fig . S9A–B , G–H , K–L ) . Bmpr1ab expression extended throughout arches 1 and 2 , while bmpr1ba and bmpr1bb expression was restricted to more intermediate and ventral domains ( Fig . S9M–O ) . Transverse sections additionally showed that bmpr1ab and bmpr1ba expression is limited to arch NC cells and not surrounding epithelia ( Fig . S9P–R ) . In dntcf3+ embryos bmpr1ab was severely reduced ( 57% n = 35 ) ( Fig . 8A–B , G ) , while bmpr1ba was slightly reduced ( bmpr1ab: 46% n = 57 ) and bmpr1bb expression was largely unaffected ( bmpr1bb: 31% n = 29 ) ( Fig . 8C–F , G ) . Changes in Bmp receptor expression in dntcf3+ embryos were further quantified by qPCR analysis . At 6 hphs we compared the relative expression of arch specific Bmp receptors ( bmpr1ab , bmpr1ba , bmpr1bb ) with ubiquitously expressed bmpr2a and tgfbr1a , a TGF-B receptor expressed in the arches unrelated to Bmp signaling [59] . There was no detectable reduction in tgfbr1a or bmpr2a expression in dntcf3+ embryos ( Fig . 8H ) . At 6 hphs , both bmpr1ab and bmpr1ba expression were reduced ( Fig . 8I ) but bmpr1bb expression showed no difference from controls ( Fig . 8H ) . A time series analysis revealed no change in bmpr1ab and bmpr1ba expression at 2 hphs , despite reduced Wnt signaling ( see Fig . 1 ) , but levels dropped dramatically by 4 hphs . bmpr1ba but not bmpr1ab expression recovered substantially by 8 hphs . This suggests differential requirements for Wnt signaling in induction of Bmp receptors . dkk1+ embryos exhibit a unique clefting of the mandible not seen with dntcf3+ . Although primarily known as a repressor of Wnt signaling , Dkk1 has also been reported to positively regulate the Wnt-PCP pathway [60] . To gain further insights into its tissue-specific functions , we examined dkk1b expression in pharyngeal arch primordia . Of the five known dkk genes in zebrafish , only dkk1b is expressed in the embryonic arches [61] . We found that between 28–48 hpf dkk1b expression localized to the pharyngeal endoderm , particularly the pouches between arches ( Fig . S10A–C ) . Consistent with this , expression was lost in van gogh ( vgo ) mutants , which lack pouches [10] ( Fig . S10D–E ) . dkk1b expression was also detected in the stomodeum ( oral ectoderm ) at 28 hpf ( Fig . S10A ) and later in the ectoderm of the mouth at 48 hpf ( Fig . S10F ) . Signals from the pharyngeal endoderm and oral ectoderm are necessary for craniofacial patterning and chondrogenesis [9] , [10] , [13]–[15] . To determine if there were gross defects in these epithelial layers in dntcf embryos , we examined nkx2 . 3 , and found that its expression in the pharyngeal endoderm was disorganized in the first two pouches ( Fig . S11B , F ) and severely reduced in the more posterior pouches ( Fig . S11B ) . In contrast , anterior pouches appeared unaffected in dkk1+ embryos , while the more posterior pouches were occasionally disorganized ( Fig . S11D , H ) . Expression of pitx2ca in the oral ectoderm ( Fig . S11I–R ) was delayed in dkk1+ embryos until 26 hphs ( Fig . S11O ) , by 30 hphs the mouth opening was abnormally elongated laterally and by 51 hphs showed a ventral midline fold ( Fig . S11P–R ) . Thus pharyngeal pouch and mouth defects differ between dntcf and dkk1 embryos , which could account for some of the differences in their effects on growth and morphogenesis of the lower jaw .
Direct Wnt responses in the ventral first and second arches resemble the pattern of TOP:Gal expression in mice , including distal ( ventral ) arch 1 [36] , [37] . Mycn , a direct transcriptional Wnt target , is also expressed in both fish and mouse arch NC cells , where it is likely to inhibit Wnt-β-catenin signaling [62] , and provide negative feedback . Murine Mycn is expressed in highly proliferative cells and mutants show hypoplasia of the mandibular arch [63] , [64] . Similarly , we find reduced proliferation in the pharyngeal arches in Wnt-deficient zebrafish and smaller craniofacial cartilages ( Fig . 3; Fig . S5 ) . Thus , Wnt signaling may promote growth of the ventral arches through induction of mycn expression . We show a critical requirement for Wnt signaling in arch growth and patterning that is distinct from its earlier roles in NC induction and migration . Earlier heat shocks of dntcf3 or dkk1 zebrafish ( 10–20 hpf ) disrupt premigratory NC formation [41] , [65] similar to Wnt1/Wnt3a mutant mice [43] . Unlike recent conditional loss- and gain-of-function studies of βcat in the pharyngeal ectoderm in mice [44] , we find no changes in cell survival in the arches in dntcf3+ embryos . D-V defects in gene expression in the arches caused by overexpressing dntcf3 or dkk1 at these later stages point to a problem with canonical Wnt signaling . Both reduce expression of canonical Wnt target genes up to 8 hphs ( Fig . 1I; Fig . S1 ) , and both lead to ventral cartilage and joint defects . However , dkk1 overexpression has more subtle effects ( restricted primarily to Mc and the jaw joint ) than dntcf3 . Defects in dkk1+ embryos are also stronger when heat shocked at slightly earlier stages ( 15–22 hpf ) , than dntcf3+ ( 22–24 hpf ) . These differences may reflect distinct functions for the two transgenes , or a delay due to the time required for Dkk1 to competitively bind with the Lrp5/6 co-receptor , whereas Tcf3 directly binds βcat and Wnt target genes . Heat shocking dntcf3 or dkk1 at earlier stages eliminates cartilage , consistent with requirements for canonical Wnt signaling in NC induction . Both mycn and axin2 expression recover by 24 hphs in heat shocked embryos indicating a transient requirement for canonical Wnt signaling prior to skeletal cell differentiation . Bmps and Edn1 secreted by the pharyngeal ectoderm both promote ventral-intermediate skeletal cell fates in the arches [19] , [20] and our results implicate Wnt as an additional ventralizing factor . Overexpression of dntcf3 leads to reduced hand2 , dlx3b , and dlx5a ventrally and expansion of dorsal jag1b expression ( Fig . 4 ) , and similar but less severe changes in gene expression result from dkk1 overexpression . Loss of the positive Wnt regulator , R-spondin , in mice also disrupts expression of Hand2 , Dlx5 , Dlx6 , and Msxe [66] , suggesting a conserved requirement for Wnt signaling in promoting ventral-intermediate cell fates . How are these different ventralizing signals integrated during D-V arch patterning ? Wnts can either activate or inhibit Bmp signaling in different developmental contexts [67]–[71] . pSmad1/5/8 expression is reduced in the pharyngeal arches of dntcf3+ embryos ( Fig . 5 ) , suggesting a novel role for Wnts upstream of Bmp signaling during arch development . Consistent with this model , microinjection of Bmp4/7 protein directly into the arch primordia at 4 hphs rescues ventral cartilages ( Mc , Ch ) in dntcf3+ embryos , but not joint fusions ( Fig . 6C , F ) . Similar injections of BMP4/7 protein into edn1−/− mutant zebrafish rescues ventral cartilages but not joint fusions , while ectopic Edn1 can rescue joint defects caused by a loss of Bmp [19] . We show that injection of EDN1 protein rescues joint fusions in Wnt deficient embryos ( Fig . 6D , L ) . Wnts also induce Edn1 expression in the pharyngeal ectoderm [66] . Taken together , these results suggest that Wnt signaling influences ventral cell fates in the arches through both Bmps and Edn1 , and joint patterning specifically through Edn1 . Another clue to specificity in the D-V arch patterning system comes from the fact that msxe expression ( a direct BMP target and marker of more intermediate identities ) is induced by Bmp4/7 in edn1−/− mutants [20] , and by Edn1 protein in the absence of Bmp signaling [19] . Surprisingly , however , msxe expression in dntcf3+ embryos is only rescued by BMP4/7 , and not EDN1 , while dlx3b and dlx5a expression is rescued in both . This suggests that Edn1 can only induce msxe expression in the arches in the presence of Wnt signaling . Thus Wnt controls the competence for arch cells to respond to Edn1 in addition to inducing expression of Edn1 itself . Mice mutant in the essential Wnt receptor co-factor , Lrp6 , also lack expression of Msx1 and Msx2 , in the arches [72] , possibly as a result of defects in Bmp signaling . Neither BMP4/7 nor EDN1 overexpression rescues hand2 expression in dntcf3+ embryos , revealing a critical requirement for Wnt in induction of hand2 [17] , [23] , [73]–[75] . Both Bmp and Edn1 induce hand2 expression and specify the ventral arch domain initially , but later Bmps maintain hand2 in the ventral domain while Edn1 promotes expression of more ventral-intermediate genes . We pinpoint a critical period for Wnt signaling in D-V patterning between 24–30 hpf , when hand2 expression is unresponsive to Edn1 . Our results suggest that Bmp signaling requires Wnt signaling to induce hand2 expression in the arches . Consistent with this model , Wnt signaling directly regulates Hand2 transcription in chondrocytes [76] . Failure of hand2 induction is not simply due to loss of cells , since ventral NC cells are still present in the arches of dntcf3+ embryos ( Fig . 4N–O; Fig . S7 ) . BMP protein can also induce hand2 throughout the D-V extent of the arch [20] . Our results suggest that Wnt signaling plays a critical role in regulating the competence of cells to respond to BMPs and to express hand2 . We propose that Wnt signaling activates a signal ( factor X ) from the pharyngeal endoderm that primes the ability of NC cells to respond to Bmp signaling , in part through the transcriptional regulation of Bmp receptors ( Fig . 9 ) . Similarly in D-V patterning of the mouse limb Wnt signaling is thought to act upstream of Bmpr1a [71] . Three type I Bmp receptors , bmpr1ab , bmpr1ba , bmpr1bb , have arch-specific expression in zebrafish , similar to mice [77] . These are expressed in nested patterns within the arches: bmpr1ab throughout and bmpr1ba/bb only in intermediate-ventral domains . Thus Bmpr1 receptors may have distinct roles in different spatial domains ( in addition to their cell-type specific roles [78] , [79] ) , but this has been difficult to test due to early lethality in traditional Bmpr knockouts [80] , [81] . We show that overexpression of dntcf3+ inhibits bmpr1ab and bmpr1ba , but not bmpr1bb and bmpr2a , expression in the arches . This reduction in Bmpr expression occurs later than most direct Wnt targets ( Fig . 8I; 4–8 hphs ) suggesting that it is indirect , consistent with our model that Wnt activates an unknown signal from the endoderm important in this process . bmpr1ba expression also recovers by 8 hphs , before bmpr1ab expression and within the period during which Wnt signaling is significantly reduced ( Fig . 1I ) indicating that bmpr1ab is particularly sensitive . This could help explain the inability of Bmp protein to rescue hand2 expression in dntcf3+ embryos if bmpr1ab plays a specific role in hand2 induction . In contrast , intermediate-ventral genes such as msxe and dlx3b , may be rescued by Bmp protein because other Bmp receptors are sufficient for their induction . Such distinct transcriptional roles for Bmp receptors could help fine-tune D-V domains within an arch despite the relatively broad expression of Bmp ligands . Both dntcf3+ and dkk1+ embryos appear to function in the pharyngeal endoderm , which is an important signaling center in craniofacial development [7] , [9] . Our chimeric analyses demonstrate a cell autonomous requirement for dntcf3 in endoderm ( Fig . S2 ) and dkk1 expression is restricted to this epithelium . Interestingly , we do not detect dkk1 expression in the first pharyngeal pouch , which lies between arches 1 and 2 ( Fig . S10 ) . This could help explain why Wnt signaling only appears to be required for D-V patterning in these arches; the more posterior ceratobranchials ( arches 3–7 ) are largely unaffected in heterozygous dkk1+ and dntcf3+ embryos ( Fig . 2B , F ) . These results suggest a previously unrecognized role for Wnts and Wnt antagonists in endoderm and the existence of an as yet unknown factor X produced by endoderm that is important in D-V patterning of the NC ( see Fig . 9 ) . The distinct and highly penetrant clefting of the mandible observed in dkk1+ embryos is never observed in dntcf3+ embryos . Disruption of the canonical Wnt pathway can cause clefting of the palate in mice [36] , [46] , [67] , but such midline clefts in the lower jaw are rare . Midline facial defects , particularly of the frontonasal process , have been reported in Wnt signaling mutants in mice [36] , [66] , [67] . Mice mutant for Dlx5/6 have cleft mandibles and Wnt9b mutants have cleft lip [82] , [83] . Humans with Richieri-Costa-Perieira syndrome also exhibit clefting of the lower jaw similar to what we describe in dkk1+ embryos [68] , [84] . Dkk1 not only inhibits canonical Wnt signaling [52] but can also activate the non-canonical Wnt-PCP pathway during zebrafish gastrulation [60] . Non-canonical Wnt signaling has also been implicated in craniofacial midline development as Wnt5a mutant mice have clefting of the secondary palate [66] . Thus , overexpression of Dkk1 may lead to both a canonical Wnt/βcat loss-of-function and a non-canonical Wnt-PCP gain-of-function to cause lower jaw clefting . Overexpression of dkk1 also leads to elongation and ventral clefting of the mouth , which is not observed in dntcf3+ embryos . Both loss- and gain-of-function mutations in mammalian Dkk1 result in midline clefts in the frontonasal and maxillary prominences [36] . NC cells fated to form the lower jaw lie adjacent to the oral ectoderm ( stomodeum ) , which secretes important skeletogenic signals such as Shh [14] , [15] and Bmps [19] . dkk1b transcripts are normally restricted to the anterior ectoderm of the mouth opening ( Fig . S10F ) where both fgf8 ( distal ) and shh ( medial ) are expressed . Thus , misexpressing dkk1b throughout the oral ectoderm may disrupt one of these other signals . Future experiments are needed to determine the roles of Wnt signaling in mouth development and the causes of mandibular clefts .
All zebrafish work was performed using protocols approved by the University of California , Irvine Institutional Animal Care and Use Committee ( Protocol # 2000-2149-4 ) . Adult Tg ( hsp701:dkk1-GFP ) ( dkk1 ) [50] and Tg ( hsp70I:tcf3-GFP ) ( dntcf3 ) [41] fish were genotyped by performing PCR analysis for gfp ( sense , GTGATGCAACATACGGAAAAC; antisense , GCCATGTGTAATCCCAGCAGC ) using genomic DNA extracted from fin clips as template . Zygosity of fish was determined by outcrossing genotyped transgenic adults with wild type adults and scoring the Mendelian ratio of GFP positive to GFP negative after heat shocking as described below . To observe NC , we outcrossed adult homozygous Tg ( hsp70I:tcf3-GFP ) fish with sox10;LynTdtomato fish to make a stable double transgenic line . Adult heterozygous Tg ( 7xTCF-Xla . Siam:GFP ) ia4 ( 7xTCF;GFP ) transgenics [49] were in-crossed and sorted by GFP expression for downstream phenotypic analysis . Tg ( BRE:gfp ) ( bre:gfp ) [19] heterozygous adults were in-crossed , selected for strong GFP expression , and used for protein injection ( see below ) . Heat shocks were performed in a thermal cycler at 39°C for either 12 min ( dntcf3 ) or 30 min ( dkk1 ) . Fluorescence was checked 1-hour post heat shock and GFP-negative embryos were separated and used as controls . Embryos were then raised in a 28 . 8°C incubator until they were fixed for in situ analysis , RNA extraction , or protein extraction at various time points after heat shock or 4 dpf for skeletal staining . Alcian blue staining of cartilage was performed on 96 hpf embryos as previously described [19] . In situ hybridization was performed on embryos fixed in 4% PFA for one hour at room temperature . Probes used include gfp [19] , dlx2a , dlx3b , dlx6a [85] , mycn [86] , hand2 [87] , jag1b [28] , msxe , dkk1b , nkx2 . 3 [17] , ednrA1 , ednrA2 ( Nair et al , 2007 ) , and pitx2ca [88] . The axin2a probe was synthesized directly from a PCR product with a T7 promoter site added at the 3′ end ( sense , AGAAGATGACCCACGTCCAC; antisense , TAATACGACTCACTATAGGGGACTGTGACCTTGTGCTGAGAC ) . The Bmp receptor probes were synthesized directly from PCR products with a T7 promoter site added at the 3′ end: Bmpr1aa ( sense , TAGCCAACCCCAATGCTTAC; antisense , TAATACGACTCACTATAGGGCCCATTTGTCTCGCAGGTAT ) , Bmpr1ab ( sense , GATGCCACAAACAACACCTG; antisense , TAATACGACTCACTATAGGGACTTTCACCGCCACATTTTC ) , Bmpr1ba ( sense , AGAATCTCTGCGGGATCTCA; antisense , TAATACGACTCACTATAGGGGCTCCGTTTCTCTTGACCAG ) , Bmpr1bb ( sense , TCACGGATTATCACGAGAGCG; antisense , TAATACGACTCACTATAGGGATTATGAGCCCAGCACTCGC ) , Bmpr2a ( sense , CCACAATGACACCTCAGTGG; antisense , TAATACGACTCACTATAGGGTTAGGGACGTTCTGCTGCTT ) , Bmpr2b ( sense , TATTGTCGCGCTGTTCTTTG; antisense , TAATACGACTCACTATAGGGGCAGATAGGCCAGTCCTCTG ) . A pcna probe was generated by a T7 promoter from a clone of the ORF ( Open Biosystems , clone ID:7000501 ) in pExpress1 after linearization with EcoRI . Immunolabeling was performed using 1∶500 rabbit anti-phosphoHistodone3 ( Upstate Biotechnology ) , 1∶1000 rabbit anti-pSmad1/5/8 ( Millipore ) , or 1∶500 chick anti-gfp ( AbCam ) antibodies diluted in 1% DMSO , 0 . 5% Triton ×100 in PBS and detected using 1∶1000 dylight donkey anti-rabbit 564 ( Jackson ImmunoResearch Laboratories ) or 1∶1000 donkey anti-chick dylight 488 ( Jackson ImmunoResearch Laboratories ) . Tg ( hsp70I:tcf3-GFP ) embryos were heat shocked as described , anesthetized , and then embedded in 1% low melt agarose in embryo medium . Human EDN1 ( Sigma-Aldrich ) was diluted to 10 µg/µl and human recombinant BMP4/7 ( R&D Systems ) was diluted to either 50 ng/nl or 10 ng/nl . A 0 . 5 nl droplet of protein solution was pressure injected into the arch region 4 hours post heat shock ( ∼26 hpf ) using a glass needle . Embryos were then carefully removed from the agarose using forceps and fixed for phenotypic analysis 4 hours later ( ∼30 hpf ) or 4 dpf for alcian stain using 4% PFA at room temperature for 1 hour . Dechorionated control , dkk1+ , and dntcf3+ embryos were placed in dishes containing 50 µm or 100 µm of BIO ( 30 mm stock in DMSO ) ( Sigma ) diluted in embryo medium ( EM ) at 2 hours post heat shock [55] . The dishes were placed in a 28 . 8° incubator for 6 hours and then washed several times with EM . Embryos were fixed for in situ hybridization , harvested for RNA , or allowed to develop to 4 dpf and then fixed for alcian blue staining . Protein was extracted from dechorionated embryos by adding 2 µl/embryo of sample buffer ( 60 mM Tris-HCl pH 6 . 8 , 2% SDS , 10% glycerol , 5% β-mercaptoethanol , 0 . 01% bromophenol blue ) and homogenizing with a pestle . The sample was boiled in a 95°C heat block for 10 min and then immediately placed on ice . Before loading into a 10% SDS gel the sample was spun down at 1300 rpm for 5 min . The membrane was blocked in 3% BSA and 3% Donkey Serum in TBST ( 1XTBS −20 mM Tris-HCl , 150 mM NaCl , 0 . 1% Tween ) for one hour at room temperature and incubated overnight at 4°C with 1∶1000 rabbit anti-pSmad1/5/8 ( Cell Signaling ) . The next day the membrane was washed several times in TBST and then incubated with 1∶5000 donkey anti-rabbit HRP ( GeneTex ) for 1 hour at room temperature . Total RNA was extracted from control , dkk1+ , and dntcf3+ embryos at various time points past heat shock using Trizol reagent ( Ambion ) . cDNA synthesis was performed with 1 µg of RNA using Protoscript M-MuLV First Strand cDNA synthesis kit ( New England Biosystems ) . qPCR was performed using Light Cycler 480 SYBR Green Master ( Roche Applied Science ) in a Light Cycler 480 Real Time PCR machine . Q-PCR primer sets used were for mycn ( Sense-AACAAGAGGGAGAATGCCA; Antisense-TAGAAGTCATCCTCGTCCG ) , axin2 ( Sense- CAATGGACGAAAGGAAAGATCC; Antisense-AGAAGTACGTGACTACCGTC ) , lef1 ( Sense-CCAGACATTCCCAATTTCTATCC; Antisense-GTGATGTGAGAACCAACCC ) , ef1alpha ( Sense- CAAGGGATGGAAGATTGAGC; Antisense- AACCATACCAGGCTTGAGGA ) , bmpr1ab ( Sense- CATGAGGGAAGTGGTATGTG; Antisense- ATGACTCGTAAGCACTCGT ) , and bmpr1ba ( Sense- GACAATATACTGGGATTTATAGCGG; Antisense – ATGATAGTCTGTGATCAGGTAGAG ) , bmpr1bb ( Sense-AACATACTGGGCTTCATCG; Antisense- CTCGTGATAATCCGTGATCAG ) , bmpr2a ( Sense-TTTCCCAGGTGAAACAGTG; Antisense-TGCATGTCCTCTATGGTAGG ) , tgfbr1a ( Sense-GCATGATCAAGCTGTCTCTG; Antisense-CAGGCTTACCCTGAGTACC ) , and edn1 ( Sense-TATGGGTGAACACACCAGAGCGAA; Antisense-CGCTTGGCAGAATGAAGAGCATGT ) . Tg ( hsp70I:tcf3-GFP ) donor embryos were injected at the 1-cell stage with 15 pg Taram-A ( Tar* ) mRNA , which drives cells to an endodermal fate , combined with a 1∶1 mixture of 6% biotin-dextran and 6% rhodamine-dextran . Cells were transplanted into WT hosts at the 30–50% epiboly stage ( 5 hpf ) to generate chimeras , as described previously [9] . Host embryos with red fluorescent cells in the pharyngeal endoderm were sorted at 22 hpf , heat-shocked and raised to either 30 hpf for ISH or 5 dpf for skeletal staining . Grafts were labeled at either time-point using a peroxidase-coupled streptavidin and diaminobenzidine . | Craniofacial abnormalities are among the most common birth defects . Understanding the molecular mechanisms underlying craniofacial disorders is crucial for developing treatment strategies . Much of the craniofacial skeleton arises from specialized embryonic structures known as pharyngeal arches . Patterning of these arches requires precise spatial and temporal expression of multiple genes , which is coordinated between tissues by secreted signals . Wnts are secreted ligands expressed throughout the pharyngeal arches yet their role in craniofacial patterning remains unclear . In this study we examine the role of Wnts in craniofacial patterning using transgenic zebrafish to inhibit downstream Wnt signaling . We show that Wnt signaling deficient embryos have lower jaw specific defects , which strongly resembles loss-of-function phenotypes in both the Bmp and Edn1 signaling pathways . Through rescue experiments we find that Wnts are upstream regulators of both Bmp and Edn1 signaling . We thus have uncovered a crucial requirement for Wnt signaling in craniofacial patterning . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"developmental",
"biology",
"model",
"organisms",
"genetics",
"biology",
"and",
"life",
"sciences",
"research",
"and",
"analysis",
"methods"
] | 2014 | Wnt Signaling Interacts with Bmp and Edn1 to Regulate Dorsal-Ventral Patterning and Growth of the Craniofacial Skeleton |
Phylogenies of highly genetically variable viruses such as HIV-1 are potentially informative of epidemiological dynamics . Several studies have demonstrated the presence of clusters of highly related HIV-1 sequences , particularly among recently HIV-infected individuals , which have been used to argue for a high transmission rate during acute infection . Using a large set of HIV-1 subtype B pol sequences collected from men who have sex with men , we demonstrate that virus from recent infections tend to be phylogenetically clustered at a greater rate than virus from patients with chronic infection ( ‘excess clustering’ ) and also tend to cluster with other recent HIV infections rather than chronic , established infections ( ‘excess co-clustering’ ) , consistent with previous reports . To determine the role that a higher infectivity during acute infection may play in excess clustering and co-clustering , we developed a simple model of HIV infection that incorporates an early period of intensified transmission , and explicitly considers the dynamics of phylogenetic clusters alongside the dynamics of acute and chronic infected cases . We explored the potential for clustering statistics to be used for inference of acute stage transmission rates and found that no single statistic explains very much variance in parameters controlling acute stage transmission rates . We demonstrate that high transmission rates during the acute stage is not the main cause of excess clustering of virus from patients with early/acute infection compared to chronic infection , which may simply reflect the shorter time since transmission in acute infection . Higher transmission during acute infection can result in excess co-clustering of sequences , while the extent of clustering observed is most sensitive to the fraction of infections sampled .
Phylogenetic clusters of closely related virus such as HIV arise from the epidemiological dynamics and transmission by infected hosts . If virus is phylogenetically clustered , it is an indication that the hosts are connected by a short chain of transmissions [1] . If super-infection is rare , and assuming an extreme bottleneck at the point of transmission , each lineage in a phylogenetic tree corresponds to a single infected individual with its own unique viral population [2] , [3] . A transmission event between hosts causes an extreme bottleneck in the population of virus in the new hosts . For infections between MSM , it is estimated that infection is initiated by one or several virions [4] , [5] . At the time of transmission , the quasispecies of virus within the transmitting host diverges and can thereby generate a new branch in the phylogeny of consensus viral isolates from infected individuals [6] . Transmissions in the recent past should be reflected by recently diverged lineages , and transmissions from long ago should reflect branches close to the root of a tree . [7] . Viruses such as HIV which have a high mutation rate relative to epidemiological spread can generate epidemics such that the correspondence between transmission and phylogenetic branching is most clear [2] . Given a phylogeny of virus reconstructed from samples , the phylogenetic clusters are a partition of the sample units into disjoint sets as a function of the tree topology . A cluster will consist of all taxa of the tree that are descended from a given lineage on the interior of the tree . There are many variations of this idea , and there is no general agreement about how to choose interior lineages for defining clusters . The most common algorithms require strong statistical support for a monophyletic clade among all taxa in a cluster [8]–[14] . These definitions may additionally require all taxa in a cluster to be connected by short branches with less than a threshold length [11] , or similarly require that the genetic sequences corresponding to each taxon be separated by a genetic distance less than a given threshold [8] , [14] . Definitions of clustering based on statistical support for monophyly are very difficult to operationalize in a mathematical model , and in particular , it is not clear how the statistical significance of internal nodes relates to population dynamics . Consequently , we have devised a conceptually similar definition of clusters that relies on the estimated time to most recent common ancestor ( TMRCA ) of a set of taxa [15] . A formal definition is provided below . The sizes of the groupings that arise from a clustering algorithm have been interpreted as a reflection of the heterogeneity of epidemiological transmission . The distribution of cluster sizes of HIV is often skewed right , and depending on the definition of clustering used , can have a heavy tail [14] , [15] . This is consistent with the prevailing view among modelers of sexually transmitted infections that there is a skewed and in some cases power-law distribution in the number of risky sexual contacts in the population , however it is not straightforward to make inferences about sexual network properties from cluster size distributions [16] . In the case of HIV , the distribution of branch lengths within clusters may also reflect the disproportionate impact of early and acute HIV infection on forward transmission , which is due to higher viral loads in the early stages of infection , higher transmissibility per act [17] , and fluctuating risk behavior [18] . When the taxa of the phylogeny are labeled , such as with the demographic , behavioral or clinical attributes of the the individuals from whom the virus was sampled , one can further analyze statistical properties of clustered taxa . Similar taxa , such as those arising from acute infections , may cluster together ( or co-cluster ) at greater rates . Patterns of co-clustering might be informative about the fraction of transmissions that occur at different stages of infection or between different demographic categories . HIV phylogenies from men who have sex with men ( MSM ) have been widely observed [12] , [13] , [19] to have individuals with early/acute HIV infection that are much more likely to appear in a phylogenetic cluster . And moreover , if early-stage individuals are in a cluster , they are much more likely to be clustered with other early infections . Both Lewis et al . and Brenner et al . [8] , [9] have hypothesized that co-clustering of early infection is caused by higher transmissibility per act during early infection . For example , in phylogenies with time-scaled branch lengths , if a large fraction of clusters have a maximum branch length of six months [8] , [15] , this suggests that at least that fraction of transmissions also occur within six months . In this article we demonstrate that the mechanisms that generate co-clustering of early infections are complex , and involve many attributes of the epidemic in addition to higher transmissibility per act [17] . To summarize , several features of the phylogenetic structure of HIV in MSM have been independently observed by several investigators: Below , we illustrate these clustering patterns using 1235 HIV-1 subtype B pol sequences collected between 2004 and 2010 in Detroit , Michigan , USA . These common clustering features motivate several questions . How informative are clustering patters about the underlying epidemic ? In particular , how does higher transmissibility per act during early infection shape the phylogeny of virus ? To address these questions , we have developed a simple mathematical framework that demonstrates the connection between epidemiological dynamics and the expected patterns of clustering from a transmission tree and the corresponding phylogeny . Our modeling work suggests that common features of HIV phylogenies are not coincidences , but universal features of certain viral phylogenies . We expect to see similar patterns for any disease such that the natural history features an early period of intensified transmission . High transmission rates during early infection may be a consequence of higher transmissibility per act due to high viral loads , but are also influenced by behavioral factors , such as fluctuating risk behavior [18] , concurrency [20] , and a lack of awareness of the infection . We do not explicitly model immunological or behavioral factors , but rather consider a compound parameter that describes the rate of transmission during the early/acute period . We find that while higher transmission rates increase the frequency of early/acute clustering , virus collected from early/acute patients clusters at a higher rate even when transmission rates are uniform over the infectious period .
This research was reviewed by the Institutional Review Board at the University of Michigan . Data used in this research was originally collected for HIV surveillance purposes . Data were anonymized by staff at the Michigan Department of Community Health before being provided to investigators . Because this research falls under the original mandate for HIV surveillance , it was not classified as human subjects research . Our analysis consists of an empirical component which establishes clustering patterns for a geographically and temporally delineated set of HIV sequences , and an analytical component which establishes a possible mechanism that could generate the observed patterns . We examined the phylogenetic relationships of 1235 HIV-1 subtype B partial-pol sequences originally collected for drug-resistance testing . All sequences were collected in the Detroit metropolitan statistical area between 2004 and 2010 . Sequences were tested for quality and subtype using the LANL quality control tool [21]–[23] , and aligned against a subtype-B reference ( HXB2 ) . Drug resistance sites [24] were treated as missing data . A maximum clade credibility phylogeny was estimated with BEAST 1 . 6 . 2 [25] . The phylogeny was estimated using a relaxed molecular clock and and HKY85 model of nucleotide substitution with Gamma rate variation between sites ( 4 categories ) . The MCMC was run for 50 million iterations with sampling every iterations . The first million iterations were discarded . The effective sample size of all parameters exceeded 50 . The phylogeny was converted into a matrix of pairwise distances between taxa expressed in units of calendar time . The distance between a pair of taxa was the TMRCA estimated by BEAST . Taxa were then classified into clusters using hierarchical clustering algorithms . A pair of taxa were considered to be clustered if the estimated TMRCA did not exceed a given threshold , and a range of thresholds was examined , from 0 . 5% of the maximum distance to the distance corresponding to the point where 90% of taxa are clustered with at least one other taxon . Co-clustering of early/acute infections was investigated using a clinical variable ( CD4 count ) and a measure of genetic diversity of the virus . Both CD4 and sequence diversity are imprecise indicators of stage of infection . Nevertheless , with a large population-based sample , even noisy indicators of stage of infection are useful for illustrating phylodynamic patterns . In most cases , CD4 counts were assessed contemporaneously with samples collected for sequencing . The CD4 cell counts can be informative about disease progression and can be used as a noisy predictor of the unknown date of infection [26] . Individuals with very high cell counts are unlikely to represent late/chronic infections , and we hypothesize that virus from these patients will be more likely to be phylogenetically clustered . Clustering of patients with high CD4 was previously observed by Pao et al . [10] Recent work [27] has also highlighted the potential for sequence diversity to be informative of the date of infection . The frequency of ambiguous sites ( FAS ) in consensus sequences provides an approximate measure of sequence diversity in the host . HIV infection is initiated by one or a few founder lineages [4] , [5]; initially the diversity of the viral population within the host is low , but diversity increases steadily over the course of infection [28] . By convention , consensus sequences report ambiguous sites as those where the most frequent nucleotide is read with a frequency less than 80% . We hypothesize that having few ambiguous sites is an indicator of early/acute infection; sequences with fewer ambiguous sites will be more likely to be in a phylogenetic cluster and to be clustered with other sequences with few ambiguous sites . A simple analysis was conducted to establish the existence of excess clustering and co-clustering in the Michigan sequences . This analysis is not designed to classify our sample into a early/acute component or to estimate the date of infection for each unit . To illustrate excess clustering of early/acute infections , we calculated the mean CD4 cell count and FAS for each sample unit in a phylogenetic cluster . Because all clustering thresholds are arbitrary , we explored a large range of values , up to the point where 90% of the sample was clustered with at least one other unit . The standard error of the estimated mean was calculated assuming simple random sampling . For small threshold distances , very few taxa are clustered , and the standard error is large , but decreases monotonically as the threshold is increased and more taxa are clustered . To illustrate excess co-clustering , we classified taxa into three categories of CD4: those with CD4 , representing AIDS cases; those with CD4 , and those with CD4 between 200 and 800 . Taxa were also classified into quartiles by FAS . We then counted the number of pairwise clusterings of taxa within and between each category . These counts were arranged in a matrix . Large counts along the diagonal ( within categories ) represent co-clustering by stage of infection . To establish excess co-clustering , we compared the counts to the expectation if clusters were being formed at random , e . g . if two taxa were selected uniformly at random without replacement . We denote the symmetric matrix of co-clustering counts as , so that represents the number of times that a taxon in category is clustered with a taxon in category . The sum of counts in the 'th row of will be denoted . Following the methods described in [29] , the expected value of under random pair formation isBelow , we illustrate the difference . We can also calculate the assortativity coefficient [29] , , which describes the total amount of co-clustering in the matrix . To construct the co-clustering matrices , we selected the value of the distance threshold which maximized the assortativity coefficient . Following the approach outlined in [6] and [30] , we develop a deterministic coalescent model derived from a compartmental susceptible-infected-recovered ( SIR ) model . A system of several ordinary differential equations describe the dynamics of prevalence of early and late HIV infection . Individuals pass from a susceptible state , to an early/acute infection state , to a chronic infection state followed by removal ( treatment or death ) . , and will denote the numbers susceptible , acute , and chronically infected respectively , and the population size will be denoted . For didactic purposes , we will suppose that treatment is completely effective at preventing forward transmissions . The HIV model is described by the following equations: ( 1 ) In these equations , and are respectively the frequency-dependent transmission rates for early and chronic infected individuals . The average duration of early and chronic infection are respectively and . Natural mortality occurs at the rate and immigration into the susceptible state occurs at the rate , which maintains a constant population size . is a term which modulates the way incidence of infection scales with prevalence . For the results presented below , we choose . This term corrects for observed patterns of decreasing incidence with prevalence; this can occur as a result of population heterogeneities ( including sexual network structure ) or as the result of decreasing risk behavior as knowledge of the epidemic spread . Many more relevant details could be included in a model of the HIV epidemic in MSM , however our purpose is to demonstrate how these simple dynamics lead to observed phylogenetic patterns . In [6] , a similar HIV model was presented along with a method to fit such models to a sequence of phylogenetic divergence times ( the heights of nodes in a time-scaled phylogeny ) . Where possible , we will use the parameter estimates from [6] . The parameters are reported in table 1 . Together , these parameters imply and that 41% of transmissions occur during the acute stage . Corresponding to an epidemic model of the form 1 , we can define a coalescent process [31] , [32] that describes the properties of the transmission tree and by extension the phylogeny of virus . The taxa descended from a lineage at time in the past form a clade , which we will also call a cluster . The number of taxa in a randomly selected cluster will be a random variable . The cluster size distribution ( CSD ) is a function of a threshold TMRCA , and describes the probability of having a size cluster if a lineage ( i . e . branch ) at time is selected uniformly at random from the set of all lineages at and the size of the cluster descended from that branch is counted . A schematic of how clusters and the CSD are constructed given a tree and a threshold is shown in figure S5 . In [6] we derived differential equations that describe the moments of the CSD . Some of the properties of phylogenies that we seek to reproduce with the model developed below are: Figure 1 shows a simple genealogy that could be generated by the HIV model . Four events can occur in this genealogy representing coalescence or the changing stage of a lineage . By quantifying the rate that these events occur using a coalescent model , we can calculate the clustering properties of these genealogies . These methods are described below and in detail in supporting Text S1 . The ancestor function is strictly decreasing in reverse time and converges to one ( a single lineage ) when the most recent common ancestor of the sample is reached . The initial value of the ancestor function ( when the population is sampled ) is equal to the sample size . For the purposes of modeling phylogenetic properties of HIV , we will be interested in phylogenies such that the taxa are labeled with the state of the sampled individual ( e . g . the individual will have early or late infection corresponding to the states in equation 1 ) . In this case , we will have two ancestor functions , since a lineage may correspond to an infected individual with either early or late infection . The ancestor functions derived from equations 1 , and which are derived in the Text S1 are as follows: ( 2 ) In these equations , is the number of lineages corresponding to early infections and is the number of lineages corresponding to late infections . These equations provide a deterministic approximation to the NLFT , which is . Each term in these equations accounts for loss or gain of lineages due to the concurrent processes of transmission ( at rates and ) and transition between states ( at rates ) . This approximation becomes exact in the limit of large sample and population size . Note that since the model is continuous in both time and state variables , the ancestor functions are not integers in contrast to most coalescent frameworks based on discrete mathematics . Real epidemics in a finite population will have transmission trees such that the number of lineages at any time is a random variable . The mean-field model presented in equation 1 can be viewed as a description of the dynamics of a stochastic system in the limit of large population size . In this case , we can adapt the coalescent to make approximate descriptions of the stochastic properties of the transmission tree in large populations . The ancestor functions will reflect the approximation of the actual ( random ) number of lineages . Previous work has demonstrated that deterministic descriptions can be excellent approximations for the number of lineages over time [6] , [33] . In the following section , we compare our deterministic coalescent to stochastic simulations , confirming that it is a good approximation over a wide range of parameters . Given a clustering threshold TMRCA , the random variable will be the number of stage- taxa descended from a given lineage that is extant at time in the past . As before , will be the number of type lineages at the time in the past . In our model , infected can be of two types ( early/acute and chronic infected ) , so there are only two types: corresponds to earl/acute and corresponds to chronic . We will denote the set of all lineages of type at time in the past as . Then we define the and 'th moment of cluster sizes descended from a type lineage to be ( 3 ) Many summary statistics that are potentially informative about transmission dynamics can be derived from these moments . The moments are difficult to interpret , so in practice we use them to calculate summary statistics such as variance and skew of the CSD . Below , we examine 30 summary statistics derived from the first three moments and multiple clustering thresholds . For example , the variance of cluster sizes counting only type taxa descended from type lineages is ( 4 ) The total variance of cluster sizes counting only stage 1 taxa is found with the weighted average over lineage types: ( 5 ) A similar set of equations can be developed for the cluster sizes aggregated over taxon types , that is , for . Detailed derivations are provided in Text S1 for differential equations that describe these moments as function of the threshold . Event-driven stochastic simulations were conducted to verify the suitability of the deterministic approximations for inference . Simulations implemented a variation on the Gillespie algorithm [34] . Populations consisted of agents , and were simulated for 15 or 30 years starting with one hundred initial infections . At the end of each simulation , a sample of either 20% or 100% of prevalent infections was taken and used to reconstruct a transmission tree . Five hundred simulations were conducted for each sample fraction and sample time . Corresponding to each simulation , 10 transmission trees were generated based on a random sampling of using distinct clustering thresholds . The CSDs were then estimated from each tree and the moments of these distributions were compared to the moment equations ( 3–5 ) . We have further conducted an investigation into the potential of various summary statistics of the viral phylogeny for inference of underlying epidemiological parameters . Of particular interest is the fraction of transmissions that occur during early HIV infection . As indicated above , it is possible that phylogenetic clustering of early infections reflects elevated transmission during early/acute HIV infection , which we will define as the infectious period from zero to six months . The following simulation experiment was carried out to identify informative statistics: The coalescent tree was simulated such that the sample size matched that of the Detroit MSM phylogeny , and the heterochronous sampling of that phylogeny was reproduced in the coalescent tree . Furthermore , the number of early/acute versus chronic taxa sampled was determined using the BED test for recency of infection for each patient [36] , and simulations were also made to match the numbers of early/acute and chronic taxa sampled . Virus from patients with early/acute infection accounted for 24% of the samples . Summary statistics were centralized around the mean and rescaled by their standard deviation ( ) . The dependent variable of interest is the fraction of transmissions attributable to the acute stage at the beginning of the epidemic , which may be defined ( 6 ) where is the expected number of transmissions generated during early/acute infection at the beginning of the epidemic , and is the expected number of transmissions over the entire infectious period . Pearson correlation coefficients were calculated for each statistic and . To give a better indication which statistics would be useful for estimating the ratio of acute to chronic transmission rates , we conducted a partial least-squares ( PLS ) regression [37] , which has been used by other investigators when estimating parameters by approximate Bayesian computation ( ABC ) methods [38] . Prediction error was assessed with 10-fold cross validation . We controlled for the sample fraction by including the prevalence of infection at the time of the most recent sample as a covariate .
The mean CD4 cell count and FAS for clustered taxa is shown in figure 2 . Consistent with our hypotheses , patients with higher CD4 count are more likely to yield phylogenetically clustered virus , and the mean CD4 count among clustered patients has an inverse relationship with the threshold TMRCA for clustering . Also consistent with our hypothesis , patients which yield virus with lower FAS ( less diverse virus ) are more likely to be phylogenetically clustered , and mean FAS has a positive relationship with the threshold TMRCA for clustering . Patients were strongly co-clustered within quantiles . Maximum assortativity values , which measures the similarity of co-clustered taxa were 13% for CD4 and 4 . 5% for FAS . The maximum assortativity also occurs at low threshold TMRCA for FAS and CD4 ( 1700 and 1467 days ) . Very little clustering is observed between the first and last quantiles . In general , the deterministic model offers an excellent approximation to the stochastic system . All trajectories pass through or close to the median of simulation predictions . Figure 3 illustrates the prevalence of early/acute and chronic infections from a typical simulation of the HIV model and the corresponding deterministic approximations . This correspondence occurs despite large fluctuations in prevalence when the number of infections is small . In [6] it was shown that the correspondence between the stochastic and deterministic systems can be very good even if the epidemic is started from a single infection and the coalescent is fit to the resulting transmission tree . In figure 3 , late infections outnumber early infections by approximately 20 to 1 . As a consequence , NLFT for late infections are more stable due to larger sample sizes , and the NLFT are more noisy for the sample of early infections . The prevalence of infection plateaus prior to the 15 year sample time , so there is not much difference in the phylogenetic features observed at 15 and 30 year sampling times . Many summary statistics calculated from an HIV gene genealogy can be informative about the fraction of transmissions attributable to early/acute infection , ( equation 6 ) . Figure 4 shows the value of four statistics as is varied . The dependancy of these summary statistics on the sample fraction is also shown in figure S4 . ( upper left ) is the Pearson correlation coefficient between the number of early/acute taxa and chronic taxa in a cluster and is most sensitive to . Also shown are the mean cluster size , the number of extant lineages at the threshold TMRCA , and the fraction of taxa in a phylogenetic cluster . As the fraction of transmissions from the early/acute stage is varied , transmission rates and are adjusted so that remains constant . The smallest value of shown in figure 4 corresponds to the point where , such that there is no excess transmission in the early/acute stage . The most recent sample is assumed to be at 35 years following the initial infection . Epidemic prevalence after 35 years is approximately constant . The threshold TMRCA was five years before the most recent sample . Sample size and distribution of samples over time was matched to the Detroit MSM phylogeny . Furthermore , the number of early/acute versus chronic taxa sampled was made to match the Detroit data by use of the BED test [36] for determining recency of infection . The fraction of taxa which are phylogenetically clustered also varies with ( figure 4 , upper left ) . The fraction of early/acute taxa clustered is more sensitive to than the fraction from chronic taxa . Early/acute taxa are always clustered at a greater rate than chronic taxa , even when corresponding to the minimum value of . This is because virus from early/acute patients was recently transmitted , making it much more likely that the lineage will coalesce in the recent past regardless of the source of the infection . Using the mathematical model , we explored many parameters including the threshold TMRCA for clustering , the sample fraction , and the time relative to the beginning of the epidemic at which sampling occurs . Figures S1 , S2 , S3 demonstrate that the deterministic model is capable of reproducing many phylogenetic signatures that have been associated with HIV epidemics in MSM . For example , figure S5 shows the fraction of the sample ( both early and late infections ) which remain unclustered with any other sample unit . When the threshold TMRCA is zero ( corresponding to the far right of the time axis ) , the entire sample remains unclustered . As the threshold TMRCA increases ( moving leftwards on the time axis ) , more sample units become clustered and the fraction of taxa remaining unclustered decreases . The time of sampling makes little absolute difference to the qualitative nature of the tree statistics if sampling occurs after the peak epidemic prevalence ( around 15 years ) . However the sample fraction ( the fraction of prevalent infections sampled ) has a large effect on all tree statistics . When the sample fraction is large , the fraction remaining unclustered drops much more precipitously than when it is small as the threshold TMRCA increases . This occurs because each transmission can cause a sample unit to become clustered; a large sample size implies that transmissions will have a greater probability of resulting in an observable coalescent event ( e . g . it results in a larger ratio ) . Early infections become clustered at a much greater rate than late infections . This corresponds to the excess clustering of early/acute infections observed in many phylogenies . By virtue of being infected in the recent past , an acute infection inevitably has a very recent common ancestor with another infection who transmitted to that individual . Mathematically , this is reflected in transmission terms of the form which appear in the ancestor function for early , but not late infections . When the sample fraction is non-negligible , the fraction of the sample in a cluster levels off for intermediate thresholds . Similar phenomena were noted by Lewis et al . [8] and Hughes et al . [14] who observed that the fraction of the sample in a cluster did not change substantially beyond a small threshold , though these studies probably had high sample fractions . The plateau is due to the bimodality of coalescence times induced by early infection dynamics . Many coalesce events occurs at thresholds close to the sampling time , which corresponds to lineages of early infection coalescing . A larger group of coalescence times occurs close to the beginning of the epidemic when the effective population size is small . We hypothesize that the amount of excess clustering of early infections can be informative for estimating the sample fraction when it is not known . Figure S2 shows the Pearson correlation coefficient for the number of co-clustered early and chronic infections as a function of the clustering threshold ( ) . Given that a sample unit is in a cluster , under certain circumstances , it is much more likely to be clustered with another unit of the same type . This is reflected by large negative correlation coefficients for the number of co-clustered early and late infections for small threshold TMRCA . But negative correlation between the number of early and late infections is only observed for small sample fractions and small threshold TMRCA . The region of negative correlation appears very briefly for a 100% sample fraction; the region is much longer for small samples . This implies that if a patient with early infection is clustered , it is much more likely to be clustered with another early infection than expected by chance alone . The skewness of the CSD shows a similar trend ( figure S3 ) . The skewness is always positive ( to the right ) and rapidly decreases as the threshold TMRCA is increased reflecting greater probability mass in the tail of the distribution . Skew is greatest for small threshold TMRCA , when most clusters are of size 1 . The distribution remains positively skewed , though it quickly levels off for intermediate threshold TMRCA . The mathematical model shows that all moments of the CSD are finite and diverge to infinity in the limit of large sample size and threshold TMRCA . A practical consequence of having an intermediate to large sample fraction is that chains of acute-stage transmission will account for many of the clusters observed at low thresholds . If a taxon is clustered with an early infection , then it is more likely that the unit will be clustered with additional early infections since such cases are highly infectious and have likely transmitted in the recent past . This provides a justification for the theory expounded in Lewis et al . [8] that high clustering of cases with recent MRCA's indicates episodic transmission; chains of transmission by early infections are interrupted by occasional long intervals until a transmission by late stage infections . Corroborating figure 4 which shows that many statistics are correlated with , the PLS regression did not single out any particular group of statistics as being informative of early/acute stage transmission rates . The first component distinguishes between statistics that describe co-clustering ( correlation of the number of acute and chronic taxa in a cluster ) and statistics that describe excess clustering ( e . g . the fraction of early/acute taxa that are not clustered with any other taxa ) . Four principal components were required to explain 42% of the variance of the transmission fraction with additional components only explaining an additional 2% . All statistics were well represented in the model with four components .
We have used coalescent models to characterize the phylogenetic patterns of a virus which produces an early stage of intensified transmission followed by a long period of low infectiousness . These patterns have been observed in multiple phylogenies of HIV-1 from MSM and IDU , and our model suggests that these should be general features for epidemics which feature early and intense transmission . These patterns are not necessarily a consequence of complex sexual network structure [14] . Complex transmission dynamics driven by sexual networks are undoubtedly taking place , but detecting the phylogenetic signature of sexual network structure will require carefully-chosen summary statistics [15] . We have characterized phylogenies using the cluster size distribution ( CSD ) which is similar to commonly used clustering methods based on strong support for monophyly but is nevertheless tractable for mathematical modeling in a dynamical systems framework . Moments of the CSD reflect a wide range of tree topologies , such as the distribution of branch lengths and tree balance , and are potentially informative of a wide range population genetic processes . For example , a highly unbalanced tree would have produce very skewed CSD , and a very star-like tree would have a CSD that is insensitive to changes in the clustering threshold . While there has been much discussion of how clustering of acute infections is caused by the intensity of transmission during the acute stage , the amount of excess clustering that will be observed is also very sensitive to the sample fraction . And even if transmission rates in the early/acute stage are equal to those in the late/chronic stage , we would still observe excess clustering of early/acute provided the sample fraction was large enough . This is a simple consequence of early/acute infections being connected by short branch lengths to the individual who transmitted infection . An advantage of the coalescent framework used in this investigation is that it is accurate even with large sample fractions [35] . Some of the statistics which are most informative of the underlying epidemiological processes are those based on co-clustering of labeled taxa , such as the correlation between the number of early and late infections in a cluster . Such statistics tend to be the most responsive to variation of the intensity of transmission during early infection , and are therefore good candidates for future estimation of the fraction of transmissions that occur during the first few months of infection with HIV . Knowing the frequency of early transmission is essential to prevention efforts , since these transmissions are the most difficult to prevent . Individuals with early and acute infection are usually not aware of the infection , and are therefore not susceptible to many interventions . Modeling to evaluate strategies such seek , test , and treat ( STT ) [39] , [40] and pre-exposure prophylaxis ( PrEP ) [41] will require good estimates for the frequency of early-stage transmission in diverse populations , and phylogenetic data promise to refine these estimates . Future work could focus on finding ways to use statistics derived from the CSD for estimation of epidemiological parameters within an approximate Bayesian framework [38] , [42] , [43] . Alternatively , advances [35] in coalescent theory may make it possible to calculate the likelihood of a gene genealogy conditional on a complex demographic history , such as those generated by the HIV model discussed here . Current techniques are limited in the amount of phylogenetic data that can be used for inference of demographic and epidemiological parameters . Estimation of the intensity of early stage transmission will likely require co-clustering statistics similar to the moments derived from the CSD . In cases where the simple compartmental models fail to reproduce phylogenetic patterns , a more complex transmission system model and its corresponding coalescent should be investigated which might involve sexual networks or geographical [44] and risk heterogeneity . We further conclude that care must be taken in using phylogenetic clusters for epidemiological inference . Mechanisms that generates clustering are often complex and counter-intuitive . We recommend that investigators shift from individual-based inference using small clusters to model-based inference using population-based surveys of sequence diversity . | Diversity of viral genetic sequences depends on epidemiological mechanisms and dynamics , however the exact mechanisms responsible for patterns observed in phylogenies of HIV remain poorly understood . We observe that virus taken from patients with early/acute HIV infection are more likely to be closely related . By developing a mathematical model of HIV transmission , we show how these and other patterns arise as a simple consequence of intensified transmission during the early/acute stage of HIV infection , however observing these patterns is highly dependent on sampling a significant fraction of prevalent infections . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"public",
"health",
"and",
"epidemiology",
"mathematics",
"effective",
"population",
"size",
"epidemiology",
"infectious",
"disease",
"epidemiology",
"genetics",
"population",
"genetics",
"biology",
"evolutionary",
"biology",
"nonlinear",
"dynamics",
"genetics",
"and",
"genomics"
] | 2012 | Simple Epidemiological Dynamics Explain Phylogenetic Clustering of HIV from Patients with Recent Infection |
Previous studies have demonstrated contact irritant and spatial repellent behaviors in Aedes aegypti following exposure to sublethal concentrations of chemicals . These sublethal actions are currently being evaluated in the development of a push-pull strategy for Ae . aegypti control . This study reports on mosquito escape responses after exposure to candidate chemicals for a contact irritant focused push-pull strategy using varying concentrations and focal application . Contact irritancy ( escape ) behavior , knockdown and 24 hour mortality rates were quantified in populations of female Ae . aegypti under laboratory conditions and validated in the field ( Thailand and Peru ) using experimental huts . Evaluations were conducted using varying concentrations and treatment surface area coverage ( SAC ) of three pyrethroid insecticides: alphacypermethrin , lambacyhalothrin and deltamethrin . Under laboratory conditions , exposure of Ae . aegypti to alphacypermethrin using the standard field application rate ( FAR ) resulted in escape responses at 25% and 50% SAC that were comparable with escape responses at 100% SAC . Significant escape responses were also observed at <100% SAC using ½FAR of all test compounds . In most trials , KD and 24 hour mortality rates were higher in mosquitoes that did not escape than in those that escaped . In Thailand , field validation studies indicated an early time of exit ( by four hours ) and 40% increase in escape using ½FAR of alphacypermethrin at 75% SAC compared to a matched chemical-free control . In Peru , however , the maximum increase in Ae . aegypti escape from alphacypermethrin-treated huts was 11% . Results presented here suggest a potential role for sublethal and focal application of contact irritant chemicals in an Ae . aegypti push-pull strategy to reduce human–vector contact inside treated homes . However , the impact of an increase in escape response on dengue virus transmission is currently unknown and will depend on rate of biting on human hosts prior to house exiting .
Dengue , transmitted primarily by Aedes aegypti mosquitoes , is the most important mosquito-borne viral disease affecting humans worldwide [1] . It is caused by four serotypes that produce a spectrum of clinical illness ranging from unapparent or mild disease , to an influenza-like illness , to a fatal shock syndrome . Due to the current lack of a licensed vaccine , dengue prevention is limited to vector control . Aedes aegypti control programs are based on two main targets: ( 1 ) the immature stages ( egg , larvae , and pupae ) through environmental management ( source reduction ) , larvicides and/or biological control; and ( 2 ) the adult stage using space or residual sprays of chemical insecticides and more recently insecticide treated materials [2] . Aedes aegypti has strong associations with human habitations , living and breeding very near or inside human dwellings [3]–[5] . This extensive use of the human indoor environment poses challenges to traditional adult control methods and as well as in devising new or improved methods to sufficiently reduce disease transmission risk [6] . Pyrethroids have commonly been employed in dengue endemic countries such as Thailand and Peru for peridomestic and/or indoor residual/space-spraying to reduce adult mosquito populations [7] , [8] . However , despite the fact that selective use of pyrethroids and other residual insecticides applied indoors have successfully controlled Ae . aegypti and dengue [9]–[13] , outdoor and peridomestic space-spraying alone has often failed to achieve any meaningful control of indoor adult Ae . aegypti populations . This is because the chemical fails to reach the preferred resting sites inside homes [14] , [15] and indoor residual or space-spraying can be resource limited and hampered by poor public perception that limits indoor access [16] therefore reducing treatment coverage . However , the high affinity of Ae . aegypti for the human indoor environment also provides unique opportunities for innovative approaches to control the adult vector [16] . During dengue outbreaks , the methods of choice for emergency interventions remain outdoor ultra-low-volume ( ULV ) application of insecticides and/or indoor thermal fogging [17] , [18] . While effective , these interventions usually follow onset of epidemics , are difficult to implement in urban environments where Ae . aegypti is most common , and more importantly very difficult to sustain [16] . For these and other reasons such as insecticide resistance [7] , [8] , there is a need for proactive measures that have potential to prevent virus transmission and avert dengue epidemics as well as reduce selection pressure for insecticide resistance . Research has shown that the impact of public health insecticides on vector populations is much more complex than simple toxic actions ( i . e . , direct mortality following exposure ) . There are other chemical actions that exist to break human-vector contact [19]–[28] . Such actions include among others , contact irritant effects , causing an escape response from homes , and initiating a spatial repellent or deterrent effect , thereby preventing house entry [26] , [28] , [29] . Both contact irritant and spatial repellent behaviors have been demonstrated in vector populations using sublethal chemical concentrations under both laboratory and field environments [26] , [28] . The importance of these findings is accentuated as both organochlorine resistant and pyrethroid tolerant Ae . aegypti were included in the vector test populations , indicating that behavior-modifying actions exist even when mosquitoes are not susceptible to their toxic actions and as such , may play a role in insecticide resistance management . Current issues of insecticide resistance are creating a necessity for development of new strategies [30] and/or chemical products that act to prevent vector biting . Such approaches will require an understanding of the range of biological actions ( independent from toxicity ) elicited by vectors exposed to chemical tools to help drive such methods for disease control . Both contact irritancy and spatial repellency are currently being evaluated in the development of a push-pull strategy for Ae . aegypti control . The goal of the push-pull strategy is to reduce the probability of human-vector contact and therefore dengue virus transmission . The approach is to target preferred indoor resting sites ( using contact irritants ) and house portals of entry ( using spatial repellents ) using minimum effective chemical concentration and treatment surface area coverage ( SAC ) to make these sites unsuitable , and thereby drive ( push ) the vector away from the treated structure and human hosts in a cost-effective manner . To further enhance the effect of a push response , an attractant trap is placed outdoors to pull the irritated/repelled vectors from the peridomestic environment thereby disrupting human-vector contact in the peridomestic environment . However , a component to system success based on a contact irritant push element is the quantification of a contact irritant response - movement of Ae . aegypti test populations away from a chemical source following tarsal contact–when less than 100% of a surface is treated ( i . e . focal application ) . Target application could prove cost-effective; however , untreated sites inside homes may serve as refuge locations for exposed vectors thereby diminishing efficacy of the approach . Our previous study quantified Ae . aegypti resting behavior under laboratory conditions following exposure to chemical-treated surfaces at various treatment coverage area ( 100% , 75% , 50% and 25% ) to define the effects of “safe sites” ( untreated surfaces ) on irritancy behavior [31] . Results indicated that when preferred Ae . aegypti resting sites were treated with irritant chemicals , even at a treatment coverage of 25% , test populations did not simply move to untreated areas but became agitated , using increased flight as a proxy indicator . It is this contact irritant response that may result in an escape behavior and could be exploited as one type of push component in a push-pull strategy . The objectives of the current study were to quantify contact irritancy ( escape ) response , knockdown , and 24 hour mortality rates in populations of female Ae . aegypti . Evaluations were carried out at varying chemical concentrations and treatment surface area coverage using three standard pyrethroid insecticides: alphacypermethrin , lambdacyhalothrin and deltamethrin . We present findings from the laboratory and two field sites ( Thailand and Peru ) where validation studies using experimental huts were conducted .
Chemicals evaluated were chosen based on current World Health Organization Pesticide Evaluation Scheme ( WHOPES ) recommendations and/or historical use in vector control programs [34] . Compounds were acquired as neat grade material purchased from Sigma-Aldrich ( St . Louis , MO ) : alphacypermethrin ( CAS 67375-30-8 ) , lambdacyhalothrin ( CAS 91465-08-6 ) , and deltamethrin ( CAS 52918-63-5 ) . Test concentrations evaluated included WHO recommended field application rates ( FAR ) and ½FAR; where FAR = 7 . 2 nmol/cm2 ( 0 . 03 g/m2 ) for alphacypermethrin and lambdacyhalothrin , and 4 . 9 nmol/cm2 ( 0 . 025 g/m2 ) for deltamethrin .
The objective of the current study was to evaluate the efficacy of contact irritant chemicals to elicit Ae . aegypti escape responses from a treated space under both laboratory and field conditions . The approach included treating at concentrations below standards for insect toxicity and at surface area coverage below 100% . The overall goal was to describe the potential of focal application of irritant compounds to drive sublethal vector behaviour ( such as escape ) that could disrupt human-vector contact . If evident , the role for these approaches in novel control strategies would be strengthened . In the laboratory , significant escape responses were observed at treatment concentrations equal to but also below corresponding WHO recommended field application rates for the test compounds based on chemical levels required for vector mortality [34] . Our field studies in Thailand corroborate this finding with a 46% increase in the percentage of Ae . aegypti exiting a hut treated with alphacypermethrin at ½FAR and significantly more mosquitoes exiting the treated hut prematurely , by four hours , compared to the control hut . The time of vector escape from inside a treated space is as important as the total density of escape since the faster irritation and escape occurs , the greater the probability that contact ( i . e . , biting ) of the human host has or can been prevented . Combined , these results are similar with previous studies conducted in Thailand that found significant escape response in an Ae . aegypti local strain to alphacypermethrin tested at concentrations below the WHO FAR and continue to support the fact that insecticides can be used to elicit irritant actions even if applied at concentrations below toxic levels [26] , [28] , [39] . The current study also evaluated the escape behavior of Ae . aegypti in response to different treatment surface area coverage ( SAC ) of a space using dark∶light fabric panels , where chemical was applied on dark surfaces only representing focal application . The approach of using contrasting colors and treating black fabric alone was based on exploiting known behavior that Ae . aegypti adults prefer to rest in dark , damp locations in households , and are generally attracted to black colors [40]–[42] . The evaluation of varying SAC was founded on the theory that taking advantage of resting characteristics could result in desired outcomes achieved with less than 100% treatment of indoor resting sites . A previous laboratory study [31] reported on the resting patterns of Ae . aegypti using different dark: light SAC ratios of sublethal concentrations of various pyrethroid chemicals . Results from these experiments indicated that Ae . aegypti preferred to rest on dark versus light-colored surfaces , even at the lowest 25% dark coverage and that when preferred Ae . aegypti resting sites ( i . e . , dark surfaces ) were chemical-treated , mosquitoes did not significantly alter their resting behavior to safe-sites ( untreated areas ) . Instead , they became agitated , and took to flight rather than seeking out alternate resting sites . However , in that study , the test populations were not provided the opportunity to escape from the test arena so measurements of exiting were not captured [31] . We report here that exposure of Ae . aegypti females to alphacypermethrin using FAR at 25% and 50% SAC resulted in escape responses that were comparable with those reported at 100% SAC . In addition , escape responses greater than controls were also observed using treatment coverage at <100% at the ½FAR sublethal concentration of all test compounds . It is noted that in most trials , escape responses were highest when using treated 100%L SAC . This agrees with previous data [31] that showed less overall Ae . aegypti resting on light versus dark fabric material . Higher escape responses on 100%L SAC may be due to the combination effect of light color ( reduction of preferred resting sites ) and the irritating chemical . Field studies from Thailand also indicated a substantial premature escape and an average of 40% increase in exiting an alphacypermethrin-treated hut using 75% SAC . Combined , these results demonstrate that , irrespective of treatment concentrations , substantial escape responses can be elicited in Ae . aegypti test populations at treatment coverage below the traditional 100% SAC used in indoor residual spray campaigns . These findings agree with studies conducted in southern Mexico using selective spraying to target the preferred resting sites of An . albimanus which showed that reduced coverage application was as effective as full spraying in controlling adult populations [43] . In addition , preferred resting sites , as determined by landing frequencies , times of resting , preferred surface types and resting heights were not modified by the insecticide applications . In essence , the preferred resting sites of the mosquitoes did not change , even on insecticide-treated surfaces after four successive spray rounds [43] . The use of this approach - sublethal concentration and <100% treatment area coverage - against Ae . aegypti , or other vector species , if effective , holds the potential for reducing programmatic costs due to reduced requirements for active ingredient . Arredondo-Jimenez et al ( 1995 ) showed that selective spraying of preferred resting sites required 46% less time and cost 67% less than conventional full-spraying . Cost and manpower constraints could be further reduced by integrating the proper active ingredients into consumer-based products ( pre-treated material strips , plastic sheeting , tiles , wallpaper , paint , etc . ) thereby sharing vector control costs with community members . Initial focus group studies evaluating acceptability of sublethal approaches to vector control have indicated that treated materials are viable options for chemical application inside homes [44] . While all trials in the current study were performed with fresh chemical applications , the pyrethroids evaluated are known to have residual properties [34] , [45]–[47] , therefore their effectiveness over a long period of time would be anticipated , further supporting a cost-effective approach . In addition , such control strategies may increase sustainability of an intervention through home ownership , and broaden the delivery platforms available for traditional house treatment . This concept recognizes that householders in many developing countries already buy pesticides to control insects in the homes [44] , [48] , and there are advantages in engaging market forces to promote such products , rather than relying solely on public health appeals . Experimental hut experiments in Peru showed only an 8 . 7% increase in escape from huts treated with alphacypermethrin at FAR using 75% SAC as compared to an untreated hut and no increase in escape rate when alphacypermethrin was used at ½FAR , irrespective of the SAC . In addition , the time of exit in Peru trials was comparable between treatments and chemical-free control . These results are in contrast to both laboratory and Thailand field trials that showed substantial increases in escape rates for the same compound and treatment coverage in relation to controls . These differences are most likely due to the naturally high exit rate of Ae . aegypti observed during baseline studies in Peru when no chemicals were used ( Castro et al . unpublished data ) , making any impact due to chemical difficult to measure , and the variation in fabric material type used between the two field sites ( polyester in Thailand versus cotton in Peru ) . Differences in material type may lead to differences in chemical uptake , absorption and release . Previous studies [49] , [50] showed less mosquito KD and mortality when locally sourced mosquito nets ( treated with K-O Tab 1-2-3 ‘dip-it-yourself’ long-lasting formulation ) were made from cotton as opposed to polyester . Although cotton retained a higher concentration of insecticide , the majority of the chemical is bound within the cotton fibers rather than remaining on the surface where the mosquitoes make contact [49] . However , the chemical was shown to be available for uptake and effect based on indoor KD rates observed within alphacypermethrin-treated huts . The difference between our laboratory findings , where minimal escape occurred from control chambers , and results from our field studies in Peru , where high escape occurred from chemical-free huts , most likely reflects influences from environmental parameters such as temperature and humidity in the micro-environment of the experimental huts . A previous study in Thailand demonstrated how Ae . aegypti exit behavior is affected by ambient environmental factors of temperature and humidity [51] . The possibility that behavioral variation between the Thailand and Peru study sites in the current study is the result of the geographically different strains is unknown . These results illustrate that despite considerable progress in the field of vector behavior as a whole , there remains much to understand on how external and genetic factors affect the biology and behavior of important disease vectors . Both laboratory and field results from the current study showed that KD and mortality were considerably low ( <90% ) in our Ae . aegypti test populations despite using FARs that are based on concentrations required to produce lethal outcomes ( i . e . LD90 ) [52] . Aedes aegypti THAI strain from Pu Tuey has been characterized as pyrethroid tolerant [27] , [33] while the PERU strain from Iquitos , Peru is most likely susceptible to the chemicals evaluated in this study ( Vasquez La Torre , personal communication ) . It was therefore expected that KD and mortality would be low in test populations from Thailand . However , the low mortality rates observed in Ae . aegypti from Peru that did not escape in our laboratory tests could be explained by the minimum exposure time compared to standard resistance testing [52] , [53] ( i . e . 10 min in a single contact irritancy replicate versus standard one hour used in toxicity assays ) ; also by the potential difference in chemical uptake , absorption and release between the cotton material used in this study and bottle and/or filter test methods used in many resistance testing [52] , [53] . More importantly , our Ae . aegypti assay cohorts from Thailand continued to exhibit a contact irritant response when exposed to test pyrethroids , despite indication of tolerance , thus indicating that a sublethal approach to vector control may be effective in locations where insecticide resistance occurs . It is worth noting that highest mortality was observed at FAR at 25%D in mosquitoes that did not escape . This finding suggests that although they did not escape , the mosquitoes rested on treated material long enough to receive a lethal dose . Combined , results suggests that escape can occur from a treated area at <100% SAC and of those that do not escape , resting on treated surfaces will continue as supported by a previous study [31] and result in mortality . As expected , we also show that KD and mortality were lower in mosquitoes that escaped as compared to those that did not escape . It is logical that killing those vectors that remain in a treated space ( i . e . inside a home ) and may have contact with humans will impact pathogen transmission . However , the role of escape and survival for disease management seems counterintuitive . We theorize that an escape and survival response could lead to reduced selection pressure for insecticide resistance which would not only potentially interrupt human-vector contact inside the home through contact irritancy but also extend the effective life of the chemical . Those mosquitoes that are irritated and exit have the opportunity to transfer their genetic material to the next generation , thereby potentially selecting for the mechanistic pathway that result in behavior modification – i . e . , irritancy . Of course , this is dependent on linking the mode of action for contact irritancy to one that is dependent on up- or down-regulation of particular cascade drivers . Such information is not yet available . The inclusion of an outdoor trap , such as in a push-pull strategy , could serve to remove irritated vectors from the peridomestic environment thereby controlling potential movement to an untreated location and hosts . Pertinent to the goal of developing a push-pull strategy , results presented here suggest a role for contact irritancy to drive vectors from a space occupied by a human host , and indicate efficacy using minimal chemical concentration ( <LD90 ) and treatment coverage area ( <100% ) . This in spite of the fact that the chemical delivery platform and formulation we evaluated were not formulated ( i . e . , optimized for chemical residuality ) . Although encouraging , current results suggest that behaviors observed in one species or strains might not translate into other mosquito species or strains found in different environmental settings . In addition , the inside environment of an experimental hut is much different than that of a local home; therefore future evaluations of contact irritant efficacy must be performed using indigenous houses against natural Ae . aegypti populations . Lastly , although contact irritancy may promote escape , it will be critical to measure the level of reduction , if any , in human-vector contact ( i . e . , biting ) to truly understand the impact of sublethal behaviors on pathogen transmission . The irritant response must occur prior to blood meal acquisition . Due to this challenge , we are therefore currently applying a similar concept of sublethal concentrations and focal treatment application to the use of spatial repellent compounds in a push-pull strategy . This is an attempt to exploit the deterrent actions of chemicals to prevent Ae . aegypti entry into homes and thereby reduce indoor densities of mosquito vectors available for human-vector contact . The latter , if successful , may provide greater protection from pathogen transmission as compared to contact irritancy as repellency functions to provide a barrier of protection that is not dependent on vector contact with a treated surface [54] . | Dengue virus is spread by Aedes aegypti , a mosquito that prefers to feed on humans . Chemicals used at toxic levels is currently the only confirmed effective strategy for dengue vector control , but insecticide resistance issues are threatening this approach . Other chemical actions that break vector-human contact exist: contact irritant effects , causing escape from homes and spatial repellent effects , preventing house entry . These actions are being evaluated in the development of a push-pull strategy to target indoor resting sites ( using contact irritants ) and/or portals of house entry ( using spatial repellents ) at sublethal concentrations to push the vector away from the treated structure and associated human hosts . When combined with an attractant trap placed outdoors to pull the irritated/repelled vectors from the peridomestic environment , the integrated stimuli form a push-pull system that can be used to further enhance disruption of human-vector contact . Here we report on the quantification of contact irritancy ( escape ) behavior , knockdown and 24 hour mortality rates in populations of Ae . aegypti under both laboratory and field conditions . Evaluations were carried out against three pyrethroid insecticides . Findings indicate an increase in escape response using sublethal concentrations and focal application suggesting a potential role for contact irritants in a push-pull strategy . | [
"Abstract",
"Introduction",
"Methods",
"Discussion"
] | [
"biology"
] | 2013 | Contact Irritant Responses of Aedes aegypti Using Sublethal Concentration and Focal Application of Pyrethroid Chemicals |
Praziquantel ( PZQ ) is a key therapy for treatment of parasitic flatworm infections of humans and livestock , but the mechanism of action of this drug is unresolved . Resolving PZQ-engaged targets and effectors is important for identifying new druggable pathways that may yield novel antiparasitic agents . Here we use functional , genetic and pharmacological approaches to reveal that serotonergic signals antagonize PZQ action in vivo . Exogenous 5-hydroxytryptamine ( 5-HT ) rescued PZQ-evoked polarity and mobility defects in free-living planarian flatworms . In contrast , knockdown of a prevalently expressed planarian 5-HT receptor potentiated or phenocopied PZQ action in different functional assays . Subsequent screening of serotonergic ligands revealed that several ergot alkaloids possessed broad efficacy at modulating regenerative outcomes and the mobility of both free living and parasitic flatworms . Ergot alkaloids that phenocopied PZQ in regenerative assays to cause bipolar regeneration exhibited structural modifications consistent with serotonergic blockade . These data suggest that serotonergic activation blocks PZQ action in vivo , while serotonergic antagonists phenocopy PZQ action . Importantly these studies identify the ergot alkaloid scaffold as a promising structural framework for designing potent agents targeting parasitic bioaminergic G protein coupled receptors .
Schistosomiasis is a neglected tropical disease that infects over 200 million people worldwide , burdening economies with an annual loss of several million disability-adjusted life years [1–3] . The disease is caused by parasitic flatworms of the genus Schistosoma and treatment is largely reliant on a single drug—praziquantel ( PZQ ) , used clinically for over 30 years [4–6] . PZQ is a synthetic tetracyclic tetrahydroisoquinoline that was initially developed by Merck while screening for compounds with tranquilizer properties , and arose from a compound that lacked sedative properties but was remarkably effective against parasitic flatworms [7 , 8] . PZQ has shown remarkable durability compared with other anthelmintics , but incidences of decreased PZQ efficacy have been reported in both the laboratory [9–11] and the field [12 , 13] , raising concerns that PZQ-resistant strains of schistosomiasis may emerge especially as eradication initiatives increase distribution of this drug [4] . Development of alternative therapies to PZQ has been hampered by the fact that the mechanism of action of PZQ remains unresolved and rationally designed derivatives of PZQ typically prove less efficacious [7 , 14 , 15] . These longstanding roadblocks impair the iteration of next generation antischistosomals needed to counter the likely emergence of PZQ-resistant isolates [16] . Resolution of the pathways engaged by PZQ in vivo is therefore a key priority . A fresh perspective toward this problem comes from the discovery of an unusual axis-duplicating effect of PZQ during regeneration of free-living planarian flatworms [17 , 18] . The striking phenotype of PZQ-evoked bipolarity ( Fig 1A ) , coupled with the genetic tractability of this system for RNAi [19] and a retained predictive value against parasitic flatworms [20] establishes a novel platform for identifying relevant in vivo effectors of PZQ action based on the molecular phenology between these systems . In planarian regenerative screens , the bipolarizing efficacy of PZQ depends on a coupling of voltage-operated Ca2+ channels to bioaminergic signals [20] , which likely regulate polarity signaling from flatworm muscle cells to coordinate regenerative outcomes [21] . Here , we apply genetic and pharmacological approaches to dissect our observation that activation of serotonergic signaling in the planarian Dugesia japonica blocks the bipolarizing ability of PZQ . This effect is shown to be unique to serotonin ( 5-HT ) , and highlights the importance of characterizing serotonergic receptors to identify 5-HT blockers that could potentiate , or phenocopy , PZQ action . Intriguingly , serotonergic screens highlight ergot alkaloids as a class of compounds that potently and penetrantly miscued planarian regeneration and schistosomule muscle function , with structure activity insight from active compounds highlighting modifications of the ergot scaffold predictive for flatworm efficacy . Based on these data , we contend that the ergot alkaloid scaffold merits further exploration to yield novel chemotherapeutics with selective efficacy against parasite musculature .
Exposure of regenerating planarian ( D . japonica ) trunk fragments to praziquantel ( PZQ ) yielded two-headed worms ( Fig 1A , [17] ) , an effect never observed in the absence of drug exposure . This effect was dose-dependent ( EC50 38±3 . 6μM , Fig 1B ) , with maximal doses being completely penetrant [17] . Strikingly , the bipolarizing action of PZQ was blocked by co-incubation with 5-HT , or the analogue O-methylserotonin ( O-MT ) , but not by co-incubation with other bioaminergic neurotransmitters ( Fig 1C ) . As this result suggests serotonergic signals functionally oppose PZQ action , we implemented genetic ( Fig 2 ) and pharmacological strategies ( Fig 3 ) to interrogate 5-HT signaling pathways in planarians . To enable interrogation of 5-HT receptor function by in vivo RNAi , we generated a de novo transcriptome assembly for D . japonica ( see Methods ) to allow a comprehensive bioinformatic identification of 5-HT receptors in this system ( Fig 2A ) . A total of 17 predicted serotonergic G protein coupled receptor ( GpCR ) sequences were identified based upon homology to previously identified sequences [22–24] . These putative 5-HT receptors clustered into three discrete clades ( S1- , S4- and S7-like , Fig 2A ) defined by homology with C . elegans serotonin receptors ( SER1 , SER4 & SER7 ) [24] . Previously identified planarian 5-HT receptors ( 5HTLpla1-4 , DjSER-7 , DtSer1 [22 , 23 , 25] ) all localized within the Ser-7 clade . To simplify nomenclature for these sequences , we assigned names to each receptor based on these three clades and transcript abundance within each grouping ( from FPKM values , fragments per kilobase of transcript per million mapped reads ) , such that the most abundant transcript in the S7 clade was named S7 . 1 and the least abundant of the eight transcripts was designated as S7 . 8 . Comparison of FPKM values for all these sequences revealed that S7 . 1 was the most abundantly expressed 5-HT receptor in this system ( Fig 2B ) , accounting for ~40% of the total FPKM values assigned to all predicted serotonergic receptors . As the most abundant receptor , S7 . 1 had previously been cloned by degenerate PCR ( 5HTLpla4 , DtSer1 [23 , 25] ) and recently demonstrated to couple to cAMP generation [25] . Expression levels of S7 . 1 mRNA changed during regeneration [23] , and we observed increased FPKM values for S7 . 1 at early regenerative timepoints ( Fig A in S1 Text ) . Although prediction of the S1 , S4 and S7-like sequences as serotonergic GpCRs is based on specific sequence features known to be important for 5-HT binding ( see below ) , as well as overall homology to other serotonin receptors ( Fig B in S1 Text ) , we do note that both the planarian receptor ( S7 . 4 , DjSER-7 [22] ) and a schistosome S7-like receptor have been successfully deorphanized following heterologous expression and shown to respond to 5-HT [25 , 26] . Alignment of the planarian sequences with human bioaminergic GpCR sequences ( Table 1 ) revealed conservation of key residues within the orthosteric binding pocket known to be important for ligand binding . With reference to molecular docking studies of 5-HT into crystal structures of human 5-HT1B ( and 5-HT2B ) receptors [27] , these include ( i ) a salt bridge between the amino group of 5-HT and D3 . 32 in the 5-HT1B receptor ( itself stabilized by Y7 . 43 , Ballesteros & Weinstein numbering [28] ) , ( ii ) a hydrogen bond from T3 . 37 to the indole ( N-H ) hydrogen of 5-HT , and ( iii ) a hydrophobic cleft formed by contributions from ( W6 . 48 , F6 . 51 , F6 . 52 , C3 . 36 and I3 . 33 ) . All these residues are well conserved in the planarian receptor sequences ( Table 1 ) . Notably , bioaminergic receptors that respond to different ligands ( e . g . dopamine , adrenaline , histamine ) present a more polar interface at resides 5 . 42 and 5 . 46 , whereas human 5-HT sequences present no more than one polar residue ( [27] , Table 1 ) . This feature has been suggested to facilitate interaction with the less polar indole group of 5-HT compared to the other bioaminergic transmitters [27] . The planarian sequences also conform to this principle with the combination of residues at this position being diagnostic of the three different clades of 5-HT receptor sequences ( e . g . ‘AA’ for S7 , ‘S/A , A/S for S4 , ‘xT’ for S1 , Table 1 ) . Another notable feature of the planarian 5-HT groupings is receptor architecture , for example the spacing between these critical residues in helix 3 and helix 5 , and helix 5 and helix 6 ( Table 1 ) appears diagnostic of the different serotonergic clades . For example , the S7 clade exhibits a consistent spacing ( ~74 residues ) between 3 . 37 and 5 . 42 , and a shorter third intracellular loop between TM5 and TM6 compared with the other clades . As S7 represented the most abundantly expressed clade of 5-HT GpCRs , we proceeded to perform RNAi against each individual receptor . First , we screened for effects on PZQ-evoked bipolarity . RNAi-mediated suppression of S7 . 1 potentiated the number of two-headed regenerants ( 82±6% ) compared to the number of bipolar worms observed in control RNAi cohorts ( 59±5% at submaximal PZQ ) . Estimation of the effectiveness of knockdown of S7 . 1 transcripts was assessed by qPCR analysis in the same cohorts used for the regenerative assays . These assays revealed a decrease of 43±3% of S7 . 1 mRNA relative to controls ( Fig 2C , inset ) . Aside from the polarity effect on regenerating fragments , S7 . 1 RNAi also impaired the movement of intact worms . Planarians subject to S7 . 1 RNAi showed decreased mobility ( Fig 2D ) , quantified by monitoring the distance traversed by S7 . 1 RNAi worms ( 61±4mm , average of 10 worms , n = 3 independent RNAi cohorts ) compared with controls ( 104±7mm ) over the same time period ( 2 mins ) . RNAi targeting other receptors in the S7 clade failed to yield a clear defect . We conclude knockdown of S7 . 1 modulated both regenerative polarity and motility outcomes . Next , we employed a pharmacological approach to manipulate serotonergic signals by screening agents with known affinity for serotonergic receptors . While diverse classes of serotonergic blockers caused regenerative bipolarity , the penetrance was typically much lower than seen with PZQ ( Table A in S1 Text ) . Results with ergot alkaloids were however of interest . Ergot alkaloids are a historically important class of compounds that realize their effects because of the close structural similarity of the ergoline scaffold to bioaminergic transmitters . Numerous ergot compounds yielded regenerative phenotypes , either phenocopying PZQ to promote bipolar ( ‘two-head’ ) regeneration or inhibiting head regeneration ( ‘no-head’ , Fig 3A ) , all at doses lower than PZQ . This broad efficacy of ergot alkaloids as a chemical class permitted structural-activity insight into features associated with specific polarity effects: for example , all ergots that caused bipolarity were either alkylated on the indole nitrogen or halogenated at the adjacent 2-position ( Fig 3A ) . In contrast , all ergots that inhibited head regeneration lacked such modifications ( Fig 3A ) . Structural studies have shown that the indole N1 hydrogen forms a key hydrogen bond with a conserved threonine residue T3 . 37 [27] within the orthosteric binding pocket of 5-HT GpCRs that is likely important for receptor activation [27 , 29] . This residue is also well conserved in the planarian 5-HT receptors sequences ( Table 1 ) . Disruption of this interaction by receptor mutagenesis interferes with 5-HT receptor activation by ergot alkaloids [29] . Similarly , alkylation of ergot derivatives at the N1 position also can cause decreased receptor activation yielding compounds that act as 5-HT receptor antagonists [30] . Therefore , this structural feature of the bipolarizing ergot compounds suggests they work through serotonergic blockade . This is consistent with observations that ( i ) structurally diverse 5-HT antagonists cause bipolarity ( Table A in S1 Text ) , ( ii ) the ergots that inhibited head regeneration act as 5-HT agonists in other systems [31–33] , ( iii ) other drugs that stimulate 5-HT signaling ( 8-OH DPAT and fluoxetine ) also block head regeneration and PZQ action [20] , and ( iv ) RNAi of tryptophan hydroxylase ( TPH ) to decrease 5-HT levels potentiates PZQ action [20] . Therefore , these data show that PZQ action mimics the bipolarizing ability of serotonergic blockers , and is opposed by 5-HT agonists . The importance of identifying new drugs from planarian regenerative screens extends beyond basic science as planarian regenerative assays can predict the efficacy of compounds against parasitic worms [20] . Exploiting this phenology may assist discovery of new drug leads and targets for treating parasitic disease . Therefore , we were interested to assess whether the same set of compounds active in regeneration assays displayed activity against schistosomules , the immature form of parasitic schistosome flatworms that exist after penetration of host skin . Schistosomes display an endogenous contractile cycle permitting drug-evoked effects to be easily screened ( paralysis versus stimulation of contractility , Fig 3B ) . In schistosome contractility assays , the compounds that caused planarian bipolarity all inhibited schistosomule motility ( just like PZQ ) , whereas the compounds that inhibited planarian head regeneration caused the opposite effect , stimulating contractile activity ( Fig 3B ) . Therefore , ergot alkaloids possess efficacy against schistosomules , with an action predictable by planarian polarity outcomes . What is the molecular basis of this phenology between planarian regenerative polarity and schistosome motility ? An appealing explanation relates to the recent identification of muscle cells as the coordinating nexus of positional signaling during planarian regeneration [21] . Specifically , a subepidermal population of myocytes was identified to coexpress all the relevant ‘position control genes’ known to regulate the planarian body plan , from which positionally appropriate transcriptional responses are engaged on injury [21] . This discovery is enlightening as it harbors the potential to rationalize a long literature on the effects of exogenous agents on regeneration dating back decades by suggesting that drugs which miscue regenerative patterning all possess a shared ability to modulate excitable cell physiology and perturb muscle function . Therefore , we examined whether PZQ and the ergot alkaloids discovered to miscue polarity , impacted planarian motility . Acute incubation of intact worms with PZQ caused worms to adopt a spastic , curled morphology with inhibitory effects on worm motion ( Fig 4A ) . This effect was dose-dependent ( Fig 4B ) , with a concentration-dependence similar to that observed for the ( longer term ) polarity effect ( EC50 = 38±3 . 6μM for bipolarity in trunk fragments versus EC50 = 23±2 . 4μM for mobility in intact worms , Figs 1B and 4B ) and reversible following drug removal ( Fig 4C ) . Just as observed with PZQ-evoked bipolarity , the immobilizing action of PZQ was also reversed by co-incubation with O-MT , but not other bioaminergic neurotransmitters ( Fig 4C ) . Finally , each of the ergot compounds discovered to cause bipolarity ( Fig 3A ) also inhibited planarian mobility ( Fig 4D ) , underscoring the association between polarity-miscuing drugs and the ability to perturb flatworm muscle function . Finally , we returned to the fundamental observation of functional antagonism between serotonergic signals and PZQ action ( Figs 1 and 4C ) . Does serotonergic activation modulate PZQ-evoked immobility in schistosomules ? To address this , we examined the ability of exogenous O-MT to reverse PZQ-evoked effects on contractility ( Fig 5A ) and morphometry ( Fig 5B and 5C ) . Addition of O-MT markedly ameliorated both the paralytic and compressed worm morphology resulting from PZQ exposure ( Fig 5 ) . Therefore , we conclude that PZQ action is functionally antagonized by serotonergic activation in schistosomule motility experiments , just as observed in planarian assays ( Figs 1 and 4C ) .
The observations that serotonergic activation opposes , while serotonergic inhibition mimics PZQ action , reveal that serotonergic signaling is an important modulator of PZQ efficacy in vivo . We have previously suggested that PZQ engages dopaminergic pathways to subvert regeneration and it is noteworthy that both dopamine and serotonin regulate cAMP turnover with opposing effects on flatworm musculature [34–36] . Levels of cAMP change during regeneration [37] and cAMP is a known mediator of flatworm muscle contraction [38] . Therefore , this ‘functional antagonism’ model ( Fig 6 , [20] ) envisages opposing Ca2+ entry pathways ( Cav1A versus Cav1B ) coupling to discrete bioaminergic neurotransmitters that differentially couple to cAMP within the excitable cell niche . Functional opposition of these bioaminergic systems are well evidenced in many systems [39] . Importantly , we demonstrate here that ergot alkaloids are efficacious modulators of planarian regeneration and motility ( Figs 3 and 4 ) . These two phenotypes are linked as surprisingly planarian polarity genes localize in a supepidermal population of muscle cells [21] . Indeed , ergot alkaloids have a well appreciated ability to modulate smooth muscle contraction based on their bioaminergic mimicry , a property that underpins several of their applications in the clinic . Beyond this ability to regulate muscle ( including opposing effects on flatworm musculature [34–36] ) , dopamine and serotonin also are known regulators of Wnt signaling . D2Rs selectively associate with both β-catenin ( to inhibit Wnt signals [40] ) and Cav channels ( to regulate their expression [41] ) . 5-HT is a well-established wounding signal [42] , long range messenger involved in regenerative proliferation [43 , 44] and a reciprocally permissive cue for Wnt signaling [45 , 46] . Such associations provide precedence for coupling bioaminergic activity to the more established players of planarian regenerative signaling that localize with a myocyte population likely regulated by bioaminergic cues . Possibly all drugs that miscue regenerative polarity share such a commonality of action on the excitable cell niche . Planarian regenerative screens hold predictive significance for discovering new drug leads and targets in parasitic flatworms [20] . Given the ease of performing drug screens in free living planarians compared to their parasitic cousins , this could be a fruitful source of novel therapeutic leads . 5-HT signaling in parasitic schistosomes is an appealing choice for therapeutic intervention given the dynamic expression of serotonergic gene products across the parasite life cycle [47–49] and a clearly evidenced role for bioaminergic signals in regulating muscle [34–36] . Parasite survival within the host requires worm muscle functionality: for example , muscle activity appears to be required for female pairing within the male gynecophoric canal , egg production and maintaining adult worm residency within the mesenteric vasculature . Paralytic agents such as PZQ have been proposed to act as antischistosomals by causing immobilized worms to shift from the mesenteric veins to the liver where they are eliminated [50] . Therefore , miscuing muscle function through bioaminergic cues is a promising route for drug intervention . Our data , revealing an ergomimetic quality to PZQ action , provide impetus for considering ergot alkaloids as potential drug leads for manipulating bioaminergic GpCRs to provide next generation antischistosomals [51] . Ergot alkaloids have been used clinically in a range of applications ( migraine , obstetrics , Parkinson’s disease , diabetes ) , although owing to their broad GpCR binding profile they are often written off as problematic , ‘dirty’ compounds [30] and therefore often deliberately excluded from drug screens . However , this may be an oversight in the context of parasitic chemotherapy . Certain ergot compounds are penetrant and potent in planarian assays compared with PZQ . Further , the clear structural-activity principles emerging from our screen in free-living and parasitic worms ( Fig 3 ) could illuminate structural differences in flatworm GpCR structure compared to their human hosts that may facilitate parasite targeting and mitigate host side effects . Based on our data from the planarian polarity and motility screens that are predictive of parasitic worm phenotypes , we contend that the ergot alkaloid scaffold merits further exploration by medicinal chemistry to identify novel chemotherapeutics with efficacy against parasite muscle .
A clonal line of Dugesia japonica ( GI strain ) was maintained at room temperature and fed strained chicken liver puree once a week [52] . Regenerative assays were performed using 5 day-starved worms in pH-buffered Montjuïch salts ( 1 . 6mM NaCl , 1 . 0mM CaCl2 , 1 . 0mM MgSO4 , 0 . 1mM MgCl2 , 0 . 1mM KCl , 1 . 2mM NaHCO3 , pH 7 . 4 buffered with 1 . 5mM HEPES ) and regenerative phenotypes archived using a Zeiss Discovery v20 stereomicroscope and a QiCAM 12-bit cooled color CCD camera [52] . Data were analyzed using two-tailed , unpaired t-tests , and presented as mean ± standard error of the mean from at least three independent assays . Commercially available ergot alkaloids were sourced as follows: Sigma ( bromocriptine , metergoline , nicergoline , ergotamine , dihydroergotamine ) ; Tocris ( LY215840 , mesulergine , methylergometrine ) ; THC Pharm ( BOL-148 , lysergol , elymoclavine ) . All other chemicals were from Sigma-Aldrich except where specified . Total RNA from regenerating and intact D . japonica was harvested in Trizol and mRNA was purified by hybridization to oligo ( dT ) beads ( Dynal ) . RNA-seq libraries were prepared according to the Illumina mRNA-Seq Sample Prep kit and Illumina TruSeq kit manufacturer protocols . Libraries were sequenced on Illumina HiSeq 2000 machines , producing 100bp paired end reads . Adapter sequences were trimmed and reads were passed through a sliding window quality filter ( window size = 4 , minimum average quality score = 25 ) using Trimmomatic version 0 . 22 [53] . Paired-end reads and singletons ≥ 50 bp in length were retained . Overlapping paired-end reads were merged using FLASH [54] . Surviving reads were combined and fed into the Trinity pipeline for de novo assembly [49] . Final assembly was carried out with a minimum k-mer coverage of 2 and the default k-mer size of 25 . Complex graphs that proved unresolvable within a 6 hour window were manually excised to allow the assembly to proceed . The minimum contig or transcript length for both assembly pipelines was set to 100 nt . Candidate D . japonica 5-HT receptor sequences were selected based upon homology to receptors predicted in the planarian Schmidtea mediterranea [24] . Alignments were performed on predicted amino acid sequences in SeaView ( version 4 . 5 . 1 ) using MUSCLE . Maximum likelihood phylogenies were generated using PhyML at 500 bootstrap replicates and visualized using FigTree ( version 1 . 4 . 0 ) . The depth of this resulting assembly proved comparable to transcriptomes generated for other planarian species [55 , 56] , as well as the predicted open reading frames of the S . mediterranea genome [57] , indicating that this resource is a reliable reference for the prediction and cloning of D . japonica gene products . The high level of coverage is evidenced by the fact that , of the 983 existing D . japonica nucleotide sequences manually cloned and deposited on NCBI , 982 are represented in our de novo assembly with a high degree of sequence identity . Reads were mapped onto the de novo assembly using RSEM [58] to obtain FPKM values reflecting transcript abundance . Sequences are provided as Supplementary material ( Datasets A and B in S1 Text ) . Total RNA was isolated from 50 starved , intact planarians using TRIzol and poly-A purified using a NucleoTrap mRNA mini kit . cDNA was synthesized using the SuperScript III First-Strand Synthesis System ( Invitrogen ) . Gene products were amplified by PCR ( LA Taq polymerase ) , ligated into pGEM-T Easy ( Promega ) for sequencing , and subcloned into the IPTG-inducible pDONRdT7 RNAi vector transfected into RNase III deficient HT115 E . coli . In vivo RNAi was performed by feeding [52] , and a Schmidtea mediterranea six-1 ( Smed-six-1 ) construct , which did not yield a phenotype in D . japonica , was used as a negative control . Cohorts of worms were fed bacterially expressed dsRNA targeting individual 5-HT receptors or the negative control over a total of five feeding cycles ( three RNAi feedings separated by 1–2 days , followed by amputation , regeneration , two more RNAi feedings , followed by excision of trunk fragments for regenerative assays ) . Targeted sequences for RNAi are provided in Supplementary Materials ( Dataset A in S1 Text ) . Knockdown was assessed by quantitative RT-PCR . Total RNA was isolated from 10 intact worms , treated with DNAse I ( Invitrogen ) and cDNA synthesized using oligo ( dT ) primers and the SuperScript III First-Strand Synthesis System . Samples ( 10 intact worms ) were homogenized in Trizol to extract total RNA which was treated with DNAse I ( Life Technologies ) and 500ng were used for cDNA synthesis using random hexamers ( SuperScript III First-Strand Synthesis System , Life Technologies ) . No RT controls were produced by using the same procedure but substituting DEPC-treated water for SuperScript RT enzyme . TaqMan qPCR reactions were performed using custom-designed TaqMan Gene Expression Assays ( Applied Biosystems ) . Assays were designed for GAPDH ( F’ GCAAAAGACTGTTGATGGACCAT , R’ CACGGAAAGCCATTCCAGTTATTTT , probe sequence CCTCTGCCATCTCGCC ) and 5-HTR 7 . 1 ( F’ CAATCTATCAAGGTTAGCTATTCCATTCGA , R’ GCTCCCACAACGATAATAAAAAATATAATCCC , probe sequence ACCAACCGGATATTTT ) and cycled in a StepOnePlus Real-Time PCR System ( Applied Biosystems ) at 50°C/2min , 95°C/10min , 40 cycles of 95°C/15sec and 60°C/1min . 5-HTR 7 . 1 mRNA abundance was quantified by the ΔΔcT method relative to GAPDH . Starved worms were exposed to drug / vehicle for five minutes , after which 10 animals were placed in drug-containing solution in the middle of a glass watchglass ( 50mm diameter , Fisher Scientific ) centered over a LED backlit light ( Edmund Optics , #83–873 ) . Movement was captured using a digital video camera ( Canon VIXIA HF R400 ) over a 2 minute period ( 30 frames per second ) . Representative images of this assay are displayed as minimal intensity z-projections ( ImageJ ) to provide a qualitative visual readout of experimental manipulations . The resulting videos were processed using custom written algorithms in Ctrax to track the motility of individual worms [59] . Motion was scored by quantifying total distance travelled ( mm ) over the fixed recording interval and averaged for the 10 worms in each assay . Errors in tracking were corrected using the Fix Errors Matlab Toolbox and descriptive statistics were computed using scripts in the Behavioral Microarray Matlab Toolbox and custom written algorithms in MATLAB . Biomphalaria glabrata snails exposed to miracardia ( NMRI Puerto Rican strain of Schistosoma mansoni ) were obtained from the Biomedical Research Institute ( Rockville , MD ) and maintained at 26°C for 4 to 6 weeks . Isolation of matured cercaria and their transformation into schistosomules were performed as previously described [20] . For contractility assays , a custom written plugin ( wrMTrck ) in ImageJ was used to resolve schistosomule body length ( major axis of an ellipse ) over time following drug exposure ( 30min ) , as previously described [20] . For experiments on PZQ and 5-HT action on schistosomules , Basch media was made without 5-HT and drugs were added to the concentrations indicted . | The parasitic infection schistosomiasis afflicts millions of people worldwide and is clinically treated using a single drug , praziquantel ( PZQ ) . Despite the fact that PZQ has served as a stalwart anthelmintic for decades , the molecular basis of action of this clinical agent is poorly understood . This lack of mechanistic information impedes the rational design of alternative therapies and highlights the need for new approaches for studying the target ( s ) and effectors engaged by PZQ in vivo . Here , we exploit the predictive phenology between free-living planarian regenerative screens and parasitic neuromuscular physiology to reveal a broad efficacy of ergot alkaloids in phenocopying the action of PZQ . In planarian regenerative screens , data highlight structural features of the ergoline scaffold that yield specific regenerative effects to promote or inhibit head regeneration . Ergot alkaloids with efficacy in regenerative assays were also found to modulate the contractility of schistosomules . Overall , these data highlight a possible therapeutic potential of ergot alkaloids as antischistosomals and the action of PZQ as an ergomimetic . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Ergot Alkaloids (Re)generate New Leads as Antiparasitics |
Cell-cycle progression is governed by a series of essential regulatory proteins . Two major regulators are cell-division cycle protein 20 ( CDC20 ) and its homologue , CDC20 homologue 1 ( CDH1 ) , which activate the anaphase-promoting complex/cyclosome ( APC/C ) in mitosis , and facilitate degradation of mitotic APC/C substrates . The malaria parasite , Plasmodium , is a haploid organism which , during its life-cycle undergoes two stages of mitosis; one associated with asexual multiplication and the other with male gametogenesis . Cell-cycle regulation and DNA replication in Plasmodium was recently shown to be dependent on the activity of a number of protein kinases . However , the function of cell division cycle proteins that are also involved in this process , such as CDC20 and CDH1 is totally unknown . Here we examine the role of a putative CDC20/CDH1 in the rodent malaria Plasmodium berghei ( Pb ) using reverse genetics . Phylogenetic analysis identified a single putative Plasmodium CDC20/CDH1 homologue ( termed CDC20 for simplicity ) suggesting that Plasmodium APC/C has only one regulator . In our genetic approach to delete the endogenous cdc20 gene of P . berghei , we demonstrate that PbCDC20 plays a vital role in male gametogenesis , but is not essential for mitosis in the asexual blood stage . Furthermore , qRT-PCR analysis in parasite lines with deletions of two kinase genes involved in male sexual development ( map2 and cdpk4 ) , showed a significant increase in cdc20 transcription in activated gametocytes . DNA replication and ultra structural analyses of cdc20 and map2 mutants showed similar blockage of nuclear division at the nuclear spindle/kinetochore stage . CDC20 was phosphorylated in asexual and sexual stages , but the level of modification was higher in activated gametocytes and ookinetes . Changes in global protein phosphorylation patterns in the Δcdc20 mutant parasites were largely different from those observed in the Δmap2 mutant . This suggests that CDC20 and MAP2 are both likely to play independent but vital roles in male gametogenesis .
Progression of mitosis in the cell-cycle is dependent upon a number of complex , sequential processes that are governed by a series of essential cell cycle regulatory proteins . Anaphase and mitotic exit is regulated by the conserved multi-subunit E3 ubiquitin ligase Anaphase Promoting Complex/Cyclosome ( APC/C ) , which targets mitotic regulators such as securin and cyclin B for destruction by the 26S proteosome [1] . Two of the major regulators of APC/C activity are cell-division cycle protein 20 ( CDC20 ) ( also known as Fizzy , p55CDC or Slp1 2–4 ) and its homologue , CDC20 homologue 1 ( CDH1 – also known as Cdh1p/Hct1p , Fizzy-related , Ste9 , Srw1 or Ccs52 [5]–[8] ) . CDC20 and CDH1 are related tryptophan-aspartic acid ( WD ) -40 repeat-containing adaptor proteins , which are highly conserved throughout eukaryotic evolution . They consist of approximately 40 amino acid-repeat motifs that often contain a C-terminal Trp-Asp ( WD ) sequence , as well as an N-terminal C-Box motif and C-terminal Ile-Arg ( IR ) residues , along with a KEN-box , Mad2-interacting motif ( MIM ) and a CRY-box [9] . CDC20 protein accumulates during S-phase , peaks in mitosis and activates the phosphorylated APC/C complex ( which is phosphorylated by cyclin-dependent kinase 1 ( CDK1 ) and other mitotic kinases [10] , [11] ) by physical association , which results in the activation of the metaphase-anaphase transition [12] and degradation of mitotic cyclins via ubiquitination [13] . Phosphorylation of APC/CCDC20 and high levels of CDK prevent CDH1 interacting with APC/C during early mitosis [14] , whereas a reduction in CDK levels by the activity of APC/CCDC20 during telophase/G1 results in CDH1 maintaining APC/C activity and cyclin degradation in proliferating cells [6] and exit from mitosis . This shows that activity of CDC20 and CDH1 in the cell cycle is temporally controlled and ensures that exit from mitosis does not occur before sister chromatid separation has been initiated . The activity of APC/CCDC20 is tightly regulated by a surveillance mechanism known as the spindle assembly checkpoint ( SAC ) [15] . The SAC is a pathway that prevents the unregulated separation of sister chromatids [16] and consists of a number of regulatory proteins including mitotic-arrest deficient ( MAD ) 1 , MAD2 , MAD3 , budding uninhibited by benzimidazoles ( BUB ) 1 and BUBR1 . The SAC negatively regulates the activity of APC/CCDC20 by preventing ubiquitination of securin and cyclin B and subsequently prolongs prometaphase until all chromosomes have been correctly oriented [15] . This process occurs at the kinetochores , where MAD2 and BUBR1 interact with APC/CCDC20 to form a mitotic checkpoint complex ( MCC ) [17] , which inhibits its activity . Once the chromatids are correctly oriented , APC/CCDC20 becomes active as it is released from the MCC and initiates anaphase , degrading securin and cyclin B and resulting in reduced CDK activity . This reduction in kinase activity promotes the formation of APC/CCDH1 and results in exit from mitosis via degradation of APC/CCDC20 , maintaining cyclin degradation in G1 prior to a new round of DNA replication [18] , [19] . Regulation of the cell-cycle and DNA replication in the unicellular apicocomplexan malaria parasite , Plasmodium , is known to be highly complex and dependent on the activity of a number of protein kinases [20] . Plasmodium is a haploid organism lacking sex chromosomes but with a complex life-cycle involving both asexual and sexual processes . Asexual multiplication occurs at three particular stages of the parasites life-cycle: blood stage schizogony , sporogony in the mosquito and pre-erythrocytic schizogony in liver hepatocytes [21] . As with some , but not all apicomplexan parasites multiplication involves repeated nuclear divisions before daughter formation by a process termed schizogony . During these stages genome duplication and segregation is accomplished using an intra-nuclear spindle while retaining an intact nuclear membrane without the formation of the typical morphological features of mitosis [22] , [23] . In contrast , DNA replication during Plasmodium sexual stages within male gametocytes occurs in the mosquito vector and involves three rounds of genomic replication resulting in eight microgamete nuclei and ultimately eight microgametes [24]–[27] . Upon ingestion of a blood meal by the female Anopheles mosquito , exposure of the male gametocyte to a slight drop in temperature , a rise in intracellular Ca2+ concentration and the mosquito-derived metabolic intermediate xanthurenic acid [28]–[30] result in rapid DNA replication ( within 12 min ) and mitosis giving rise to eight gametes , which egress out of the cell in a process termed exflagellation . This process is known to be dependent upon two protein kinases – calcium-dependent protein kinase 4 ( CDPK4 ) and mitogen-activated protein kinase 2 ( MAP2 ) [20] , [28] , [31]–[33] . Activation of CDPK4 results in genome replication , mitosis and axoneme assembly [28] and in a subsequent step; MAP2 is activated and results in axoneme motility and cytokinesis [32] . However , the cell division cycle proteins that interact with these kinases are unknown . As described earlier , in human and yeast cells CDC20 and CDH1 are known to play a major part in cell cycle regulation [9] particularly during early mitosis , and interact with regulatory kinases and phosphatases [3] , [7] , [34] . To examine the function of a single homologue of CDC20/CDH1 ( termed CDC20 for simplicity ) in the complex life-cycle of Plasmodium we used a rodent malaria model , P . berghei ( Pb ) in laboratory mice , which is very amenable to analysis by reverse genetics and where the entire life cycle , including within the mosquito vector , can be analysed . The results presented here suggest that CDC20 has an essential role in Plasmodium male gamete formation , possibly through interacting with the kinase regulator MAP2 , but has no essential involvement in asexual multiplication .
Sequence analyses of P . berghei identified a cdc20 gene ( PBANKA_051060 ) comprised of one exon . The protein contains a classical KEN-box , RVL-cyclin binding motif , IR motif and seven WD-40 repeat motifs as found in CDC20 and CDH1 of other organisms ( Figure 1A ) , but does not contain a C-box , D-box or a Mad2-interacting motif . We were only able to identify a single CDC20/CDH1 homologue coded in the genomes of Plasmodium species , which has also been suggested for Trypanosomatidae [35] . To assess the evolutionary relationships between these CDC20 homologues we aligned ( using ClustalW – Figure S1 ) the WD domains and used the alignment to draw a phylogenetic tree using the meiotic APC/C activator from yeast as an out group ( Figure 1B ) . In the resulting tree we see four clusters . Two clusters , as expected , represent the CDC20 and CDH1 homologues from a range of eukaryotic species . Another cluster contains the CDC20/CDH1 homologues from Trypanosomatidae species . The final cluster includes all the CDC20/CDH1 homologues from Plasmodium species . These results suggest that Plasmodium species contain only a single CDC20/CDH1 homologue and that the Plasmodium APC/C has only one regulator . Little information is available regarding CDC20 expression and localisation in the malaria parasite in both vertebrate and mosquito hosts . Therefore , we generated a C-terminal green fluorescent protein ( GFP ) fusion protein from endogenous cdc20 using a single crossover recombination strategy ( Figure S2A–D ) . Correct targeting was confirmed using integration PCR and Southern blot ( Figure S2B , C ) . Expression of CDC20-GFP in transgenic parasites was confirmed by Western blotting using an anti-GFP polyclonal antibody ( Figure S2D ) . A protein band of ∼92 kDa was present for all analysed CDC20-GFP samples , which corresponds to the predicted mass of the CDC20-GFP fusion protein ( 92 . 4 kDa ) . The line expressing the unfused GFP [36] produced a band at 29 kDa and was used as a control ( Figure S2D ) . Expression of the CDC20-GFP fusion protein resulted in no visible abnormalities . Low intensity CDC20-GFP expression was detected during all stages of the life-cycle ( data not shown ) apart from activated male gametocytes , which had the highest intensity of GFP expression that co-localized with Hoechst nuclear staining ( Figure 2 ) . We also generated a parasite line whereby CDC20-GFP was expressed episomally ( using the same plasmid utilised to target the endogenous locus ) under the control of the cdc20 promoter . This line showed high GFP fluorescence intensity at all stages , which co-localised with Hoechst nuclear staining in asexual , gametocyte and oocyst stages and an additional cytoplasmic localisation in ookinetes ( Figure S3 ) . To discover the function of CDC20 in the Plasmodium life-cycle , we used a double crossover homologous recombination strategy to knockout the gene . This was achieved by replacing the endogenous gene with a pyrimethamine resistant allele of the dihydrofolate reductase-thymidine synthetase ( dhfr/ts ) gene from Toxoplasma gondii ( Figure S2E ) . Successful integration of the gene was confirmed by a diagnostic PCR across the junction of the expected integration site , as well as by Southern blot , pulsed-field gel electrophoresis ( PFGE ) and quantitative reverse transcription PCR ( qRT-PCR ) to indicate an absence of transcription ( Figure S2F–I ) . Analysis of two cdc20 deletion mutant clones , N10 cl7 and N10 cl9 ( henceforward called Δcdc20 ) , showed no developmental abnormalities during asexual proliferation or gametocyte formation , as assessed on blood smears ( data not shown ) . However , in vitro cultures analysed for differentiation into ookinete stages [20] , [37] showed complete ablation of ookinete development ( Figure 3A , B ) . To ascertain whether the block in ookinete formation was a defect along the male or female line , we performed genetic crosses as previously described [37] , [38] . Crossing of Δcdc20 with a cdpk4 deletion mutant ( a previously characterised male mutant [28] , henceforward called Δcdpk4 ) showed no rescue of the phenotype . Conversely , crossing with a nek4 deletion mutant ( a previously characterised female mutant [38] , henceforward called Δnek4 ) resulted in 36% ookinete formation ( Figure 3C ) . These data prove that Δcdc20 parasites are defective along the male line . As a result of this observation , we analysed exflagellation of the activated male gametocytes [26] , which was completely blocked in Δcdc20 parasites . To substantiate the in vitro findings , we fed mosquitoes on mice infected with either wild-type or Δcdc20 parasites and analysed oocyst development . Wild type parasites developed normally and oocysts were detected in the mosquito gut , whereas no oocysts were found in the guts of mosquitoes fed on Δcdc20 parasites and analysed 14 or 21 days after feeding ( Figure 3D ) . This result confirms that CDC20 is vital to male gamete development and that fertilization/zygote formation/ookinete development is completely blocked in the Δcdc20 parasites , preventing oocyst formation . Exflagellation of the activated microgametocyte proceeds via a number of sequential steps prior to the formation of male gametes [24] , [25] . These steps are dependent upon two protein kinases; calcium-dependent protein kinase 4 ( CDPK4 ) , which is involved in cell-cycle progression to S phase and mitogen-activated kinase 2 ( MAP2 ) , which is essential for replication and mitosis to be completed before cytokinesis commences [20] , [28] , [31]–[33] . Both of these kinases have previously been shown to be essential for male gamete development and the exflagellation process [28] , [32] . As the cdc20 deletion mutant line shows a similar phenotype , we decided to analyse mRNA expression of cdc20 , map2 ( PBANKA_093370 ) and cdpk4 ( PBANKA_061520 ) in our Δcdc20 line as well as the previously characterised Δmap2 and Δcdpk4 mutant lines . Transcription of these three genes in total asexual blood , schizont and gametocyte stages of wild type parasites showed a similar profile , with highest mRNA levels found in gametocytes ( Figure 3E ) . When compared to wild-type , expression of both map2 and cdpk4 was not significantly altered at any stage in the Δcdc20 mutant; however , striking differences were found in cdc20 mRNA levels in both the Δcdpk4 and Δmap2 mutants . cdc20 was found to be significantly down-regulated in Δcdpk4 asexual blood and schizont stages ( p = 0 . 037 and 0 . 009 respectively ) . In contrast , expression in activated Δcdpk4 activated gametocytes was significantly up-regulated ( p = 0 . 001 ) , but was not altered in non-activated blood stage gametocytes . The greatest change in cdc20 expression was observed in both non-activated and activated gametocyte stages of the Δmap2 parasites , where expression was significantly up-regulated ( p = <0 . 001 and 0 . 001 respectively ) . Expression of map2 in the Δcdpk4 line was shown to be significantly down-regulated in schizont and non-activated and activated gametocyte stages ( p = <0 . 01 for all ) , whereas no significant alteration in cdpk4 was observed at any stage of the Δmap2 parasites analysed ( Figure 3F ) . Due to the ablation of exflagellation in the Δcdc20 line and the significant alteration in cdc20 expression in the Δmap2 line , we analysed axoneme formation and DNA replication in both mutants by direct immunofluorescence and fluorometric estimation of DNA content respectively . Staining of α-tubulin in both Δcdc20 and Δmap2 lines revealed normal formation of axonemes and their characteristic circling of the nucleus by the axonemes in concentric rings 8 min post activation ( mpa ) ( Figure 4A ) . However , differentiation and shortening of the spindle microtubules did not occur in either mutant 15 mpa . Furthermore , nuclear DNA in the enlarged nucleus of activated microgametocytes remained uncondensed in both mutants at 15 mpa; whereas wild-type microgametocytes had started to undergo exflagellation and nuclear divison resulting in the release of normal microgametes containing haploid nuclei with condensed DNA . These observations suggest that development of mutant microgametocytes after activation is blocked at a very late stage , possibly after the third round of DNA replication . To test whether mutant microgametocyte development was blocked after completing the three rounds of DNA replication , we analysed DNA replication by determination of the DNA content of activated microgametocytes by fluorescence microscopy and by FACS after staining with the DNA-specific dyes 4 , 6-diamidino-2-phenylindole ( DAPI ) and Hoechst 33258 , respectively . The DNA content of activated microgametocytes at 8 mpa , as determined by fluorescence microscopy , was similar in wild-type and mutant parasites , with nuclei of mutant parasites also increasing their DNA content to the octoploid level at 8 mpa ( Figure 4A , B; upper and middle panels ) . At 15 mpa the activated Δcdc20 and Δmap2 microgametocytes still contained a single enlarged nucleus with octoploid DNA content , but in contrast , in wild type microgametocytes nuclear division and gamete formation resulted in the formation of gametes with haploid DNA content ( Figure 4A , B; lower panels ) . Genome replication in activated microgametocytes was confirmed using FACS analysis of purified gametocytes that were stained with Hoechst 33258 . At 8 mpa both Δcdc20 and Δmap2 microgametocytes showed strongly increased DNA content similar to that of wild-type parasites ( Figure 4C , D ) . Purified gametocytes of the previously characterised Δcdpk4 parasite line [28] were used as a control and did not undergo DNA replication . Together these results suggest that CDC20 acts downstream of CDPK4 and has an essential role in axoneme motility , DNA condensation and cytokinesis , similar to MAP2 [32] , but does not play a role in activation of genome replication . Due to the similar morphology and dynamics of DNA replication of cdc20 and map2 mutants as analysed by direct immunofluorescence and DNA content analysis , respectively , we next examined whether deletion of the endogenous cdc20 locus resulted in structural defects that were similar to those associated with the map2 mutant line by electron microscopy . Ultrastructure analyses were performed on wild-type , Δcdc20 and Δmap2 gametocytes at 15 and 30 mins after activation . The appearance of the cytoplasm was similar for all three lines with the formation of a number of axonemes ( Figure 5A , B , C ) . The microgametocyte nucleus also appeared similar in all three strains with homogeneous electron lucent nucleoplasm and the formation of nuclear poles with radiating microtubules and attached kinetochores ( Figure 5B , C , E , and F ) . However , only in the wild-type was it possible to observe nuclear poles with condensed chromatin consistent with later stages in microgamete nucleus formation ( Figure 5A , D ) , whereas the mutants showed defects in chromosome condensation . To identify any quantitative differences , 100 parasites of each line and time point were examined with nuclear appearance divided into four categories . When the number of microgametocytes displaying the various nuclear appearances was counted , while the mutants appeared similar , significant differences were observed between the mutants and the wild type ( Table 1 ) . It was observed that wild-type parasites exhibited all stages of microgamete nuclear development with reduced numbers of the early stages with tubules and kinetochores ( 21% compared to 10% ) and increased numbers of the later stages with condensed chromatin ( 42% compared to 29% ) at 30 mins compared to 15 mins post-activation ( Table 1 ) . In both mutants approximately half the nuclei exhibited early stage microtubules and attached kinetochores ( 49–60% ) and parasites with dense nuclear pole or chromatin condensation ( 0–3% ) were rarely observed irrespective of the time point ( Table 1 ) . While no structural abnormality was observed , the quantitative differences are consistent with the two mutants being “frozen” at the nuclear spindle kinetochore formation stage . Furthermore , chromosome condensation was not observed in either cdc20 or map2 mutants as compared to wild type parasites . Reversible phosphorylation is an important regulatory mechanism in mitotic progression . In human and yeast cells , phosphorylation of CDC20 is known to be an essential step during anaphase and early mitosis [11] , [39] . As exflagellation of Δcdc20 parasites is completely ablated , but DNA replication and axoneme motility in activated microgametocytes was indistinguishable from wild-type parasites , we hypothesised that phosphorylation of CDC20 could be a vital regulator of Plasmodium gametogenesis . Analysis of CDC20 phosphorylation was performed before , during and after completion of microgametogenesis ( i . e . in schizont , activated gametocyte and ookinete stages respectively ) in CDC20-GFP parasites metabolically labelled with 32P-orthophosphate [40] and immunoprecipitated by GFP-trap . 32P-orthophosphate labelling in whole cell lysates of schizonts , activated gametocytes and ookinetes showed similar profiles and confirmed efficient uptake of 32P-orthophosphate in all stages . Autoradiography showed that CDC20 is phosphorylated at all three stages ( Figure 6A ) but phosphorylation levels were higher in activated gametocytes and ookinetes compared to schizonts ( 1 . 70 and 2 . 48 times higher respectively ) ( Figure 6A , B ) . The GFP-tagged CDC20 protein appeared as a doublet by Western Blot in schizonts ( Figure 6B ) , whereas only a single band was detected on the corresponding autoradiograph ( Figure 6A ) , suggesting that the upper band on the Western Blot may represent a phosphorylated form of CDC20-GFP and the lower band a non-phosphorylated form of the protein . Interestingly , in activated gametocytes and ookinetes , only the upper GFP-immunoreactive band is present , which may reflect a higher degree of phosphorylation of CDC20-GFP in sexual stages compared to schizonts . In order to examine whether or not CDC20 has a role in pathways of protein phosphorylation similar to those of the kinase MAP2 , we compared the global phosphorylation profile of wild type activated gametocytes with that of Δcdc20 and Δmap2 lines using metabolic labelling with 32P-orthophosphate [40] . This approach employs metabolic labelling of parasites followed by fractionation by ion exchange chromatography . The experiment was performed in triplicate and in each experiment 20 fractions were collected , resolved by SDS-PAGE and an autoradiograph obtained for seven of them to reveal the phosphorylation profile . Shown in Figure 7 are three fractions from the ion exchange fractionation where differences in the phosphorylation profile between the wild type and mutant parasite strains were observed . Importantly , the Coomassie blue stain of the SDS-PAGE gels demonstrated that the overall protein expression profiles of the wild type and mutant parasite lines were very similar ( Figure 7 ) . Despite this similarity , there were clear differences in the phosphorylation profile between the parasite lines . The phosphorylated band labelled A in Figure 7 was significantly decreased in the Δmap2 mutant , whereas the Δcdc20 mutant showed increased phosphorylation . Bands C , D , F , G , H and J showed altered phosphorylation status only in the Δmap2 mutant , whereas bands A , E and I were changed only in the Δcdc20 mutant . Only one band ( band B ) showed a similar change in both the Δmap2 and Δcdc20 mutants . This analysis indicated that although the phosphorylation profile of the parasite was affected by the deletion of map2 and cdc20 , the proteins that showed a change in phosphorylation status in the two mutant lines were ( with the exception of one protein ) different . It seems unlikely therefore that MAP2 and CDC20 regulate the same network of phospho-proteins . We are currently investigating this result further using mass spectrometry-based phosphoproteomic approaches .
Mechanisms to control cell division and the cell cycle are essential parts of the cell regulation machinery . These processes are not well understood in unicellular protozoa such as the malaria parasite Plasmodium . Plasmodium undergoes two distinct mitotic processes; one involving repeated DNA duplication , in which karyokinesis occurs after each replication and is associated with asexual proliferation and the other involving endoreduplication , with three rounds of replication prior to the simultaneous formation of eight microgamete nuclei during microgametogenesis . Here , we describe a CDC20/CDH1 orthologue in Plasmodium as an important regulator of mitosis during male gametogenesis , but interestingly it has no effect on the mitotic process undergone during schizogony . Our bioinformatic studies suggest that in Plasmodium there is only one gene representing CDC20 and its homologue CDH1 , and that the protein is a true structural homologue of CDC20/CDH1 , even though we could not complement CDC20 function in yeast ( data not shown ) . Although we cannot exclude the possibility that we failed to detect a second highly spliced Plasmodium cdc20/cdh1 homologue , the phylogenetic clustering of all the Plasmodium CDC20 homologues gives confidence that there is only a single CDC20 orthologue in Plasmodium species . This suggests that Plasmodium diverged from other eukaryotes prior to the duplication event that presumably gave rise to CDC20 and CDH1 genes . It is interesting to note that the Plasmodium cluster is distinct from the Trypanosomatidae cluster where there is also a single corresponding gene in each genome . Furthermore , this orthologue has a classical KEN box-like domain at the N-terminus and an RVL domain and IR motif at the C-terminus , all of which are required for cyclin degradation and binding to the APC/C core [7] . The presence of these domains suggests that CDC20 in Plasmodium could influence the cell cycle in a similar manner to other systems , such as yeast , mammals and plants [9] , [41] . The lack of a D-box and presence of a KEN-box are consistent with the structure of CDC20 in humans , with the presence of a KEN-box suggesting that Plasmodium CDC20 is a prime target for ubiquitination , as suggested in a recent study [42] . Alternatively , as Plasmodium CDC20 is the only orthologue of both CDC20 and CDH1 present in other systems , it is plausible that ubiquitination of CDC20 in Plasmodium is self-regulating , as CDC20 is known to be degraded by APC/CCDH1 via its KEN-box [43] and could therefore act as a “negative feedback” mechanism as seen in human cells [44] , [45] . The seven conserved WD repeats in the Plasmodium CDC20 protein also suggests that it does bind an as yet unknown multi-protein complex . Plasmodium CDC20 shows some differences from the ccs52 homologue reported in plants , such as Medicago sativa [8] , since it lacks a MAD-binding box and also the D-box that appears to be specific for CDH1 and is not conserved in CDC20 and FZY proteins . It has been reported recently that in Arabidopsis thaliana there are five isoforms of CDC20 , and two of them are functional [41] . We did not observe any such expansion of genes for this protein in Plasmodium . Our CDC20-GFP expression studies showed that CDC20 is highly expressed in activated male gametocytes ( with gametocytes showing highest expression at the mRNA level , in agreement with previous transcriptomic studies [46] ) but it is also present throughout the life-cycle and located mainly in the nuclear compartment , with some cytoplasmic localisation , consistent with expression in other systems [47] , [48] . However , although previous studies have shown cdc20 transcripts and protein to be highly expressed in sporozoites of P . falciparum [46] , [49] , we did not observe high protein expression levels of CDC20-GFP in sporozoites . Functional studies using a gene deletion strategy showed that CDC20 controls male gamete development and deletion mutants are impaired during transmission of the parasite to the mosquito vector . Further in-depth analysis of these mutants using a cross fertilisation approach showed that this defect is limited to male gamete differentiation ( exflagellation ) and formation since Δcdc20 macrogametocytes are fully capable of cross fertilization with microgametes from donor strains . Hence , CDC20 has an essential function for the transition of male gametocytes to gametes . Gametogenesis in Plasmodium involves three rounds of mitotic division in male gametocytes resulting in eight gametes [24]–[26] , [50] . We have previously shown that CDPK4 is involved in cell cycle progression to S phase and MAP2 may be essential for replication and mitosis to be completed before cytokinesis commences [28] , [32] , [33] ( although it is important to note that MAP2 is essential for asexual development in P . falciparum [51] , so there may be species-specific differences in the roles of different kinases ) . As cdc20 mRNA levels are up regulated in both Δcdpk4 and Δmap2 mutants , this suggests that CDC20 may be interlinked with these kinases and orchestrates the process of male gametogenesis and is perhaps up-regulated to compensate for the loss of these two kinases , but this suggestion requires further investigation . The cdc20 deletion mutants formed axonemes and mitotic spindles but failed to undergo karyokinesis or cytokinesis and also did not form motile , flagellar gametes , a phenotype similar to what we have observed with map2 deletion mutants . The requirement for CDC20 during karyokinesis is consistent with the known function of CDC20 and CDH1 in other systems [7] . As described earlier , CDC20 is active during early mitosis in other cells and its up-regulation in gametocytes suggests that it has an essential role in the multiple rounds of DNA replication and the chromosome separation specifically associated with this process . However , mutant cdc20 parasites do not arrest during asexual proliferation and this suggests that Plasmodium CDC20 is specifically required for microgametogenesis . Functional studies in human systems have shown that a deficiency of CDH1 results in delayed mitotic exit as well as an accumulation of mitotic errors and difficulty in completion of cytokinesis [52] , [53] , similar to what is observed in our cdc20 and map2 mutants . Therefore we suggest that CDC20 in Plasmodium fulfils the function of both CDC20 and CDH1 . Moreover , loss of cdc20 results in arrest during metaphase to anaphase transition [12] , [54] , [55] , with sister chromatids failing to form . How the single CDC20 protein may fulfil the roles of both CDC20 and CDH1 requires further investigation . Our ultrastructure studies for both Δcdc20 and Δmap2 lines , reported for the first time to our knowledge; show that these mutants have a similar arrest in cytokinesis and karyokinesis detected by EM , with defects in nuclear spindle/kinetochore movement and chromatin condensation , confirming our initial light microscopy observation of Δmap2 . Unlike the Δmap2 line , we never observed any exflagellation in the Δcdc20 line . As suggested before [56] , classical spindle checkpoints are not present in Plasmodium since blockage of microtubule organisation does not appear to block DNA synthesis . Therefore , MAP2 and CDC20 may be involved in a critical cell cycle checkpoint during microgametogenesis that controls DNA replication and mitosis , prior to karyokinesis and cytokinesis and is summarised in Figure 8 . DNA replication in the Δcdc20 line was similar to that in the Δmap2 line , with both mutants undergoing octoploidy 8 mpa , but not undergoing karyokinesis . As a result , we analysed whether phosphorylation of CDC20 could be involved in mitotic progression during microgametogenesis . In other systems , phosphorylation of CDC20 can be achieved by BUB1 , CDK1 , MAPK [57] , [58] and also NEK2 [59] , another protein kinase required for zygote development in Plasmodium [60] and this modification is an essential step for CDC20 inhibition by the SAC [61] , [62] . Here , we have shown that CDC20 is more phosphorylated in activated gametocytes and ookinetes ( i . e . sexual stages ) compared to schizont ( asexual stages ) , which suggests that phosphorylation of CDC20 may be a possible mechanism involved in gametogenesis . Interestingly , the global phosphorylation profile of Δcdc20 parasites suggests that CDC20 regulates the phosphorylation of specific proteins within the gametocytes . The proteins that are regulated by CDC20 are; however , largely different from those that appear to be regulated by MAP2 . This would suggest that at the level of phosphorylation CDC20 and MAP2 regulate different pathways . It would be interesting in future studies to dissect out the proteins regulated by MAP2 and CDC20 and in this way build a network of phospho-proteins that regulate male gametogenesis . In conclusion , this study identified significant differences in the control of mitosis during asexual development compared to microgametogenesis in the malaria parasite . We have also shown that CDC20 and MAP2 may play independent but essential roles in the mitotic division associated with microgametogenesis but are not essential for mitosis during asexual stages in the malaria parasite .
All animal work has passed an ethical review process and was approved by the United Kingdom Home Office . Work was carried out in accordance with the United Kingdom ‘Animals ( Scientific Procedures ) Act 1986’ and in compliance with ‘European Directive 86/609/EEC’ for the protection of animals used for experimental purposes . The permit number for the project licence is 40/3344 . Either Tuck's Original ( TO ) ( Harlan ) or CD1 ( CRUK ) outbred mice were used for all experiments . The targeting vector for cdc20 was constructed using the pBS-DHFR cassette , in which polylinker sites flank a Toxoplasma gondii dhfr/ts expression cassette conveying resistance to pyrimethamine . PCR primers N10-1 ( 5′-CCCCGGGCCCGAGCTGTCTACTGCTCTGGTAAAGCC-3′ ) and N10-2 ( 5′-GGGGAAGCTTCATTATTCTGGATCATAGCTCTC-3′ ) were used to generate a 452 base pair ( bp ) fragment 5′ upstream sequence of Pbcdc20 from genomic DNA , which was inserted into ApaI and HindIII restriction sites upstream of the dhfr/ts cassette of pBS-DHFR . A 579 bp fragment generated with primers N10-3 ( 5′-CCCCGAATTCGGAACTTCTCTTGTTTCTGGATCTCC-3′ ) and N10-4 ( 5′-GGGGTCTAGAGCATGCTAATTAGCTTCACATCCG-3′ ) from the 3′ flanking region of Pbcdc20 was then inserted downstream of the dhfr/ts cassette using EcoRI and XbaI restriction sites . The linear targeting sequence was released using ApaI/XbaI . For GFP-tagging by single homologous recombination and generation of the plasmid for episomal expression , a 2435 bp region of Pbcdc20 starting 812 bp upstream of the start codon and omitting the stop codon was amplified using primers T36-1 ( 5′-CCCCGGTACCCTTATTTATGAAAACGATTATAAGG-3′ ) and T36-2 ( 5′-CCCCGGGCCCCCTGATTATTTCATAATAATTTTCAAAGGG-3′ ) , producing an amplicon 2435 bp in length . This was then inserted upstream of the gfp sequence in the p277 vector using KpnI and ApaI restriction sites . The p277 vector contains the human dhfr cassette , also conveying resistance to pyrimethamine . Before transfection , the sequence was linearised using HindIII and P . berghei ANKA line 2 . 34 was then transfected by electroporation [36] . Briefly , electroporated parasites were mixed immediately with 200 µl of reticulocyte-rich blood from a phenylhydrazine ( Sigma ) treated , naïve mouse and incubated at 37°C for 30 minutes and then injected intraperitoneally . From day 1 post infection pyrimethamine ( 7 mg/ml ) ( Sigma ) was supplied in the drinking water for four days . Mice were monitored for 15 days and drug selection repeated after passage to a second mouse , with resistant parasites used for cloning by limiting dilution and genotyping . Chromosomes of wild type and gene knockout parasites were separated by pulsed field gel electrophoresis ( PFGE ) on a CHEF DR III ( Bio-Rad ) using a linear ramp of 60–500 s for 72 hr at 4 V/cm . Gels were blotted and hybridized with a probe recognizing both the resistance cassette in the targeting vector and , more weakly , the 3′-untranslated region ( UTR ) of the P . berghei dhfr/ts locus on chromosome 7 . For the gene knockout parasites , two diagnostic PCR reactions were used as illustrated in Figure S2 . Primer 1 ( INT N10 , 5′-GTGTGCAATTTGGGAATTTAGCTAG-3′ ) and primer 2 ( ol248 , 5′-GATGTGTTATGTGATTAATTCATACAC-3′ ) were used to determine correct integration of the selectable marker at the targeted locus . Primers 3 ( N10 KO1 , 5′- GATAATAATTGGAATAGTCATT-3′ ) and 4 ( N10 KO2 , 5′- TTACATGTATAACTATTCCA-3′ ) were used to verify deletion of the target gene . Having confirmed integration , genomic DNA from wild type and mutant parasites was digested with HindIII and the fragments were separated on a 0 . 8% agarose gel , blotted onto a nylon membrane ( GE Healthcare ) , and probed with a PCR fragment homologous to the P . berghei genomic DNA just outside of the targeted region . For the C-fusion GFP tagging parasites , one diagnostic PCR reaction was also used as illustrated in Figure S2 . Primer 1 ( INT T36 , 5′- CATTCCAAACTAGTATTATAAAATTTGTTG -3′ ) and primer 2 ( ol492 , 5′- ACGCTGAACTTGTGGCCG-3′ ) were used to determine correct integration of the gfp sequence at the targeted locus . Having confirmed correct integration , genomic DNA from wild type and transgenic parasites was digested with EcoRI and the fragments were separated on a 0 . 8% agarose gel , blotted onto a nylon membrane , and probed with a PCR fragment homologous to the P . berghei genomic cdc20 sequence using the Amersham ECL Direct Nucleic Acid Labelling and Detection kit ( GE Healthcare ) . Parasites were also visualized on a Zeiss AxioImager M2 ( Carl Zeiss , Inc ) microscope fitted with an AxioCam ICc1 digital camera ( Carl Zeiss , Inc ) and analysed by Western blot to confirm GFP expression as described . Western blot analysis was performed on cell lysates prepared by re-suspending parasite pellets in a 1∶1 ratio of PBS containing Protease inhibitor ( Roche ) and Laemmli sample buffer , boiling and separating on a 12% SDS-polyacrylamide gel . Samples were subsequently transferred to nitrocellulose membranes ( Amersham Biosciences ) and immunoblotting performed using the Western Breeze Chemiluminescent Anti-Rabbit kit ( Invitrogen ) and anti-GFP polyclonal antibody ( Invitrogen ) , according to the manufacturer's instructions . The protein sequences of the highly conserved WD domain from CDC20 and CDH1 homologues from a range of eukaryotes were downloaded from NCBI . ClustalW was used to align the sequences and construct a phylogenetic tree . Phenotypic screening of cdc20 mutants was performed as previously described [20] , [37] . Briefly , asexual proliferation and gametocytogenesis were analysed using blood smears . Gamete activation , zygote formation and ookinete conversion rates were monitored by in vitro cultures using a marker for the surface antigen P28 as previously described [37] , [38] . Hoechst 33342 was used to stain parasite nuclei . Stained cells were analysed on a Zeiss AxioImager M2 microscope ( Carl Zeiss , Inc ) fitted with an AxioCam ICc1 digital camera ( Carl Zeiss , Inc ) . For mosquito transmission triplicate sets of 50–100 Anopheles stephensi mosquitoes were allowed to feed on anaesthetized infected mice on days 4 to 5 following blood infection for 20 min at 20°C . Guts were analysed 14 and 21 days post infection for production of oocysts and sporulating oocysts respectively . Parasite-infected blood was re-suspended in ookinete medium as previously described [32] , [60] . After 24 hours , samples were re-suspended in ookinete medium containing Hoechst DNA dye and anti-P28 Cy3-conjugated 13 . 1 antibody [32] , [37] and examined with the Zeiss AxioImager M2 microscope fitted with an AxioCam ICc1 digital camera ( Carl Zeiss , Inc ) . The percentage of ookinetes to all 13 . 1-positive cells ( unfertilised macrogametes ( round cells ) and ookinetes ) was then calculated . Gametocytes in parasite-infected blood ( as described above ) were activated in ookinete medium , resuspended in 4% paraformaldehyde ( PFA ) ( Sigma ) diluted in microtubule stabilizing buffer ( MTSB ) [32] and added to poly-L-lysine coated slides . Immunocytochemistry was performed on the fixed parasite material using primary mouse monoclonal anti-alpha tubulin antibody ( Sigma , used at 1 in 500 ) . Secondary antibody was Alexa 547 conjugated anti-mouse IgG ( Molecular probes , used at 1 in 1000 ) . The slides were then mounted in Vectashield with DAPI ( Vector Labs ) . Parasites were visualized on a Zeiss AxioImager M2 microscope ( Carl Zeiss , Inc ) fitted with an AxioCam ICc1 digital camera ( Carl Zeiss , Inc ) . To measure nuclear DNA content of activated microgametocytes by direct immunofluorescence , images of parasites fixed and stained as above were analyzed using the ImageJ software ( version 1 . 44 ) ( National Institute of Health ) as previously described [32] . To confirm nuclear DNA content of activated microgametocytes by FACS , purified gametocytes were transferred to standard ookinete culture medium for activation of gamete formation . At 8 mins after activation cells were pelleted by centrifugation ( 5 sec; 10 , 000 rpm ) , fixed in 0 . 25% glutaraldehyde/PBS solution and stained with 2 µM Hoechst-33258 . The Hoechst-fluorescence intensity ( DNA content ) of the gametocytes was analyzed by FACS using a LSR-II flow cytometer ( Becton Dickinston ) . Cells were analyzed at room temperature with the following filters ( parameters/thresholds ) : UB 440/40 ( Hoechst ) ( 400/5000 ) ; FSC ( 250/2000 ) ; SSC ( 200/5000 ) . The cells for analysis were selected on size by gating on FSC and SSC . A total of 10 , 000–500 , 000 cells were analyzed per sample and all measurements were performed on triplicate samples . To determine the Hoechst-fluorescence intensity ( DNA content ) from the populations of activated female and male gametocytes , gates were set as in [63] . Data processing and analysis was performed using the program FlowJo ( http://www . flowjo . com ) . Samples of wild type , cdc20 mutant and map2 mutant microgametocytes cultured as described above were fixed in 4% glutaraldehyde in 0 . 1 M phosphate buffer and processed for routine electron microscopy as described previously [64] . Briefly , samples were post fixed in osmium tetroxide , treated en bloc with uranyl acetate , dehydrated and embedded in Spurr's epoxy resin . Thin sections were stained with uranyl acetate and lead citrate prior to examination in a JEOL12EX electron microscope ( Jeol AB ) . Quantitation of the nuclear features observed by electron microscopy was carried out at 15 and 30 minutes . This was based on the examination of 100 microgametocytes identified by axoneme formation at each time point . The features identified were 1nuclei with no specific features in the plan of section , 2early stage exhibiting nuclear poles with spindle microtubules and kinetochores , 3mid stage with nuclear pole but no attached kinetochores , and 4late stages with the nucleus exhibiting areas of condensed chromatin . Purification of gametocytes was achieved using a modified protocol from [65] . Briefly , mice were treated by intra-peritoneal injection of 0 . 2 ml of phenylhydrazine ( 6 mg/ml ) ( Sigma ) in PBS to encourage reticulocyte formation four days prior to infection with parasites . Day four post infection ( p . i . ) mice were treated with sulfadiazine ( Sigma ) at 20 mg/L in their drinking water for two days to eliminate asexual blood stage parasites . On day six p . i . mice were bled by cardiac puncture into heparin and gametocytes separated from uninfected erythrocytes on a NycoDenz gradient made up from 48% NycoDenz ( 27 . 6% w/v NycoDenz in 5 mM Tris-HCl , pH 7 . 20 , 3 mM KCl , 0 . 3 mM EDTA ) and coelenterazine loading buffer ( CLB ) , containing PBS , 20 mM HEPES , 20 mM Glucose , 4 mM sodium bicarbonate , 1 mM EGTA , 0 . 1% w/v bovine serum albumin , pH 7 . 25 . Gametocytes were harvested from the interface and washed twice in RPMI 1640 ready for activation of gamete formation . Blood from day 5 pi mice were cultured for 24 hrs at 20°C for ookinetes as described above . Ookinetes were purified on a 63% NycoDenz gradient and harvested from the interface , washed and labelled . Parasites were purified as described and frozen in Trizol ( Sigma ) prior to RNA extraction . RNA was isolated according to manufacturer's instructions . Isolated RNA was treated with DNase I ( Promega ) and used in reverse transcription reactions ( SuperScript III Reverse Transcription kit , Invitrogen ) from 1 µg of total RNA . Gene expression was quantified by SYBR green PCR using Fast mastermix on an ABI 7500 QPCR System ( Applied Biosystems ) . Primers were designed using the PerlPrimer software program [66] to be 18–22 bps in length , with 30–60% GC content , to amplify a region 50–150 bp long and when possible , to bind within 600 bp of the 3′ end of the genes of interest . Primer efficiencies were all between 90–110% , with qRT-PCR resulting in no detectable primer dimers , as determined by dissociation curves . cDNA was diluted 1∶20 with DEPC-treated water before use . Reactions consisted of 3 . 6 µl of diluted cDNA , 5 µl SYBR green fast mastermix ( Applied Biosystems ) , 0 . 2 µl each of forward and reverse primer and 1 µl of DEPC water . Cycling conditions were: 95°C for 20 sec followed by 40 cycles of 95°C , 3 secs , and 60°C , 30 secs , followed by dissociation curve . Three biological replicates , with three technical replicates from each biological replicate were performed for each assayed gene . Endogenous gene expression was determined using the comparative cycle threshold method [67] , whereas relative quantification in mutant lines was determined using the Pfaffl method [68] . Both methods used hsp70 ( PBANKA_081890 ) ( forward , 5′-GTATTATTAATGAxACCCACCGCT-3′; reverse , 5′-GAAACATCAAATGTACCAxCCTCC-3′ ) and arginyl-tRNA synthetase ( PBANKA_143420 ) ( forward , 5′-TTGATTCATGTTGGATTTGGCT-3′; reverse , 5′-ATCCTTCTTTGCCCTTTCAG-3′ ) as reference genes . cdc20 primers were: forward , 5′-ATGTTTGGTAACTATTTGGCGG-3′; reverse , 5′-ATCCCATATTTCTACTGCACCA-3′ . map2 ( PBANKA_093370 ) : forward , 5′-AATGAAGAACCAGGGCCA-3′; reverse , 5′-ACCATCTAGTAACTACATGGCT-3′ . cdpk4 ( PBANKA_061520 ) : forward , 5′-AAATGTTGATGTACACAAGTGC-3′; reverse , 5′-ATGTTCTAATGCATCTCTCTTGCT-3′ Blood aliquots from infected mice were incubated overnight , from which schizonts and ookinetes were purified by using Nycodenz protocols as described previously [36] , [65] . Gametocytes were purified and activated for 25 min at 20°C in ookinete medium as described above . Schizonts , activated gametocytes and ookinetes were then washed in phosphate-free Kreb's buffer and metabolically labelled with 3–5MBq 32P-orthophosphate in phosphate-free Kreb's buffer for 30 min at 20°C . After two washes in phosphate-free Kreb's buffer , the labelled parasites were lysed for 30 min at 4°C in lysis buffer ( 10 mM Tris pH 7 . 5 , 150 mM NaCl , 0 . 5 mM EDTA , 0 . 5% NP-40 ) supplemented with protease and phosphatase inhibitors ( Roche ) , the resulting lysate was centrifuged at 20 , 000×g for 5 min and the supernatant collected . GFP-tagged CDC20 proteins were then immunoprecipitated using GFP-TRAP beads ( ChromoTek ) . The immunoprecipitated proteins were then resuspended in Laemmli sample buffer and separated by SDS-PAGE . 32P-labelled proteins were visualized using a phosphorimager ( Molecular Dynamics ) and GFP-tagged proteins analysed by Western Blot as described above , using an anti-GFP polyclonal antibody ( Invitrogen ) . The relative CDC20-GFP phosphorylation levels in activated gametocytes and ookinetes with respect to schizonts were obtained by taking the normalized ratio between the intensity of the phosphorylation signal from the phosphorimager and the intensity of the GFP immunoreactive signal from the corresponding Western Blot by using the ImageJ software ( National Institute of Health ) . Gametocytes from wild type , cdc20 and map2 mutant parasites were purified by using a Nycodenz protocol as described above from the blood of infected mice . Purified gametocytes were placed for 25 minutes in ookinete medium at 20°C to activate both male and female gametocytes to form gametes . For metabolic labelling , the parasites were washed once with 1 ml of phosphate-free Kreb's buffer: 118 mM NaCl , 4 . 7 mM KCl , 4 . 2 mM NaHCO3 , 1 . 2 mM MgSO4 ( 2H2O ) , 11 . 7 mM glucose , 10 mM HEPES , 1 . 3 mM CaCl2 ( 2H2O ) , pH 7 . 4 and resuspended in 500 µl of the same buffer . 20–25 µl 32P-orthophosphate ( 7–9 . 25MBq ) was added to the suspension and incubated at 37°C for 30 min . The labelled parasites were then lysed in lysis buffer: 50 mM Tris , 0 . 5 mM EDTA , 5% β-glycerolphosphate , pH 7 . 6 , supplemented with protease/phosphatase inhibitors ( Roche ) and 1% NP-40 . Following incubation on ice for 10 min , the samples were centrifuged 3 min at 20000×g and the supernatants were collected for further fractionation . Fractionation was carried out on an AKTA chromatographer ( Amersham Pharmacia Biotec ) using Resource Q ( Amersham Pharmacia Biotec ) anion-exchange column ( matrix volume 1 ml ) . The proteins were eluted using a linear gradient of 0–1 . 0 M NaCl in running buffer: 10 mM Tris , 5 mM EDTA and 20 mM β-glycerolphosphate , pH 7 . 4 . Fractions ( 1 ml ) were collected and analysed further by resolution on SDS-PAGE gels . 32P-labelled proteins were visualised by autoradiography . All statistical analyses were performed using GraphPad Prism ( GraphPad Software ) . For ookinete conversion rates , non-parametric t-tests were used . For relative quantification of qRT-PCR reactions , two-way ANOVA was performed . | Malaria parasites are single cell organisms that multiply via asexual division at different stages in the life-cycle: in the liver and red blood cells of the vertebrate host and gut of the mosquito vector . The precursor sexual stages ( male and female gametocytes ) form in red blood cells , then following ingestion in a blood meal the male gametocytes undergo three mitotic divisions resulting in eight male gametes in the mosquito gut . In many organisms including yeast and mice it has been shown that cell division and mitosis are tightly regulated by a set of cell division cycle proteins , namely CDC20 and CDH1 . We studied the function of the single homologue of CDC20/CDH1 expressed in the rodent malaria parasite , Plasmodium berghei . We found that P . berghei CDC20 is not required for asexual multiplication but is essential for male gamete formation . Analysis of these mutant parasites revealed a defect in male gametocyte division and differentiation resulting in no male gamete formation with major defects in cytokinesis . This phenotype is similar to that of a kinase mutant ( map2 ) , suggesting that they play an independent but essential role in progression of the sexual stage . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biology"
] | 2012 | A Putative Homologue of CDC20/CDH1 in the Malaria Parasite Is Essential for Male Gamete Development |
Translation of RNA to protein is a core process for any living organism . While for some steps of this process the effect on protein production is understood , a holistic understanding of translation still remains elusive . In silico modelling is a promising approach for elucidating the process of protein synthesis . Although a number of computational models of the process have been proposed , their application is limited by the assumptions they make . Ribosome profiling ( RP ) , a relatively new sequencing-based technique capable of recording snapshots of the locations of actively translating ribosomes , is a promising source of information for deriving unbiased data-driven translation models . However , quantitative analysis of RP data is challenging due to high measurement variance and the inability to discriminate between the number of ribosomes measured on a gene and their speed of translation . We propose a solution in the form of a novel multi-scale interpretation of RP data that allows for deriving models with translation dynamics extracted from the snapshots . We demonstrate the usefulness of this approach by simultaneously determining for the first time per-codon translation elongation and per-gene translation initiation rates of Saccharomyces cerevisiae from RP data for two versions of the Totally Asymmetric Exclusion Process ( TASEP ) model of translation . We do this in an unbiased fashion , by fitting the models using only RP data with a novel optimization scheme based on Monte Carlo simulation to keep the problem tractable . The fitted models match the data significantly better than existing models and their predictions show better agreement with several independent protein abundance datasets than existing models . Results additionally indicate that the tRNA pool adaptation hypothesis is incomplete , with evidence suggesting that tRNA post-transcriptional modifications and codon context may play a role in determining codon elongation rates .
The process of protein synthesis is central to all living organisms . It has been actively researched for over five decades , and by now the individual steps of this process are known in great detail at the molecular and mechanistic levels [1] . Gene adaptation to the tRNA pool , mRNA secondary structure strength , codon order and local amino acid charge were independently implicated in shaping rates of protein production [2–4] . However , many disciplines would benefit from a holistic view of how these factors collectively influence translation . In particular , in biotechnology this knowledge would allow for tuning protein expression as desired with ramifications for cost-effective production of medicines and biofuels using microbes [5] . However , owing to the biological complexity of the process and the difficulty of measuring kinetic rates of the individual steps of protein synthesis , the development of computational models that would enable such applications lagged behind . Only recently , the accumulated knowledge was integrated into several state-of-the-art models of increasing complexity . Zhang and Ignatova [6] proposed a “static” model for predicting the local speed of translation within a gene from codon-specific elongation rates derived from tRNA concentrations; their approach was extended by Reuveni et al . [7] , who suggested using a “dynamic” model in which ribosomes initiate translation at the first codon and block each other while moving towards the end of the mRNA transcript . Siwiak and Zielenkiewicz [8] and Shah et al . [9] independently proposed static and dynamic full-cell models that additionally integrated the intracellular concentrations of ribosomes , mRNA and tRNA molecules , and their diffusion inside the cell in a single model . While predictions made by these models are usually in accordance with the current understanding of translation , most of their core assumptions ( e . g . codon translation rates ) have not been subjected to comparison against measured data . Ribosome profiling ( RP ) [10 , 11] , a relatively new technique based on high-throughput sequencing of ribosome-protected RNA fragments ( footprints ) , is nowadays often employed for studying translation [12–15] . It provides noisy snapshots of the locations of actively translating ribosomes attached to mRNA transcripts , thereby convolving the number of ribosomes and their speed of translation ( a few stalled ribosomes can generate similar sets of footprints as many ribosomes involved in rapid translation ) . While in principle these data allow for simultaneously reasoning about ribosome counts and their local speed , such analysis is hampered by the limited understanding of the error model and biases of RP data [16] . To date RP measurements have been analyzed either at the level of full genes [8 , 9] or at single codon resolution [4 , 17] . While only the latter allows for analyzing the dynamics of translation , it is not clear whether codon-resolution measurements are sufficiently reliable for such quantitative analysis ( see S1 Text ) . To overcome the measurement reliability issue several studies [18–20] performed “meta-codon” analysis by pooling observations from different occurrences of a particular codon together to produce an estimate of the codon elongation time . It is unclear , however , to what extent such estimates are affected by ribosomal interference . We propose a set of methods for deriving full translation kinetics of an organism from RP data ( see Fig 1 ) . Our approach is conceptually similar to Ciandrini et al . [21] , who inferred translation initiation rates of yeast genes from polysome profiling data , except that we use RP for deriving these rates and additionally determine the translation elongation rates . The method is based on a novel “segment tree” multi-scale interpretation of the RP data that captures ribosome translation dynamics along mRNAs without sacrificing reliability due to measurement noise . We use this interpretation to simultaneously extract , for the first time , per-gene translation initiation rates and per-codon translation elongation rates for the bakers yeast Saccharomyces cerevisiae by fitting two version of the TASEP ( Totally Asymmetric Exclusion Process ) , a simple dynamic model of translation [22] , on the segment tree estimates . To make fitting tractable , we devised a highly efficient initiation rate approximation scheme and combined it with a novel Monte Carlo simulation strategy inside an evolutionary optimization algorithm . Fitted TASEP models match the RP data significantly better than the state-of-the-art models , and their predicted protein production rates are confirmed by several independent protein abundance ( PA ) datasets . In particular our models show significantly better agreement with PA than existing models when the measurements are corrected for mRNA levels , i . e . when only the effect of translation on protein levels is considered . Interestingly , the fitted codon elongation rates deviate significantly from the tRNA pool adaptation hypothesis .
RP data for yeast Saccharomyces cerevisiae strain S288C [23] containing ribosome footprint reads ( ribo-seq ) and fragmented mRNA reads ( RNA-seq ) measured in duplicate were obtained from the NCBI Short Read Archive ( accession SRP028552 ) . Reads were trimmed and mapped to the latest S . cerevisiae strain S288C reference genome taken from the Saccharomyces Genome Database ( SGD , Cherry et al . [24] ) in two stages , and assigned to gene coding sequences ( CDSes ) obtained from SGD . Aligned ribosome footprint and mRNA reads were assigned to single positions within the CDSes based on respectively their inferred A-sites or the centre position of the read ( see S1 Text for details ) . To obtain a high-resolution map of mRNA and ribosome density without sacrificing measurement accuracy , for each gene we construct a segment tree of density measurements from nested parts of the CDSes ( Fig 1 , left ) . By pooling reads from all segment positions , average densities per segment can be calculated more reliably than would be possible at single codon resolution ( see also S1 Text ) , while recording these densities for nested segments of decreasing lengths allows for indirectly capturing the change in density along a transcript . Starting from an initial segment [l , r] equivalent to the complete CDS we count the number of ribo-seq reads R[l , r] and RNA-seq reads M[l , r] assigned to this segment . These counts are normalized by the total number of ribo- and RNA-seq reads aligned to all CDSes ( NR and NM respectively ) and the segment length L[l , r] = r − l + 1 to obtain ribosome and mRNA densities d [ l , r ]Ribo = R [ l , r ] L [ l , r ] N R and d [ l , r ]mRNA = M [ l , r ] L [ l , r ] N M for the current segment . To obtain the sought per transcript ribosome density ( later referred to as density ratio ) the ratio of the two measurements ω [ l , r ] = d [ l , r ]Ribo d [ l , r ]mRNA is calculated . The average segment ribosome density given by this ratio is normalized for transcript abundance and allows for directly comparing segments from different genes to each other . A cut point p is then chosen and the process is repeated recursively for segments [l , p] and [p + 1 , r] ( see Fig 1 , left ) . The aim behind calculating d [ l , r ]mRNA for each segment independently instead of estimating a single gene-specific value is to remove any local sequencing bias ( presumed to be identical between RNA- and ribo-seq since very similar protocols are used for library preparation [23] ) from the ratio estimates . Density measurements are computed for each replicate individually , but the same segment cut points are used in order to merge replicates later . Cut points are chosen such that the combined number of RNA- and ribo-seq reads across replicates is divided equally between the left and the right segments ( see S1 Text for details ) . The recursive tree construction continues while there are sufficient reads for making reliable density estimates ( at least 128 reads in the two replicates summed together for RNA-seq and ribo-seq , separately; see S1 Text for details on choosing these thresholds ) and segment length is large enough , L[l , r] ≥ 20 codons . The segment length cutoff aims at keeping the segments long enough to average out any measurement error due to incorrect read assignment or sequence bias . Prior to interpreting the measurements , we additionally remove a systematic density-dependent bias present in the density and ratio measurements using the available replicate information ( see S1 Text ) . This procedure was used to construct segment trees for 4 , 892 genes with a total of 60 , 466 nested density estimates left after removing genes classified as dubious or located on the mitochondrial chromosome . In order to accurately capture variance of RP data , we assume that the measured segment densities follow a log-normal distribution around the density values . A similar assumption is often made for transcriptome measurements and is justified by the observation that inter-replicate errors ( i . r . e . ) , i . e . the ratios of replicated mRNA and ribosome density measurements , follow a log-normal distribution ( S1 Fig and Ingolia et al . [10] ) . It then holds that density ratios ω[lj , rj] ( j ∈ Jg , where Jg is the set of all segments of gene g ) from different replicates also follow a log-normal distribution ln𝓝 ( μj , σj ) as ratios of log-normally distributed random variables—the mRNA and ribosome segment densities . Here μj and σj are used as shorthands for μ[lj , rj] and σ[lj , rj] respectively . To determine the parameters of this distribution we estimate μj for the j-th segment from the available replicated measurements as the log of their geometrical mean . Ideally , a separate shape parameter σj should also be estimated per segment , but , given the number of replicates , doing so would not yield reliable estimates . Instead it was chosen to group segments from all genes based on their length , and estimate shape parameters σ kgroup for group k from the i . r . e . of measurements from that group ( see S1 Text and S2 Fig ) . The proposed measurement distribution ln 𝓝 ( μ j , σ k jgroup ) , where kj denotes the length group of the j-th segment , is used throughout this paper as an error model for fitting TASEP models of translation on RP data and for comparing different models with the data . Computational models of translation typically provide the ability to extract steady-state codon occupancy probabilities obtained from model simulations , i . e . estimates of the chance that a particular position of an mRNA is occupied by an actively translating ribosome . Similar to the ribosome profiling measurements these occupancy profiles are determined by the local speed of translation and the number of ribosomes translating an mRNA . This allows for evaluating how well a given model matches the RP data by comparing the average segment occupancies and the segment tree ratio estimates ( see Fig 1 , right ) . Quantitative measurements obtained via high-throughput sequencing such as the mRNA and ribosome densities ( and hence their ratios ) are measured in arbitrary units . Without explicit assumptions on the physiological characteristics of the analyzed organism , such as the full size of its transcriptome [8] or the number of ribosomes per cell [9] , and on the efficiency of individual experimental steps , it is impossible to estimate sequencing depth of the RP measurements ( i . e . the average number of reads per ribosome or the average number of reads per kilobase of transcript ) and therefore impossible to express the measured values in physiologically meaningful units ( e . g . number of ribosomes per transcript ) . Additionally , this unit mismatch complicates the comparison of modeled ribosome occupancies to the measured densities . To derive a model evaluation criterion , we first assume that an unknown scaling factor C that transforms model output into measurement data units is given , and propose a method for calculating it later . Let n ig be the model-predicted ribosome occupancy at position i of gene g and T g = { ( μ jg , σ jg ) ∣ j ∈ J g } be the set of ratio distribution parameters for segments [ l jg , r jg ] . Here the upper index g denotes the gene , and for a more succinct notation we use the lower index j in place of [ l jg , r jg ] . For segment j on gene g the probability of the predicted occupancies given the segment ratio estimates can be expressed as p ( C , N j g | μ j g , σ j g ) ∝ f C ( N j g ; μ j g , σ j g ) , ( 1 ) where N jg ≡ ∑ i = l jg r jg n ig / ( r jg − l jg + 1 ) is the predicted average occupancy on segment j of gene g , and f C ( x ; μ , σ ) = 1 x σ 2 π e − ( ln x + ln C − μ ) 2 2 σ 2 is the log-normal probability density function describing the density ratio measurement error scaled by factor 1 C . This formulation is used for comparing the predicted occupancies to the estimated values in a probabilistic fashion . Assuming independence between ratio estimates of the same gene and between genes , the probability of observing all estimates , denoted by n , can be expressed as p ( C , n | T ) ∝ ∏ g ∏ j ∈ J g f C ( N j g ; μ j g , σ j g ) , ( 2 ) In practice these calculations are more easily performed in log space and the constant factors are dropped: ψ ( C , n | T ) = ∑ g ∑ j ∈ J g ln f C ( N j g ; μ j g , σ j g ) ∼ ∑ g ∑ j ∈ J g [ - 1 2 ( σ j g ) 2 ( ln N j g - μ j g + ln C ) 2 - ln N j g] ( 3 ) We use ψ ( C , n∣T ) as the objective function for quantifying how well model-predicted ribosome occupancies match measured data . To choose the scaling factor C , we note that it is the only free parameter of ψ ( C , n∣T ) if model output n and segment tree estimates T are given . In that case , the value of C maximizing ψ can be determined analytically: ln C = ( ∑ g , j ∈ J g 1 ( σ j g ) 2 ( μ j g - ln N j g ) ) / ( ∑ g , j ∈ J g 1 ( σ j g ) 2 ) ( 4 ) Throughout this paper , different models are evaluated at a scaling factor C maximizing their fit to the data ( i . e . maximizing ψ ) . While the unknown true scaling factor is determined by the physiological properties of the cell , the efficiency of the experimental protocols and characteristics of the high-throughput sequencing measurements ( see section “Initiation rate approximation” and S1 Text ) , evaluating models at the best possible scale allows for a more fair evaluation as it does not penalize models in cases when the model and the true scales mismatch . TASEP ( Totally Asymmetric Exclusion Process ) models mRNAs g as one-dimensional lattices of length Sg and ribosomes as abstract “particles” occupying L sites corresponding to codons ( Fig 2 ) . These particles hop on ( translation initiation ) and off ( translation termination ) the lattice at the first and last sites with rates k 0g and k S gg respectively . They only move towards the end of the lattice ( hence the totally asymmetric ) by hopping one site at a time with site-specific elongation rate k ig . Ribosomes interact with each other by “excluding” a volume of L sites that they cover on the lattice , meaning that a ribosome cannot continue to the next codon if it is already covered by another ribosome . The exact location of the active site among the L covered codons does not change the rules governing ribosome motion [22] , but the choice of L may influence simulation dynamics in cases of high ribosome queueing . Typically , values 9 ≤ L ≤ 11 are used [8 , 9 , 16 , 21]; L = 10 was chosen for our simulations as it best matches the RP footprint size distribution [10] . TASEP captures the high-level physical interaction between ribosomes and transcripts by describing the ribosomes as travelling on the mRNAs . While in practice a number of varying translation scenarios are possible ( e . g . RER-bound translation with ribosomes glued to the endoplasmic reticulum and moving very slowly while the mRNA is instead pulled though the ribosomes [25] ) , the rich set of behaviors attainable by TASEP makes it a suitable framework for modelling translation . It requires specification of a large number of parameters , namely the gene- and site-specific elongation rates k ig ( with the stop codon elongation rate functioning as the termination rate ) and the gene-specific initiation rates k 0g . To reduce the number of parameters we assume that the site-specific elongation rates are codon-specific and do not differ between genes . This commonly made assumption [7 , 16 , 21 , 26] is necessary for determining model parameters from RP data as it makes the model fitting problem tractable . Depending on the experiment , either elongation rates consistent with the tRNA pool adaptation hypothesis were fixed to allow fitting the initiation rates only , or all model parameters were fit on the available data . Evaluation and fitting of the TASEP model requires an efficient way of obtaining steady-state ribosome occupancies . TASEP models allow limited analytical tractability and , to our knowledge , no analytical results for the steady-state codon occupancy probabilities are available for the general case . Additionally , existing TASEP mean-field approaches poorly approximate codon occupancies [27] , a quantity of particular importance to this study , leaving stochastic simulations as the only suitable approach . TASEP steady-state codon occupancies were obtained by simulating the model using a Monte Carlo algorithm , i . e . by randomly selecting an event ( translation initiation , elongation or termination ) in every simulation step and , if no other ribosomes interfere with the event , executing it with a probability proportional to its rate . To speed up simulation we developed a continuous time simulation method similar to the Gillespie algorithm [28] , but based on the use of the Erlang distribution to only sample times between state-changing events , i . e . events that change the configuration of ribosomes attached to an mRNA . Formally , the times between consecutive initiation or elongation events at position i are assumed to be exponentially distributed with rates k 0g and k ig respectively ( i . e . the corresponding model rate parameters , Fig 2 ) . Let oi , i = 1 , … , Sg be the current state of the simulated molecule: o i = { 1 , codon i is occupied by a ribosome ( is at its A-site ) 0 , otherwise . ( 5 ) Then the time between any two consecutive events is also exponentially distributed with rate k = k 0g + ∑ i = 1 S g o i k ig as the minimum of independent exponentially distributed random variables . Once an event occurred , the probability that it was event j is given by p j = o j k jg / k ( it is assumed that ribosomes are always available to initiation translation , i . e . o0 = 1 ) . Some of the events cannot be executed due to ribosomes blocking each other and do not lead to a state change . If k+ is the sum of rates of events leading to a state change , then the number of events between consecutive state changes , denoted as e , follows a geometric distribution with parameter p+ = k+/k and the time Δt between state changing events follows the Erlang distribution with shape e and rate k as the sum of iid exponential random variables . The simulation proceeds by repeated random sampling of the number of events , the time between events and the event type s from the appropriate probability distributions; and updating ribosome locations in accordance to the sampled event: s ∼ Categorial ( p 0 , p 1 , … , p S g ) , e ∼ Geometric ( p + ) , Δ t ∼ Erlang ( e , k ) . ( 6 ) Simulating only state-changing events allows the simulation to progress faster , especially in cases of high ribosome queueing . The total time T ig spent by ribosomes at position i and the total simulation time Tg are recorded to estimate the per-transcript ribosome occupancy at this position as n ig = T ig / T g , which is then used for comparing the model to RP data . Similarly the total number of translation terminations Fg is used to estimate the protein production rate Jg = Fg/Tg . To reach steady-state distribution of ribosomes on mRNA irrespective of the CDS length , each mRNA was simulated until 1000 translation termination events occurred . After that the model was further simulated for up to 107 additional steps or until the average ribosome occupancy in the segments of interest was estimated with high precision ( absolute error ϵ < 10−3 ) . The latter stopping criterion is based on the observation that average ribosome occupancy over a fixed segment of the mRNA can be reliably estimated before per-position occupancies can . Segment densities were first estimated after 5 × 105 simulation steps and then every 106 steps . Simulation was stopped if the absolute error between consecutive estimates was smaller than ϵ . In addition to the elongation rates , large TASEP models require specification of hundreds gene translation initiation rates prior to simulation . Direct measurements of the initiation rates rates are unavailable and instead their values are often inferred from other sources such as ribosome profiling [8 , 9] or polysome size measurements [21] data . Initiation rates estimated in such a way depend on the rates of translation elongation used in the analysis , and hence need to be optimized together with the elongation rates of the TASEP model . This leads to an explosion of the number of parameters that need to be determined , stressing the need for highly efficient initiation rate approximation strategies if the initiation and elongation rates are to be determined from the RP data simultaneously . The problem of determining initiation rates was previously tackled by approximations neglecting ribosome queueing [8 , 9] , and by near-exhaustive computational search [21] . We propose a method that is a compromise between the two approaches—it allows approximating gene initiation rates for the TASEP model from RP data at a fraction of the computational cost of an exhaustive search . Briefly , we add an additional parameter C ˜ , the “proposed” scaling factor , to the list of model parameters that need to be estimated . This parameter is identical to the scaling factor C from Eq ( 4 ) , but is used within the model to obtain biologically meaningful initiation rates . We calculated the value of C ˜ from the number actively translating ribosomes [29] and the number of mRNA molecules [30] per cell using a procedure proposed by Siwiak and Zielenkiewicz [8] . Given some estimate of the elongation rates and C ˜ we then find optimal initiation rates using a novel numerical approximation of ribosome density for TASEP models that is based on the observations of Cinandrini et al . [21] . This approach allows us to decouple initiation rates from elongation rates and greatly reduces the number of model parameters that need to be fitted explicitly ( next section ) . We used this method to efficiently ( re- ) approximate initiation rates of genes for each new set of elongation rates k ig . A full description of the approach is available in the S1 Text . When fitting the TASEP models , translation rates that maximize ψ ( C , n∣T ) are sought . Lacking a closed-form solution , we employed the Covariance Matrix Adaptation Evolutionary Strategy ( CMA-ES [31] ) to find these rates . We considered two different TASEP models: TASEPinit and TASEPelong . In TASEPinit the elongation rates are fixed at values consistent with the tRNA pool adaptation hypothesis and initiation rates are approximated as described earlier . In the TASEPelong model none of the parameters are fixed: also the codon-specific elongation rates are optimized with the CMA . Since TASEP simulation output is invariant to scaling of translation rates , many equally good solutions exist . To constrain the search the elongation rate of codon GAA was fixed at its initial tRNA pool adaptation hypothesis value . The codon was chosen as it is present in many genes and segments ( S5 Fig ) . Further details regarding the use of CMA can be found in the S1 Text . Despite the efficient Monte Carlo simulation and translation rate search strategies , model fitting remains a very CPU-intensive task . To speed up computations in practice , the models were fitted using hundreds of CPUs in parallel as individual genes can be simulated independently . Because TASEP simulations of different genes are independent of each other , it may be unclear how to interpret the fitted elongation and initiation rates , as they must depend on such global biophysical quantities as the number of tRNAs or ribosomes within the cell . Nevertheless , the final simulation results are compared to whole-genome RP measurements . We can therefore expect that if our TASEP simulations agree well with RP data , the fitted translation rates used within the simulations account for the necessary biophysical parameters . Thus they should be regarded as the effective initiation and elongation rates that account for the relevant biophysical characteristics of the cell and growth conditions . We note that translation rates determined in such a way are condition-specific , and would likely change if fitted on a dataset obtained under different growth conditions . To obtain a baseline for evaluating the performance of fitted TASEP models we also evaluated several existing state-of-the-art static and dynamic models of translation and compared them to each other based on their agreement with the RP data as given by Eq ( 1 ) . SMoPT [9] and Zhang’s model [6] were chosen for evaluation on the segment tree data as other state-of-the-art models , namely the Ribosome Flow Model [7] and the model from Siwiak and Zielenkiewicz [8] , do not provide ribosome occupancy profiles compatible with the segment tree interpretation . The latter model was however compared to the fitted TASEP models based on several independent PA datasets . When comparing models’ predictions using independent protein abundance datasets , the “initiation frequency” P , “total amount of protein molecules produced from transcripts of particular type” B and the “total time of translation of one protein molecule from a given transcript” T from Siwiak and Zielenkiewicz [8] were respectively treated as translation initiation rate , the product of J and mRNA levels , and the inverse of J; the average gene total elongation time from SMoPT [9] was treated as the inverse of J; 𝓟 from Ciandrini et al . [21] was treated as J . Since the sets of genes included in SMoPT and the segment trees differ , to facilitate comparison , all models were evaluated on a set of 3 , 617 genes ( 49 , 894 segments ) that were in common between all models after removing very long genes to speed up TASEP simulations ( 31 genes longer than 2 , 000 codons ) . This set of genes was used to fit TASEP models inside a 5-fold stratified cross-validation ( CV ) loop over genes , in which the CV folds were chosen to balance the number of genes and segments between folds . In every step of the CV 1 fold was used for fitting ( training set ) and 4 folds were used for model evaluation ( test set ) . Smaller training sets were used to reduce model fitting time . To evaluate predictions of the proposed TASEP models we also fitted them on all segment tree estimates . And to further reduce fitting time on this large dataset , codon elongation rates of the TASEPelong model were set to the geometric mean of elongation rates from CV folds , and only the initiation rates were estimated from the data . To simplify comparison of different models , we computed CV objectives for all evaluated models , including the models that did not require any parameter fitting ( i . e . SMoPT and Zhang’s model ) . While the static Zhang model does not explicitly model the translation initiation step , SMoPT and TASEP models require initiation rates to be defined for every gene in the test sets in order to calculate the CV objective . We used the original initiation rates inferred from the RP data [9 , 10] for SMoPT , and approximated TASEP initiation rates using the test set segment tree measurements . Some of the experiments required the translation elongation rates to be defined . For those experiments we used translation elongation rates kAAA , … , kGGG consistent with the tRNA pool adaptation hypothesis , which could be seen as a statement that codons recognized by more abundant tRNAs are translated faster . The exact values for the elongation rates were defined based on the tRNA Adaptation Index ( tAI [32] ) , which quantifies the decoding efficiency of a codon by simultaneously considering abundances of all tRNA species recognizing it and the strength of wobble base pairing between the codon and the anticodons of the isoacceptor tRNAs . The elongation rates kAAA , … , kGGG were calculated as the inverse of the codon translation times taken from the Ribosome Flow Model [33]; and translation termination rates ( i . e . kTAG , kTAA , kTGA ) were set to 1 . The tAI and CAI ( Codon Adaptation Index [34] ) are the most commonly used codon indices . They quantify respectively the extent to which a particular sequence consists of codons recognized by abundant tRNAs , and the extent to which a particular sequence consists of codons present in highly expressed ( e . g . ribosomal and glycolytic ) genes . These indices are often used as a proxy for translational efficiency of a gene and are employed to optimize its sequence for expression in a different host organism . Having determined elongation rates for the TASEPelong model , we sought to understand whether these rates suggest a different optimization scheme than the one given by tAI or CAI . For each codon the tAI ( CAI ) assigns a number—the absolute adaptiveness of that codon to the tRNA pool ( codons used in highly expressed genes ) . To facilitate comparison between the different indices , following the definition of the CAI , we define the relative adaptiveness of a codon as its absolute adaptiveness normalized by the maximum adaptiveness among synonymous codons . We then use the relative adaptiveness for CAI , tAI and an index based on the TASEPelong elongation rates ( described below ) , when comparing optimization schemes . We note that from the definitions of tAI [32] and elongation rates consistent with the tRNA pool hypothesis ( previous section and [7] ) it follows that the tAI absolute codon adaptiveness and the elongation rates are proportional to each other , and use this observation to define a codon index based on the fitted TASEPelong elongation rates . We define the relative adaptiveness of a codon according to TASEPelong as its elongation rate normalized as described above . Protein abundance measurements were taken from Newman et al . [35] ( YEPD and SD media ) and Ghaemmaghami et al . [36] . 5′- and 3′ UTR lengths were determined based on Nagalakshmi et al . [37] and Yassour et al . [38] as the mean length obtained from the two studies .
Segment density ratios are estimates of the average number of ribosomes engaged in translation of a given segment ( measured in arbitrary units ) , and are expected to become more reliable if the segment length is increased . Fig 3 shows that estimates obtained for longer segments are indeed more reliable ( smaller σ values ) with the longest segments ( rightmost group ) being nearly as reliable as the full-CDS estimates from all genes ( S2 Fig ) . We note that although group widths increase almost exponentially , potentially collecting segments with different i . r . e . in the top group , the constructed groups map very well to individual levels of the segment trees because lengths of segments with each new level are halved on average . This mapping thus provides important additional information to the segment trees about the increasing reliability of measurements that are located higher within the tree . In this way , segment trees establish a tradeoff between measurement reliability and measurement resolution by combining the use of trustworthy estimates high in the tree ( corresponding to longer segments , describing high-level gene behavior ) with the use of many less reliable estimates located lower in the tree that describe the local density variation . As can be seen from the visualization of the raw data for gene YLR449W and its segment tree reconstruction in Fig 4 , our multi-scale approach , that combines measurements from different scales ( segment lengths ) , allows for implicitly capturing changes in ribosome density along transcripts , while at the same time keeping the average ribosome density across larger segments close to the observed levels . This representation also encodes uncertainty about the density ratio at a particular region of the gene , even if that region is not directly represented by a segment from the tree . For example , region ( 85 , 104 ) ( highlighted in the figure ) is covered by 6 segments ( i . e . has depth 5 within the tree ) and has one of the tightest confidence intervals ( CIs ) in the reconstruction . At the same time region ( 105 , 120 ) was not measured at the two lowest scales ( has a depth 3 ) and its average density has to be derived from the density values of other segments and our uncertainty about them , leading to a wider CI . This example demonstrates how segment trees capture changes in ribosome density along the transcript , which are crucial for fitting translation rates and evaluating competing models . Small standard deviations of the scaling factors and objective scores ( determined using CV ) of the evaluated models shown in Table 1 suggest that the ( fitted ) models perform consistently across different folds . The objective scores also show that knowledge-based models ( i . e . the SMoPT and Zhang models ) based on a manual choice of numerous translation-related parameters explain the ribosome density measurements significantly worse than the two models fitted on RP data . This can also be concluded from a visual inspection of the predictions made by these models for one of the genes in Fig 4C , which shows that their ribosome occupancies tend to “miss” the measured density ratios . For the Zhang model this could be explained by the absence of gene-specific initiation rates in the model , whereas SMoPT often overshoots the measured density ratios , presumably because it over-estimates initiation rates by neglecting ribosome queueing . The TASEPinit model simulated with tAI-based elongation rates and fitted initiation rates achieves a significantly higher objective scores than the two state-of-the-art models . It is further improved by the TASEPelong model , for which the elongation rates are additionally fit on the segment tree measurements . Fig 5 shows that superior objective function values of the fitted models translate to better predictions of the measured ribosome density ( Pearson correlation coefficient r = 0 . 77 vs . 0 . 45 , p < 10−293 ) . Although the predictions are generally better for longer segments , improvements can be observed at all scales ( see S3 Fig ) . While due to its relative simplicity only a weak positive correlation was expected for the Zhang model , for reasons unclear , a highly significant ( p < 10−293 ) negative correlation is observed ( Fig 5 , left ) . This demonstrates that current knowledge-based models are not supported by RP measurements and highlights the importance of a critical evaluation of existing translation models against independent measurements . Although TASEPinit and TASEPelong outperformed existing models in the CV experiments , care has to be taken when interpreting these results as only the TASEP models were fitted directly on the segment tree measurements . We sought to obtain additional confirmation of the models’ performance and to determine if they make biologically meaningful predictions . To this end we compared the protein production and translation initiation rates given by TASEP models fitted on all segment tree estimates to several independent large-scale PA datasets ( see Materials and Methods ) . Most importantly , we found that for both models the predicted protein production rates ( PPRs ) J positively correlate with the PA measurements ( Table 2 ) . As expected , because J describes PPR per transcript , these correlations improve when the product of J and mRNA levels ( J × mRNA; mRNA levels taken from the RP data ) is considered . Even when both J and PAs are corrected for mRNA levels ( thereby removing transcriptional regulatory influences in order to study translational regulation in isolation ) , the remaining ( partial ) correlation between J′ and PA′ is still significant , indicating that our TASEP models adequately capture the effects of protein translation on protein levels . These correlations are superior compared to correlations observed for state-of-the-art models ( Table 3 ) , especially when the partial correlations are considered . While strong positive partial correlations would be expected , we find these only for the fitted TASEP models . Unexpectedly low and negative partial correlations between PA′ and J′ for other models , together with strong correlations between PPR and mRNA levels ( Table 4 ) suggest that existing models are overfit on transcript levels and may not accurately model translation . These findings provide an independent confirmation that the TASEP models with fitted translation rates accurately capture the dynamics of the S . cerevisiae translation machinery . Looking more in detail ( Table 4 ) , we find that for both models the fitted initiation rates agree well with the rates inferred by the existing full-cell models of Shah et al . ( SMoPT ) , and of Siwiak and Zielenkiewicz . However , we did not find the previously reported strong negative correlation between initiation rates and CDS length [9 , 21] . We note that this correlation is also not supported by the model of Siwiak and Zielenkiewicz . The initiation rates also exhibit a weak correlation with the 3′ UTR lengths ( similar correlations also found for several other models ) , supporting the hypothesis of more efficient translation re-initiation in genes with longer 3′ UTRs . Interestingly , we did not find the tendency for genes with short 5′ UTRs to exhibit high initiation rates suggested by Shah et al . and supported by Ciandrini et al . [21] in our models or the model of Siwiak and Zielenkiewicz . We also note that no relationship or a negative relationship can be observed between initiation rates and 5′ UTR lengths corrected for CDS lengths can be found in most considered models . This suggests that the previously observed inverse relationship between 5′ UTR lengths and initiation rates may not be indicative of a 5′ UTR-mediated initiation rate regulation mechanism , but could be merely a consequence of a positive correlation between 5′ UTR lengths and CDS lengths . While correlations observed for the fitted models do not change between TASEPinit and TASEPelong ( Table 4 ) , the latter model makes considerably better ribosome occupancy predictions . It can be seen from the example in Fig 4C that fitting the elongation rates allows the segment-averaged ribosome occupancy of TASEPelong to follow the reconstructed density considerably better than any of other model . Since the TASEPelong model achieves a significantly better fit to the RP data compared to TASEPinit with tAI-based rates ( Table 1 ) , having fitted its elongation rates on different CV folds , we sought to interpret the obtained values and their variance . We first , however , confirmed that elongation rates determined from different RP datasets agree qualitatively with each other by fitting a new TASEPelong model on the dataset of Ingolia et al . [10] and comparing its translation rates to the original model ( see S1 Text ) . It can be seen from Fig 6 that despite the generally large SDs , for many codons the elongation rates fitted in different folds of the CV are spread compactly around codon-specific values . This is clearly visible for codons with smaller SDs ( green and blue ) , for which similar rates were found in different folds . Nonetheless the rate SDs differ considerably between codons . While the majority of the fitted elongation rates are consistently different from tAI-based rates , only for 13 codons this difference is statistically significant ( single sample t-test for population mean difference , p < 0 . 05; Fig 6 , S2 Table ) : GAC , TTG , CCA , CAA , GCC , GGT , GAT , TTT , CAG , GTG , ACG , CCT and CGA . Although these differences between the tAI-based and fitted elongation rates are challenging to explain , their presence suggests that additional unknown factors are shaping these rates . Having identified differences in elongation rates between the TASEPinit and TASEPelong models , we sought to understand their effect on models’ predictions . As could be expected from the similar correlations in Table 4 and Fig 5 , the two models make very similar PPR and ribosome density predictions ( S4 Fig ) . However , ribosome density predicted by the TASEPelong model with fitted elongation rates agrees better with RP measurements . To understand the cause of this improvement we looked for genes whose fit to the RP data improved when fitted elongation rates were used . These genes can be classified into two groups: ( i ) genes that have a very similar initiation rate in both models ( Fig 7 , left ) ; and ( ii ) genes that have a considerably lower initiation rate in the TASEPelong model ( Fig 7 , right ) . Because all 13 codons with significantly different elongation rates were predicted to be slower , their presence in CDSes generally leads to higher predicted ribosome occupancy , especially if the genes initiation rate remains unchanged . For genes from the first group , such as YOR202W shown on the left panel of Fig 7 , this already results in a more accurate ribosome occupancy prediction . For most other genes , the second group , this increase in codon elongation times yields ribosome occupancy that is too high under the current initiation rate . For these genes ( e . g . YGR284C on the right panel of Fig 7 ) a smaller fitted initiation rate is required to reduce ribosome occupancy that would otherwise be too high due to the effects of slow codons and high ribosomal flux ( due to high initiation rate ) . Together these effects allow the model to better match the ribosome density changes along the transcript . Codon optimization , the process of substituting codons with synonymous alternatives that are elongated faster , thus contributing to the overall protein production rate , is routinely used to improve protein expression [39 , 40] . Nonetheless , it remains a controversial tool because the same optimization techniques can lead to contradicting results when applied to different proteins [41] . Here we compare our fitted elongation rates to codon optimality estimated by the commonly used tAI [32] and CAI [34] indices . We considered the relative adaptiveness of a codon ( see Materials and Methods ) given by the CAI , the tAI and the fitted elongation rates of the TASEPelong model . Fig 8 shows that the three measures of codon adaptation often agree on the optimal codon for a particular amino acid ( relative adaptiveness of 1 . 0 , dark blue ) , which further demonstrates that our findings are in line with the earlier work . In particular , despite significant differences between the fitted elongation rates and elongation rates given by the tRNA adaptation hypothesis , the two sets agree on optimal codons for all but four amino acids . Only for isoleucine , leucine , lysine and serine the TASEPelong model suggests codons ATC , AAA , TTA and TCG instead of ATT , AAG , TTG and TCT respectively . An interesting observation is that the bottom row in Fig 8 is much more blue than the top ones , suggesting codon optimization is less black-and-white than suggested by tAI and in particular CAI , meaning that many more codons are “reasonably good” , i . e . there may be less to gain by codon optimization than thought before . This observation is also corroborated by Leavitt and Alper [42] , who noted that the level of control achievable in yeast through codon optimization is considerably smaller than what can be achieved through transcriptional regulation . It is still unclear whether translation of endogenous yeast genes is limited by initiation or elongation [43 , 44] . To test whether translation is limited by the initiation rates or by the elongation rates we artificially increased the initiation rate of each gene from the TASEPelong model by 10% . To obtain robust results the experiment was repeated 5 times with different random initializations and the average increase in PPRs was calculated for every gene . Fig 9 shows the relative differences in PPRs for all genes . In almost all cases ( except 7 genes ) the PPR increased substantially ( relative difference > 0 . 02 ) when increasing the initiation rate , supporting the hypothesis that under exponential growth in the rich medium translation in S . cerevisiae operates in an initiation-limited regime . This also explains why fitting the codon elongation rates in TASEPelong did not improve the PA correlations compared to the TASEPinit model . Elongation-limited production for these genes can be explained by the very high initiation rates predicted for them , which shift the rate-limiting step from translation initiation to translation elongation . Interestingly , groups of genes that had a low , medium and high PPR increase are enriched for several biological functions ( FDR < 0 . 05 , Fig 9 ) . Notably , genes in the high increase group are involved in negative regulation of various biosynthetic and metabolic processes . This suggests that yeast cells may have evolved to rapidly “switch on” negative regulation by keeping a buffer of the required mRNA transcripts that are efficiently translated only once there is demand .
For the first time , we described an approach that derives complete translation kinetics of an organism from ribosome profiling data and used it to simultaneously infer the translation elongation , translation initiation and protein production rates all together without neglecting the effects of ribosomal interference . We applied our methodology to the ribosome and RNA sequencing data of the baker’s yeast Saccharomyces cerevisiae . The fitted yeast translation models agree considerably better with independent protein abundance datasets than existing models . In particular , our TASEP models are the only ones that maintain strong correlations with protein abundance after removing the effect of transcriptional regulation . While translation initiation rates provided by the models are similar to rates from other studies , we did not find the previously reported negative correlation between initiation rates and CDS lengths . The observed negative correlations between PA and CDS length , which one would expect to see as a result of this correlation , can alternatively be explained by transcriptional regulation , i . e . the strong negative correlation between mRNA levels and CDS lengths ( Table 4 ) . An alternative explanation can be offered by a mechanism driven by amino acid chain elongation rather than translation initiation . For example , abortive translation or the degradation of misfolded proteins [45] , since the chance of producing a misfolded protein is expected to increase with protein length . We also found that translation elongation rates deviate considerably from the widely accepted tRNA pool adaptation hypothesis , for 13 codons significantly so . Differences in elongation rates of these codons between the tRNA pool adaptation hypothesis and TASEPelong may be partially explained by nucleotide modifications of their respective tRNAs , which are known to modify the specificity and efficiency of messenger decoding [46] . As such , some of these 13 codons were shown to be affected by post-transcriptional nucleotide modifications of tRNAs in different organisms [47] . We speculate that for these codons the concentration of ( un ) modified tRNAs , rather than the total tRNA concentration , plays a non-negligible role in determining their elongation rates [18] . An additional factor that possibly contributes to the observed deviation from the tRNA pool adaptation hypothesis is its implicit assumption that different tRNA genes from the same family contribute equally to determining the rate of translation . This assumption should be revisited in light of the recent finding of Bloom-Ackermann et al . [48] that the contributions of different gene copies from the same tRNA family to the tRNA pool and cellular fitness are far from equal . In our experiments we found that SDs of elongation rates from different CV folds differ markedly between codons . In order for the elongation rates to be specified with high precision by the RP data , small changes in the rates must lead to detectable differences in ribosome density . However , in light of our finding that yeast translation is initiation-limited and the observation of Bloom-Ackermann et al . [48] that S . cerevisiae is robust to deletions of tRNA genes , especially in rich medium used to produce the ribosome profiling measurements analyzed here , it is unlikely that in the considered physiological conditions the elongation rates exert a strong enough effect on ribosome density to allow the RP data to specify elongation rates with high precision . We speculate that found SDs reflect the robustness of the yeast translation system w . r . t . the codon translation rates , with the system being more sensitive to changes in rates of those codons that have smaller SDs . In this case , yeast translation appears to be robust to fold changes in codon translation rates and , consequently , to the aminoacyl-tRNA availability that these rates are thought to be determined by [44] . Alternatively , the SDs may reflect the extent to which codon translation rates change between CV folds due to codon context , i . e . the local sequence around a codon which may alter its elongation rate ( see S1 Text , translation rate reproducibility analysis ) . It is unlikely that the TASEP model captures the full complexity of the translation process by assuming that codon elongation rates are determined solely by the codon identity , and not also by the sequence surrounding the codons as was previously suggested [2 , 3] . Such a constraint limits the models ability to capture the underlying translation dynamics and may bias it towards fitting different rates on different sets of genes ( e . g . CV folds ) with varying codon contexts , thereby inflating the SDs . The observed variation in fitted elongation rates puts forward codon context as a factor that may significantly modulate the baseline elongation rates . Using our models we found that under exponential growth in rich medium translation initiation appears to be the main limiting factor of protein production of endogenous genes in Saccharomyces cerevisiae , with protein production being limited by initiation rates for all but 7 genes with very high initiation rates . These findings suggest that rational design of 5′ UTRs involved in translation initiation [49 , 50] may be a more promising avenue for achieving protein overexpression than the routinely used codon optimization techniques . It is likely , however , that further overexpression could be achieved using codon optimization . Because once the gene is put under the translational control of an efficient 5′-UTR , which is usually the case in heterologous gene expression , translation elongation is expected to become a rate-limiting factor . In such cases we recommend performing codon optimization using the fitted TASEPelong elongation rates , which , while mostly agreeing with existing techniques , also demonstrate several differences . Although we found that translation initiation appears to be the main factor limiting protein production in yeast under exponential growth in rich medium , it is possible that different mechanisms are dominant in other organisms . For example , Li et al . [51] and Guimaraes et al . [52] discuss greater contribution of protein elongation respectively by anti-Shine-Dalgarno sequences and codon usage in E . coli . Our method could be applied to ribosome profiling data of other organisms to delineate the relative contribution of initiation and elongation . All translation models proposed to date , including TASEPinit and TASEPelong , assume that translation elongation rates are not influenced by codon context , i . e . the sequence around a particular codon , although various factors affecting the speed of elongation have been suggested [2–4] . Variation in fitted elongation rates and the highly varying codon translation times recently observed by Dana and Tuller [53] suggest that codon context may play a more compelling role in determining translation rates than previously thought . Fortunately investigations of codon context are becoming feasible thanks to the growing adoption of ribosome profiling as a standard technique for studying translation . With the increasing amount of ribosome profiling measurements , data-driven approaches , such as the one described here , will become instrumental for delineating the effects of multiple competing translation mechanisms , for generating new hypothesis , and for constructing predictive models for use in other fields . These goals can be achieved by incorporating alternative translation mechanisms as sequence- and position-specific effects altering the codon elongation rates . | Translation , the process of synthesizing proteins from mRNA templates , is an essential biological process in all living organisms . A better understanding of this process will have ramifications in various fields—from gene regulation , disease understanding and medicine to biotechnology and synthetic biology . Nonetheless , a holistic understanding of the processes remains elusive , making computational modelling a promising approach for studying it . However , accurate modelling of translation is challenging due to many assumptions made by such models and due to the sheer number of parameters that need to be specified . Here , we propose to fit models of translation onto ribosome profiling measurements , which record snapshots of the locations of actively translating ribosomes on mRNAs from millions of cells . We develop statistical and computational methods for fitting the Totally Asymmetric Exclusion Process ( TASEP ) models of translation on these measurements and verify them by deriving highly accurate translation models for the baker’s yeast Saccharomyces cerevisiae , which outperform existing models on independent datasets . We find that fitted elongation rate parameters from the derived models deviate significantly from the widely accepted tRNA pool adaptation hypothesis . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | Unbiased Quantitative Models of Protein Translation Derived from Ribosome Profiling Data |
Trichome patterning in Arabidopsis serves as a model system to study how single cells are selected within a field of initially equivalent cells . Current models explain this pattern by an activator–inhibitor feedback loop . Here , we report that also a newly discovered mechanism is involved by which patterning is governed by the removal of the trichome-promoting factor TRANSPARENT TESTA GLABRA1 ( TTG1 ) from non-trichome cells . We demonstrate by clonal analysis and misexpression studies that Arabidopsis TTG1 can act non-cell-autonomously and by microinjection experiments that TTG1 protein moves between cells . While TTG1 is expressed ubiquitously , TTG1–YFP protein accumulates in trichomes and is depleted in the surrounding cells . TTG1–YFP depletion depends on GLABRA3 ( GL3 ) , suggesting that the depletion is governed by a trapping mechanism . To study the potential of the observed trapping/depletion mechanism , we formulated a mathematical model enabling us to evaluate the relevance of each parameter and to identify parameters explaining the paradoxical genetic finding that strong ttg1 alleles are glabrous , while weak alleles exhibit trichome clusters .
During the development of animals and plants , specific cell types need to be placed in a regular pattern within a field of cells . In the simplest scenario , this occurs in a two-dimensional sheet of cells . Mathematical modeling of such a spacing pattern has uncovered two general principles . Both rely on the assumption that the factor promoting the formation of the specific cell type is autocatalytic . In the “activator–inhibitor” mechanism autoactivation is counteracted by the production of an inhibitor . In contrast , in the “substrate-depletion” mechanism , a substrate is consumed by the autocatalysis of the cell type promoting factor . A common requirement of both principles is significantly reduced mobility of the autocatalytic species compared to that of the inhibitor and the substrate , respectively [1] . The activator–inhibitor system is thought to generate the regular spacing pattern of leaf trichomes in Arabidopsis [2–4] . Trichomes are regularly distributed on the leaf surface without any reference to morphological landmarks , and clonal analysis indicated that cell lineage is not involved [5 , 6] . Therefore , trichomes are an ideal model system to study how single cells become regularly spaced within a sheet of equivalent cells . Current models assume that the R2R3 MYB transcription factors GLABRA1 ( GL1 ) and MYB23 [7–9] , the bHLH factors GLABRA3 ( GL3 ) and ENHANCER OF GLABRA3 ( EGL3 ) [10–12] , and the WD40 repeat protein Transparent Testa Glabra1 ( TTG1 ) [13 , 14] form a trichome-promoting trimeric complex due to the binding of one R2R3 MYB factor and TTG1 to a bHLH factor . Formally , this complex acts as the activator described in the theoretical models [1] . The activity of this complex is thought to be counteracted by the single R3 repeat MYB-like transcription factors TRIPTYCHON ( TRY ) [15] , CAPRICE ( CPC ) [16] , ENHANCER OF TRY and CPC1 ( ETC1 ) [17] , ETC2 [18] , TRICHOMELESS1 [19] , and CAPRICE LIKE MYB3 ( CPL3 ) [20] through competition for binding of the R2R3 MYB factors to the bHLH protein [21] . The single R3 repeat MYB proteins are collectively considered to represent the inhibitor in the theoretical models . The active complex ( AC ) is postulated to activate the inhibitors , which can move into neighboring cells , where they repress the activators . This type of model is generally consistent with most data though several aspects have not been confirmed experimentally [3 , 4 , 6 , 22 , 23] . The role of TTG1 in trichome patterning is obscure , as the glabrous phenotype of strong alleles suggests that it promotes trichome development , whereas the formation of trichome clusters in weak alleles suggests that it is involved in the inhibition of trichomes [5 , 24] . This dual function of TTG1 suggested to us that TTG1 has a central function in the patterning process . In this work , we identified TTG1 as the key component of a newly discovered depletion mechanism , likely to act in parallel to the above-described activator–inhibitor mechanism . We demonstrate that TTG1–YFP depletion depends on GL3 , suggesting an underlying trapping mechanism , such that GL3 captures TTG1 in trichomes . Finally , we provide a mathematical model to evaluate the properties of this new GL3/TTG1 trapping/depletion mechanism .
TTG1 is expressed in most tissues of the plant [14 , 25] . To determine the TTG1 expression in young leaf parts , where trichome initiation takes place , we created transgenic plants , in which the β-glucoronidase ( GUS ) reporter gene was driven by a 2 . 2 kb promoter fragment including the 5′ UTR of TTG1 ( pTTG1:GUS ) . This fragment is sufficient to rescue completely the ttg1–13 null-mutant phenotype when driving the TTG1 cDNA ( Table 1 ) . pTTG1:GUS is ubiquitously expressed in young leaves with slightly elevated levels in incipient trichomes , and expression ceases in more mature leaf parts ( Figure 1A and 1B ) . To determine the localization of TTG1 protein , we created a C-terminal fusion of TTG1 with yellow fluorescent protein ( YFP ) and an N-terminal fusion with green fluorescent proten ( GFP ) , which both rescued all aspects of the ttg1–13 mutant phenotype , including the seed coat mucilage , transparent testa , and trichome number when expressed under the TTG1 promoter ( unpublished data; Table 1 and Figure 2A–E ) . We further substantiated the functionality of this rescue construct by demonstrating that protein–protein interactions of TTG1–YFP with GL3 are indistinguishable from TTG1 in yeast two-hybrid interaction assays ( unpublished data ) . Both fusion proteins were found in the nucleus and in the cytoplasm ( Figure 2F ) . The integrity of the TTG1–YFP fusion protein was confirmed by western blot analysis ( Figure 2G ) . The distribution of the pTTG1:TTG1–YFP fusion protein differed strikingly from pTTG1:GUS expression . Initially , in very young leaf regions , in which trichomes are not yet initiated , TTG1–YFP is detected in all cells reflecting the gene expression pattern ( Figure 1C ) . In slightly older leaf regions , TTG1–YFP accumulates in incipient trichomes ( Figure 1C , 1D , and 1E ) . In the cells adjacent to young trichomes , TTG1–YFP levels are the lowest , and fluorescence gradually increases with the distance from the trichome ( Figure 1D and 1E ) . This initial observation was confirmed by quantifying the fluorescence intensity , using the Leica Confocal software ( Figure 1F ) . On average , cells next to a trichome showed 39% of the fluorescence of that in the trichome , the cells in the second tier around a trichome 76% , and cells of the third tier 93% ( n = 31 ) . As a control , we measured the distribution of fluorescence of a nuclear-localized GFP under the control of the TTG1 promoter ( pTTG1:GFP-NLS , Figure 1G and 1H ) . The fusion to the nuclear localization signal ( NLS ) reduces or completely prohibits the movement of proteins [26–28] , and therefore the distribution of GFP–NLS should reflect the expression pattern of the TTG1 promoter in this assay system . Consistent with the pTTG1:GUS lines , TTG1 expression is elevated in trichome initials and ubiquitously distributed in the surrounding cells ( first tier 74% , second tier 76% , third tier 77% , n = 30 ) . Depletion next to the trichome cell was not found , demonstrating that the relative distribution of TTG1–YFP differs significantly from its expression pattern . Using the Mann–Whitney U test , the strong fluorescence reduction in the first tier is highly significant ( p < 0 . 0001 ) . The difference between the homogeneous TTG1 reporter expression and the non-homogeneous protein distribution could be explained in principle by two mechanisms . First , the protein stability could be controlled spatially , such that TTG1 is more stable in trichomes than in the neighboring cells . Second , the uneven distribution could result from TTG1 movement from neighboring cells into trichomes . To determine whether TTG1–YFP depletion around trichomes is regulated by protein degradation , we treated whole pTTG1:TTG1–YFP plants with epoxomicin , a specific and irreversible inhibitor of the proteasomal degradation machinery [29] . The TTG1–YFP protein depletion around trichomes was not affected by epoxomicin treatments , suggesting that uneven distribution of the TTG1–YFP is not caused by a difference in TTG1 stability in trichome initials and its adjacent cells ( Figure 3E–G ) . As a control to show that TTG1 is an actual target of the 26S proteasome and that the proteasomal inhibitor was active , we used cotyledons of the same plants analyzed for the depletion of TTG1–YFP around trichomes on rosette leaves . TTG1 is expressed in cotyledons ( [25] , our own observation ) ; however , TTG1–YFP protein is not detectable in cotyledons of untreated plants or control plants ( Figure 3A and 3C ) . In plants treated with 20 μM epoxomicin for 24 h , TTG1–YFP protein could be detected in cotyledons , showing that the epoxomicin treatment was effective ( Figure 3B and 3D ) . Control plants treated with the solvent DMSO showed no YFP-specific fluorescence in cotyledons ( Figure 3C ) . The concept that TTG1 moves from neighboring cells into trichomes was proved by the following series of experiments . First , we demonstrated movement of the TTG1–YFP fusion protein from non-trichome cells into trichome cells , using the #232 activation tag line from the Poethig collection ( http://enhancertraps . bio . upenn . edu/default . html , line #232 ) . This line was identified as a line , driving the expression of the GAL4/VP16 activator , triggering expression of a UAS promoter driven mGFP5-ER , a GFP form localized to the endoplasmatic reticulum ( ER ) as a cell-autonomous marker . GFP-ER was expressed in an apparently random pattern but never in trichomes at any stage of development ( Figure 4A–C ) . In contrast , the TTG1–YFP fusion under the control of the UAS promoter in this enhancer trap line showed additional YFP-specific fluorescence in initiating trichomes next to epidermal cells expressing the GAL4/VP16 activator ( Figure 4A–C ) . This suggests that the TTG1–YFP fusion moved from the trichome neighboring cells , where it was expressed , into the trichome . Second , we asked whether TTG1 exerts its function in a non-autonomous manner . We used the Cre-LoxP recombination system to create ttg1 mutant sectors in plants , where wild-type-expressing cells were marked by GUS expression [30] . This was achieved by cloning the TTG1 and the GUS genes , each under the control of the CaMV 35S promoter , between the two LoxP recombination sites and by introducing this construct into ttg1–13 mutants , containing the Cre recombinase under the control of a heat-shock inducible promoter ( Figure 4D ) . These plants showed a wild-type trichome pattern due to the rescue of ttg1 by 35S:TTG1 ( Table 1 ) and ubiquitous expression of GUS . Heat shocks were applied when the first two leaves emerged . After a saturating heat treatment of 1–2 h , no GUS staining and no trichomes were detected on leaves three and four ( unpublished data ) . Heat-shock conditions ( 5–15 min ) were chosen such that a recombination event excising the 35:TTG1/35:GUS occurred rarely . These cells subsequently developed into large clonal sectors on leaves number three and four . As shown in Figure 4E and 4F , GUS-negative and therefore ttg1 mutant sectors were found that clearly exhibited trichomes . This shows that TTG1 can rescue the ttg1 mutant in a non-cell-autonomous manner . Third , we analyzed whether TTG1 protein can actively move between cells . It has been shown that soluble GFP , 2 × GFP , and 3 × GFP ( 27 , 54 , and 81 kDa , respectively ) move passively between cells with higher capacity at early stages and restricted mobility later in development [31 , 32] . Therefore , the size of a protein is not the main criterion for its ability to move between cells . Transport of molecules between plant cells is mainly regulated through plasmodesmata ( PDs ) , plant-specific channels that span the cell wall and connect plant cells with each other . In recent years , several proteins have been shown to move between cells , most likely by using the PD pathway [33 , 34] . Hence , the potential of TTG1 to act non-cell-autonomously and to move between cells raises the question whether the 38 kDa TTG1 protein moves by actively opening the PDs . To test this general biological property of TTG1 , we used microinjections in tobacco mesophyll cells ( Figure 5 and Table 2 ) . This system can be used to monitor changes in the symplasmic connectivity after injection of proteins [35] . Each set of experiments on a given leaf includes four steps . First , the injection of the small fluorescent tracer molecule acridine orange and lucifer yellow confirmed that the leaf tissue was healthy and that cells were symplasmically connected ( Figure 5A ) . Second , 11-kDa rhodamine–dextran or 12-kDa F-dextran were injected to show that molecules larger than the plasmodesmatal size exclusion limit ( SEL ) for this tissue do not move into the neighboring cells ( Figure 5B and Table 2 ) [36] . Third , the coinjection of the normally cell-autonomous 12-kDa F-dextran and TTG1 protein was done to test whether TTG1 can increase the SEL for this tracer . As shown in Figure 5C and 5D , the F-dextran moved out of the injected cell into neighboring cells in these coinjection experiments , suggesting that TTG1 increases the SEL . Fourth , to test directly whether the 38-kDa TTG1 protein can move , it was labeled with fluorescein isothiocyanate ( FITC ) or rhodamine . After injection , the fluorescent signal emitted by labeled TTG1 protein appeared within minutes in adjacent cells ( Figure 5E and 5F and Table 2 ) . GST–rhodamine and NtMPB2C–FITC were used as negative controls in these experiments [37 , 38] . Both proteins did not move and did not trigger movement of the tracer , indicating that the injection procedure as such did not change the movement behavior of the tracer or proteins in general . Thus , recombinant TTG1 protein shows an equivalent behavior in microinjection assays as the non-cell-autonomous KN1 protein [36] . These data indicate that TTG1 similar to KN1 increases the plasmodesmatal SEL and moves actively to neighboring cells via the intercellular transport pathway established by PDs . Finally , we tested the movement ability of TTG1 between cell layers . For subepidermal expression studies , we used the mesophyll-specific phosphoenolpyruvate carboxylase promoter from Flaveria trinervia ( ppcA1 ) [39] . To corroborate the specificity of the promoter in Arabidopsis , we used it to express a GFP–YFP fusion , which does not move between leaf tissue layers in Arabidopsis [40] . The GFP–YFP signal was exclusively detected in subepidermal tissue from early primordia stages on ( Figure 6A and 6B ) . In contrast , lines expressing TTG1–YFP under the ppcA1 promoter showed additional fluorescence in the epidermal layer , showing that TTG1–YFP moved from mesophyll to epidermal tissue ( Figure 6E and 6F ) . Consistent with this , cDNA expressed under the ppcA1 promoter rescued the ttg1 mutant trichome phenotype equally well as under the endogenous TTG1 promoter . Also the TTG1–YFP fusion rescued the ttg1 mutant phenotype , though less efficiently ( Table 1 ) . In young leaves , the TTG–YFP signal was found in all epidermal cells ( Figure 6E ) , whereas in older leaves it was found only in trichomes ( Figure 6F ) . This finding is consistent with the earlier observation that TTG1 is expressed only in subepidermal tissues during embryo development but is required in the protodermal tissue ( the embryonic epidermis ) [41] . To test whether trichomes can generally attract proteins or whether this is a specific property of TTG1 , we also expressed YFP under the control of the ppcA1 promoter ( Figure 6C and 6D ) . The YFP protein was observed in all cell layers in young tissues ( Figure 6C ) . However , YFP did not accumulate in trichomes ( Figure 6D ) . These data indicate that trichome-specific localization is a property of the TTG1 protein rather than due to trichome characteristics , such as a larger SEL of PDs or generally higher import rates of molecules . To understand the mechanism leading to the depletion , we tested the hypothesis that TTG1–YFP might be trapped by GL3 in trichomes . This seemed reasonable because GL3 expression is increased in trichomes relative to the surrounding cells and because GL3 strongly binds to TTG1 in yeast two-hybrid assays [12] . If the hypothesis is correct , then one would expect that TTG1–YFP would not show depletion in gl3 mutants . As shown in Figure 1I and 1J , TTG1–YFP is ubiquitously distributed in the epidermis in plants lacking functional GL3 . The quantification revealed elevated fluorescence in trichome initials and ubiquitously similar levels in the surrounding cells ( first tier 79% , second tier 77% , third tier 79% , n = 40 ) . These data strongly suggest that TTG1–YFP is depleted through trapping in trichome cells by GL3 . We used mathematical modeling to evaluate the properties of a patterning mechanism solely based on GL3/TTG1 depletion . Therefore , we neglected the influence of additional inhibitors on the patterning mechanism . The model is based on the following assumptions: ( i ) TTG1 is constantly and ubiquitously expressed ( shown in this work ) . ( ii ) TTG1 moves nondirectionally between cells . Although we show that TTG1 can actively open the PDs , there is no evidence for regulated transport affecting the actual rates . ( iii ) TTG1 forms a dimer with the GL3 protein as indicated by yeast two-hybrid results [12] . ( iv ) The AC enhances the expression of GL3 cooperatively . This is assumed because nonlinearity of the positive feedback is absolutely necessary for pattern formation . The data toward this end are not clear . At the whole plant level , it appears that GL3 is involved in a negative feedback loop [42]; however , at the current experimental resolution , these data do not contradict our assumption . Moreover , the GL3 homolog TT8 was shown to act in an autoactivation [43] . ( v ) GL3 and the AC are cell-autonomous . This assumption is based on the observation that GL3 protein does not move in the leaf ( unpublished data ) . ( vi ) All components are degraded by first-order kinetics . The corresponding interaction scheme is shown in Figure 7A . Because the model parameters are unknown , we employed a two-step approach . First , a rescaling of model variables allowed the confinement of the parameters to relevant ranges . Second , we fitted the resulting model to the experimentally obtained relative fluorescence intensities of TTG1 in the vicinity of the trichomes . Fitting of the parameters also took into consideration the mean trichome density in the initiation zone . For parameter values and details of the optimization , see the Materials and Methods section . A typical simulated concentration pattern of total TTG1 ( i . e . , TTG1 + AC ) is presented in Figure 7B . The highest TTG1 levels are found in the trichomes where it is completely bound to GL3 . In cells adjacent to trichomes , the level of unbound TTG1 is significantly lowered by depletion , while the level increases with distance from the trichomes . Our rescaling and fitting procedure enabled us to estimate the model parameters and in turn to judge their relevance . We focused on the dependence of trichome density and clustering on parameters related to TTG1 function ( Figure 7C ) . Here , trichome density is defined as the ratio of trichome cells to the total number of epidermal cells in the initiation zone of the young leaf . A decrease of the degradation rate λ3 of the AC ( cyan line , circles ) or of the transport rate d of TTG1 ( green line , squares ) results in an elevated trichome density/clustering . Conversely , an increase in the complex formation rate β ( blue line , diamonds ) raises the trichome density/clustering . Surprisingly , the trichome density/clustering is unaffected by a decreased degradation rate λ1 of TTG1 ( red line , triangles ) . The increase of trichome density is correlated with a corresponding change of the percentage of the trichomes found in clusters ( Figure 7C , inset ) . Note that blunt ends correspond to a loss of the trichome pattern; e . g . , a decreased complex formation rate leads to glabrous plants . These data provide for the first time an explanation for the apparently paradoxical observation that strong ttg1 alleles are glabrous ( suggesting a positive function ) and weak ttg1 alleles show clusters ( suggesting an inhibitory function ) . While it is trivial that the absence of TTG1 in this model causes a glabrous phenotype , surprisingly , simulations of the depletion mechanism revealed that alterations of all parameters , except for the protein degradation rate , can lead to clusters .
In this study , we focus on the functional analysis of TTG1 in trichome patterning on Arabidopsis leaves . We show that TTG1 is ubiquitously expressed with slightly higher levels in developing trichomes . The distribution of TTG1–YFP differs from the expression pattern such that the signal is strongly reduced in cells immediately next to the trichome . In showing that the proteasome inhibitor epoxomicin does not affect the protein distribution , we exclude the possibility that differential protein degradation results in the local depletion of TTG1–YFP around incipient trichomes . We demonstrate that TTG1 acts non-cell-autonomously by clonal analysis and that the TTG1–YFP protein can move within the epidermis into trichomes by using a GAL4-based expression system . Further , we show that TTG1–YFP can move between cell layers and that the TTG1 protein can open actively PDs in a heterologous system . Together these data suggest that TTG1 is redistributed from neighboring cells into the trichome by intracellular movement . What is the underlying mechanism of the observed depletion/attraction of TTG1 ? One possibility is that TTG1 moves freely and becomes trapped in trichomes . Alternatively , the redistribution could be achieved by directional movement into the trichomes , although both mechanisms do not necessarily rule out each other . The latter scenario is similar to that proposed for the function of auxin in the positioning of primordia in the meristematic region [44] . In this system , directional transport of auxin by the transporter PIN1 leads to an accumulation of the hormones in primordia and a reduced level of auxin in the neighborhood [44 , 45] . A directional transport similar to auxin is unlikely for TTG1 because TTG1–YFP can move from the cells , expressing it not only into trichomes but also into other epidermal cells ( Figure 4A–C ) . We therefore hypothesized that TTG1 accumulates in trichomes , because it binds to another protein , as suggested for SHORT ROOT ( SHR ) in the root [46] . SHR is expressed in the stele and moves specifically into the endodermis , where it is required and sequestered in the nucleus due to interaction with SCARECROW [46] . In support of this hypothesis , we find no depletion of TTG1–YFP in gl3 mutants , indicating that TTG1 binding to GL3 causes the depletion . Current models explaining trichome patterning on Arabidopsis leaves are based on the activator–inhibitor-like mechanisms described above [2–4 , 47] . These mechanisms can explain the generation of a pattern in the absence of pre-existing positional information . However , not all aspects of the model have been shown experimentally . The mobility of the inhibitors was shown for CPC in the root system [48] , but nothing is known about the mobility in leaves . Moreover , the theoretical requirement that the activators can autoactivate lacks experimental proof . Another problem with the current models is that various genetic data cannot be explained [3] . Our finding that in addition to the activator–inhibitor mechanism a substrate-depletion-like mechanism is operating during trichome patterning may provide some missing clues . In general , a substrate-depletion mechanism is superficially similar to the activator–inhibitor mechanism . Instead of producing an inhibitor that laterally suppresses trichome development in cells next to a developing trichome , a factor necessary for trichome development is removed from these cells . When simulating this type of mechanism , however , it turned out that the system properties are different [1 , 49] . In particular , it was noted that new peaks are formed at the maximum distance by the activator–inhibitor mechanism and by splitting already existing ones by the substrate-depletion mechanism [1 , 49] . To understand the properties of the GL3/TTG1 trapping mechanism , we formulated a mathematical model and fitted it to our experimental data to obtain a biologically relevant parameter range . This strategy enabled us to test how parameter changes affect patterning . In particular , we aimed to simulate the weak ttg1 cluster phenotype as this genetic finding was the most confusing , because the lack of trichomes in strong ttg1 mutants suggested that TTG1 functions as a trichome-promoting factor and the cluster phenotype in weak ttg1 mutants pointed toward a role as a negative regulator [5 , 13 , 24 , 50 , 51] . The simulations of the GL3/TTG1 trapping mechanism revealed that changes of several parameters related to TTG1 function can result in a clustering phenotype . Thus , we can offer for the first time explanations for the apparently paradoxical genetic results on TTG1 with our new GL3/TTG1 trapping/depletion model . However , our reduced model can only partially capture the experimental observations . For example , the simulated mean trichome density as predicted by the optimal parameter set is still substantially larger than that in the wild type . We expect that more complex models involving additional patterning genes will improve the agreement between theory and experiment . As GL3 is also a central component of all activator–inhibitor-based models , it is conceivable that the two models act in concert . We can recognize TTG1–YFP depletion at the earliest stages of morphologically recognizable trichome development . This would suggest that the trapping/depletion mechanism becomes relevant after the activator–inhibitor mechanism already has started the selection of trichomes . However , it is well possible that more sensitive microscopic techniques and more sophisticated imaging analysis tools will reveal the depletion much earlier , so we consider the relative timing of the two processes to be elusive at the moment . It will be a future challenge to combine both principles in a single model . To operate in biologically reasonable parameter ranges , it will be crucial to base such a model not only on qualitative but also on quantitative data .
In this study , the wild-type ecotypes Landsberg erecta ( Ler ) and RLD were used . The ttg1-1 , -9 , -10 , and -13 and gl3-1 mutant lines have been described previously [14 , 24 , 52] . The Poethig activation tag line #232 ( Columbia ecotype ) was a kind gift from Scott Poethig , University of Pennsylvania ( http://enhancertraps . bio . upenn . edu/default . html ) . The heat-shock inducible HSP:CRE3 line containing the pCGNHCN construct in a Nossen ecotype background [30] was crossed into the ttg1-13 mutant line ( RLD background ) , and plants homozygous for both the transgene and the ttg1-13 allele were isolated and crossed to TTG1-Lox lines . The TTG1-Lox construct is a descendant from the pCGNLox2a construct [30] , introducing a 35S:TTG1:NOSpA cassette into the PmeI site of pCGNLox2a . The resulting plants of these crossings were used for heat-shock treatments . Plants were grown on 1 × Murashige Skoog agar ( 1% sucrose ) plates for approximately 10 d at 20 °C under 16 h light/8 h dark conditions . Heat shock was performed by placing the plates into an illuminated incubator at 41 °C for 10–15 min . All transgenic lines were produced using the floral dip method [53] . The TTG1 promoter ( position −2227 to −1 from the start codon and includes the 110 nucleotide of the 5′ UTR ) was isolated from Arabidopsis thaliana ecotype Ler by PCR ( forward primer , 5′-AAAGCTTAACCGAGAATGTCTCCCGACTTCTAT-3′; reverse primer , 5′-AGTCGACTCAAACTCTAAGGAGCTGCATTTG-3′ ) and cloned into pGEM-T vector ( Promega Corporation ) ( pTTG-pGEM ) . An AscI restriction site was added by adapter ligation ( 5′-CTAGAATGGCGCGCCATT-3′ ) into the SpeI site of the vector . To generate the pTTG:GUS construct , the pTTG-pGEM was digested with AscI and SalI , and the resulting fragment was cloned into the binary gateway vector pAM-PAT-GW-GUS ( GenBank accession AY02531 ) to replace the existing CaMV 35S promoter between the AscI and the XhoI sites . To create the pPPCA1-pAMPAT binary vector , the 2117 bp promoter fragment of the phosphoenolpyruvate carboxylase 1 gene ( ppcA1 ) from Flaveria trinervia ( GenBank accession X64143 ) [39] was removed from ppcA1-pBS 5′ with HindIII and religated using an oligonucleotide linker to generate an AscI restriction site . The resulting AscI–XhoI fragment was inserted into pAM-PAT-GW using the same restriction enzymes . The yeast UAS promoter was PCR-amplified with the attachments of AscI for the forward primer and XhoI for the reverse primer . The corresponding fragment was ligated into pAMPAT-GW by exchanging the existing CaMV 35S promoter using AscI and XhoI , giving rise to pUAS-pAMPAT . The TTG1 cDNA ( GenBank accession AT5G24520 . 1 ) was PCR-amplified with attB1 forward and attB2 reverse linker primers for Gateway BP recombination with the pDONR201 vector ( Invitrogen ) . To create the TTG1–YFP fusion , the TTG1 cDNA was PCR-amplified again to add a SalI site at the 5′ and a XhoI site at the 3′ of the coding sequence deleting the stop codon ( forward primer , 5′-AGTCGACATGGATAATTCAGCTCCAGA-3′; reverse primer , 5′-ACTCGACAACTCTAAGGAGCTGCATTT-3′ ) . The digested fragment was ligated into the SalI site of pUC18 , then a XbaI–SacI EYFP fragment ( Clontech ) was fused C-terminally to TTG1 using the same sites . The fusion was isolated using XhoI and EcoRI and ligated into pEN1a SalI–EcoRI fragment . The resulting construct was called TTG1–YFPpEN . pEYFP ( Clontech ) was digested with SalI and NotI and ligated into pEN1a to create EYFPpEN . The GFP–YFP fusion was constructed using an NcoI fragment of mGFP4 , which was ligated in frame into the NcoI site of EYFPpEN . All constructs were sequenced . To create all of the binary constructs or yeast two-hybrid vectors , the Gateway LR Reaction System was used according to the user's manual ( Invitrogen ) . GUS activity was assayed as described previously [54] . After adding the X-Gluc-solution ( 5-bromo-4-chloro-3-indolyl-β-d-glucuronic acid ) , plants were vacuum-infiltrated for 15 min and then incubated at 37 °C overnight . The tissue was cleared by an ethanol series ( 15% , 30% , 50% , and 70% EtOH solutions at 37 °C for several hours ) . Seed coat mucilage staining was done with a 0 . 01% ruthenium red solution for 15 min . Light microscopy was performed using a Leica DMRE microscope using differential interference contrast optics . Images were taken using a KY-F70 3-CCD JVC camera and DISKUS software ( DISKUS , Technisches Büro ) . Confocal laser scanning microscopy was done with a Leica TCS-SP2 confocal microscope equipped with the Leica software Lite 2 . 05 ( LCS , Leica Microsystems ) . Z-stacks in steps of 1 or 2 μm were taken and processed using deconvolution tools of the Leica software . Quantification of fluorescence was performed using the same software . Plants were incubated for 10–15 min with a 10 μg/ml propidium iodide solution to visualize cell walls . Transverse sections were generated by embedding the tissues in 4% low-melting-point agarose and by hand sectioning using a razor blade as described by [55] . Images were assembled and processed using GIMP 2 . 2 software ( http://www . gimp . org ) . Recombinant TTG1 protein was produced in Escherichia coli , labeled , purified , and microinjected as previously described [38 , 56] . The protein concentration used for microinjection was 2 μg/μl . A Leica SP2 AOBS UV confocal microscope was employed to detect the fluorescent probes after microinjection . Tissues were scanned in sequential mode to excite and detect fluorescence probes in their specific wavelengths , and the resulting Z-stack ( 5 μm distance ) images were merged using the NIH image software ImageJ ( version 1 . 32j ) ( http://rsb . info . nih . gov/ij/ ) . pTTG1:TTG1–YFP plants were grown on Murashige Skoog agar plates containing 1% sucrose at 22 °C for 6 d under 16 h light/8 h dark conditions and then transferred into liquid ½ MS medium containing 1% sucrose . The medium contained either 2% DMSO ( control ) or 20 μM epoxomicin ( Sigma-Aldrich , stock solution in DMSO ) . The samples were vacuum-infiltrated for 15 min and incubated under the same growth conditions as previously for 24 h . After being washed with ½ MS ( 1% sucrose ) , plants were analyzed using confocal laser scanning microscopy ( see above ) . Yeast two-hybrid interaction assays were performed as described previously [9] . Fusions with the GAL4 activation domain and GAL4 DNA-binding domain were performed in the pACT and pAS plasmids ( Clontech ) . TRY , GL3 , and a truncated version of GL3 lacking 96 amino acids at the N-terminus were fused to the GAL4 activation domain in the pACT vector . TTG1 and TTG1–YFP were fused to the GAL4 DNA-binding domain of pAS . None of the constructs or empty vectors showed any self-activation in yeast . Fifteen 10 d old plants ( long day conditions , 24 °C ) were harvested without roots , frozen in liquid nitrogen , and afterwards ground . The powder was mixed and boiled in 300 μl of sample buffer ( 50 mM Tris/HCl . pH 6 , 8 , 2% ( w/v ) SDS , 8 M urea , 30% ( v/v ) glycerol , 5% ( v/v ) β-mercaptoethanol , and 0 . 5% ( w/v ) bromphenol blue ) for 15 min followed by centrifugation ( 16 , 000g at 4 °C ) for 15 min . Approximately 25 μl of the supernatant was analyzed by 12% SDS-PAGE , which was followed either by Coomassie staining or by western blotting and subsequent immunodetection with anti-GFP monoclonal IgG mouse antiobody ( Roche ) . Detection was done by electrochemiluminescence . On the basis of the interaction diagram presented in Figure 7A , a system of coupled ordinary differential equations was derived that describes the temporal evolution of the protein concentrations of TTG1 , GL3 , and the AC inside each cell . The model was formulated on a two-dimensional grid of hexagonal cells with the cell index j = ( y , x ) , where 1 ≤ y ≤ N and 1 ≤ x ≤ M . N and M denote the number of cells in the y and x directions , respectively . Periodic boundary conditions were chosen for model simulation and analysis . The nondirectional transport of TTG1 between cell j and its six neighboring cells is characterized by the coupling term The model includes parameters αi for the expression of TTG1 and GL3 and parameters λi for the degradation of the single proteins and the active complex . The parameter d is the transport rate of TTG1 between neighboring cells and the parameter β is the rate of active complex formation . To allow an assignment of reasonable parameter ranges and to reduce the number of model parameters a rescaling of the model variables was applied . All concentrations were multiplied by the factor β/λ3 , and the new dimensionless time was expressed as τ = tλ3 . The transformed , but mathematically equivalent , dimensionless equations are The relation between the dimensional and the dimensionless parameters k1 to k5 is given in Table 3 . Let ν0 ( i ) = ( [ttgl]0 ( i ) , [gl3]0 ( i ) , [ac]0 ( i ) ) T denote the ith uniform steady state . Equations 1–3 have three uniform steady states given by where f = ( ( k1k4 – 1 ) 2 – 4k2k4k5 ) 1/2 . For biological relevance , all three steady states must be real and positive , which restricts the range of possible values for parameters ki . In a pioneering work , Turing introduced the concept of pattern formation from homogeneous conditions by a diffusion driven instability; a uniform steady state that is stable for a single cell can be driven unstable by the interaction between cells [57] . On the basis of the idea of Turing , the criteria for pattern formation from a uniform steady state were derived in two steps: ( i ) criteria for the stability of the steady state without TTG1 mobility and ( ii ) criteria for an instability of the uniform steady state when adding TTG1 mobility . The stability of the steady in the absence of TTG1 mobility was analyzed by a linearization of equations 1–3 leading to ∂τΔν ( i ) = J ( i ) Δν ( i ) . Here , Δν ( i ) = ν ( i ) – ν0 ( i ) are small deviations from the ith steady state , and J ( i ) is the Jacobian matrix evaluated at steady state ν ( i ) . A steady state is stable if small deviations from it decay with time . This is the case if all eigenvalues of the Jacobian matrix have negative real parts [58] . The eigenvalues of J ( i ) are the roots of the characteristic equation λ3 + a1 ( i ) λ2 + a2 ( i ) λ + a3 ( i ) = 0 with the coefficients All three roots have negative real parts if the following three necessary and sufficient criteria for the coefficients of the characteristic equation are fulfilled [58] Next , we considered the stability of the steady state ν0 ( i ) including the mobility of TTG1 . The temporal evolution of small spatially inhomogeneous deviations Δνj ( i ) = ν0 ( i ) – νj from the uniform steady state ν0 ( i ) are again described by a linearization of Equations 1–3 , now including the cellular coupling ∂τΔνj ( i ) = J ( i ) Δνj ( i ) + D〈Δνj ( i ) 〉 . The matrix of transport coefficients is D and has a single entry for [ttg1] at D11 = k3 . Fourier analysis was used to study the temporal evolution of spatially periodic solutions of the form Δνj ( i ) = ∑s=1N∑r=1Mφs , r ( i ) e2πisy/N e2πirx/M . The transformed linear equations read ∂τφs , r ( i ) = ( J ( i ) – 4Dg ( s , r ) ) φs , r ( i ) with the function g ( s , r ) = sin2 ( πs/N ) + sin2 ( πr/M ) + sin2 ( π ( s/N – r/M ) ) . The uniform steady state ν0 ( i ) becomes unstable to small spatial variations if any of the eigenvalues of the matrix J ( i ) – 4Dg ( s , r ) has a positive real part . The eigenvalues of J ( i ) – 4Dg ( s , r ) are roots of the characteristic equation λs , r3 + b1 ( i ) ( s , r ) λs , r2 + b2 ( i ) ( s , r ) λs , r + b3 ( i ) ( s , r ) = 0 with the coefficients: b1 ( i ) ( s , r ) = ( a1 ( i ) + 4Dg ( s , r ) , b2 ( i ) ( s , r ) = a2 ( i ) + 4Dg ( s , r ) ) ( 1 + k5 + [ttgl]0 ( i ) ) , and b3 ( i ) ( s , r ) = a3 ( i ) + 4Dg ( s , r ) ( k5 + [ttgl]0 ( i ) – 2k4[ttgl]0 ( i ) [ac]0 ( i ) ) . If any of the three necessary and sufficient criteria are violated , then the ith steady state gives rise to a Turing instability . For the analysis , we restricted all parameters ki to be real and positive . Analysis of steady state ν0 ( 1 ) revealed that conditions C1a–C1c and C2a–C2c are always fulfilled . Furthermore , if both steady states ν0 ( 2 ) and ν0 ( 3 ) are real and positive , then only ν0 ( 3 ) fulfills conditions C1a–C1c . Therefore , only steady state ν0 ( 3 ) was considered in the following . For a given parameter set , all six conditions were verified numerically . Here , it is sufficient for Turing instability if conditions C2a–C2c are violated at the maxima of g ( s , r ) . The parameter optimization was confined to the region in parameter space that gave rise to a Turing instability of steady state ν0 ( 3 ) as defined by the criteria given above . Additionally , parameters were restricted to the biological reasonable ranges given in Table 3 . Parameters were estimated by fitting the model Equations 1–3 to the experimentally determined relative fluorescence intensities of TTG1 in the vicinity of the trichomes as well as the mean trichome density in the initiation zone of the young leaf . The optimized function was with k = ( k1 , k2 , k3 , k4 , k5 ) . The trichome number T ( k ) was determined from a numerical solution of Equations 1–3 . The uniform steady state ν0 ( 3 ) plus a small inhomogeneous perturbation were used as the initial conditions . The average total [ttg1] level of the cells in tier j around trichome i is Pi , j ( k ) . It was normalized by the total [ttg1] level in trichome i; i . e . , Pi ( k ) . Rj is the experimentally determined average relative TTG1 level in tier j , and σR , j is the corresponding standard deviation . The levels are R = ( 0 . 387 , 0 . 765 , 0 . 935 ) , and the standard deviation is σR = ( 0 . 14 , 0 . 22 , 0 . 183 ) . For the mean trichome density in the initiation zone , we used μD = 0 . 075 with the corresponding standard deviation σD = 0 . 035 . Both values reflect the experimental observation that the mean trichome distance in the initiation zone is between 3 and 5 cells . Because the numerical solution of T ( k ) and Pi , j ( k ) depends on the initial conditions , the optimal parameter set also depends on the initial conditions . Therefore , optimal parameters were averaged across 10 optimizations to determine the mean and standard deviation given in Table 3 . For each of the 10 optimizations , a different random perturbation of the initial conditions was chosen . Parameters k4 and k5 cannot be determined simultaneously from the data . To resolve this nonidentifiability , we fixed k5 = 1 . Global optimization was performed using an algorithm based on adaptive simulated annealing ( Lester Ingber , http://www . ingber . com ) in combination with the MATLAB interface ASAMIN by Shinichi Sakata ( http://www . econ . ubc . ca/ssakata/public_html/software/ ) . All numerical analysis was performed with MATLAB from Math Works , Inc . The predicted mean trichome density and mean percentage of the trichomes in clusters given in Figure 7C were determined from an average over 100 simulations for each parameter set . Accession numbers for genes mentioned in this paper from the National Center for Biotechnology Information ( http://www . ncbi . nlm . nih . gov ) are: pAM-PAT-GW-GUS ( AY02531 ) , ppcA1 ( X64143 ) , and TTG1 ( AT5G24520 . 1 ) . | Trichomes , the specialized hair cells found on plant leaves , represent a model system to study how cellular interactions coordinate the development and arrangement of a collection of initially equivalent cells into regularly placed specialized cells . It was assumed that a regulatory feedback loop of positively and negatively acting factors governs these decisions . In this work , we show that trichome spacing also is controlled by the local depletion of the trichome-promoting protein TTG1 . We provide evidence that binding of TTG1 to a second trichome-promoting protein , GL3 , causes a depletion of TTG1 in the neighborhood of cells with elevated GL3 levels . We postulate that this leads to trichome fate determination in cells containing high GL3/TTG1 levels and prevents trichome formation in surrounding cells because of the reduced TTG1 levels . We show by theoretical modeling that this mechanism alone is capable of creating a spacing pattern and has properties that can explain even apparently paradoxical genetic observations . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"cell",
"biology",
"developmental",
"biology"
] | 2008 | Two-Dimensional Patterning by a Trapping/Depletion Mechanism: The Role of TTG1 and GL3 in Arabidopsis Trichome Formation |
Many species have experienced dramatic changes in their abundance and distribution during recent climate change , but it is often unclear whether such ecological responses are accompanied by evolutionary change . We used targeted exon sequencing of 294 museum specimens ( 160 historic , 134 modern ) to generate independent temporal genomic contrasts spanning a century of climate change ( 1911–2012 ) for two co-distributed chipmunk species: an endemic alpine specialist ( Tamias alpinus ) undergoing severe range contraction and a stable mid-elevation species ( T . speciosus ) . Using a novel analytical approach , we reconstructed the demographic histories of these populations and tested for evidence of recent positive directional selection . Only the retracting species showed substantial population genetic fragmentation through time and this was coupled with positive selection and substantial shifts in allele frequencies at a gene , Alox15 , involved in regulation of inflammation and response to hypoxia . However , these rapid population and gene-level responses were not detected in an analogous temporal contrast from another area where T . alpinus has also undergone severe range contraction . Collectively , these results highlight that evolutionary responses may be variable and context dependent across populations , even when they show seemingly synchronous ecological shifts . Our results demonstrate that temporal genomic contrasts can be used to detect very recent evolutionary responses within and among contemporary populations , even in the face of complex demographic changes . Given the wealth of specimens archived in natural history museums , comparative analyses of temporal population genomic data have the potential to improve our understanding of recent and ongoing evolutionary responses to rapidly changing environments .
Rapid environmental change threatens global biodiversity and has led to population declines in many species [1–5] . Although phenotypic plasticity may enable populations to track rapidly changing climates , evolutionary adaptation will often be essential for long-term persistence [6] . Disentangling plasticity from evolutionary responses ultimately requires resolving the genetic basis of adaptation . However , it remains challenging to differentiate recent or ongoing positive selection from stochastic genetic drift in contemporary populations that are also undergoing extreme demographic changes [7 , 8] . Natural history museum collections may hold the key to overcoming many of these difficulties by providing crucial temporal information on species distributions , phenotypes , and population genetic variation spanning periods of recent environmental change [9–11] . Temporal genomic contrasts have begun to yield powerful insights into recent evolutionary responses in humans [12 , 13] and other species [14–17] , indicating that genetic analyses of biological archives will be an effective tool for understanding evolutionary responses to rapid anthropogenic climate change [18] . Using contrasts between early 20th century and modern museum surveys , Moritz and colleagues [1] showed that the ranges of several high elevation small mammal species in the Yosemite National Park ( YNP ) region of the Sierra Nevada mountains ( California , USA ) have retracted upward over the past century . This and associated studies [19 , 20] demonstrated the potential of using museum archives to understand species and community-level ecological responses during periods of recent climate change . Contemporary range shifts towards higher latitudes , elevations , or both have now been documented in many terrestrial species [5 , 19 , 21 , 22] , and are generally thought to reflect direct and indirect population responses to warming temperatures [22 , 23] . However , these works have also highlighted that closely related species can differ markedly in the magnitude and direction of their ecological responses , for reasons that are often not clear [1 , 20] . Currently , we know very little about how recent range shifts have affected evolutionary processes within species , or the extent to which evolutionary genetic responses have been synchronous within and between co-distributed species . Here we focus on two chipmunk species within the YNP montane mammal community that show different ecological and phenotypic responses over the last century of climate change ( Fig 1 ) . Western chipmunks have long been considered models for niche partitioning by elevation and habitat type [24–26] . The alpine chipmunk ( Tamias alpinus ) is an ecological specialist endemic to the high elevation alpine habitats of the Sierra Nevada Mountains . The lodgepole chipmunk ( T . speciosus ) occurs more broadly across mid- to high-elevation subalpine coniferous forests of California . Tamias alpinus has undergone severe range contraction driven by extirpation of lower elevation populations [1 , 20] combined with pronounced shifts in diet and cranial morphology [27] across the alpine zone of YNP and elsewhere in the Sierra Nevada mountains . Spatial modeling of current versus historical ranges across YNP suggests that the strong upward contraction of T . alpinus is best explained by increases in minimum winter temperatures; competing models including co-occurrence with other species of chipmunks or changes in the distribution of preferred vegetation types did not improve prediction of the observed range contraction in this species [28 , 29] . That increasing minimum temperature alone had the strongest impact makes sense in that there has been little change in vegetation across the high elevation talus slopes preferred by this species . By contrast , range contraction in a mid-elevation chipmunk species , T . senex , is best explained by changes in its preferred vegetation types [28 , 29] . For YNP T . alpinus , there is also evidence of strong directional selection on skull morphology over the past century [30] , whereas the range , diet , and morphology of the partially overlapping lodgepole chipmunk ( T . speciosus ) has remained relatively stable within YNP [20 , 27 , 31] . As is generally the case [32] , it remains unclear why the montane specialist , T . alpinus , has contracted during the past century whereas its more widely-distributed congener , T . speciosus , has not . These two species set up a natural contrast , allowing us to understand how differing ecological responses during periods of rapid environmental change correspond to differing evolutionary responses . A previous temporal survey of eight microsatellite markers revealed increased subdivision and declining allelic diversity in YNP T . alpinus , but no significant changes in overall population genetic variation of T . speciosus over the same interval [33] . Synthesizing descriptions of phenotypic [30] , behavioral [31] , and genetic variation [33] into a detailed understanding of demographic and evolutionary responses in these species requires genomic data . Towards this end , Bi et al . [9] used a custom exon capture platform to enrich and sequence ~11 , 000 exons ( ~4 Megabases or Mb across 6 , 249 protein-coding genes ) from 20 early 20th century and 20 modern YNP T . alpinus . These genome-wide data confirmed signatures of increasing population subdivision in this retracting species and illustrated the potential for targeted genomic experiments to generate high quality data from archived specimens . However , neither of these preceding analyses had sufficient sampling to determine whether evident increases in population structure were due to reductions in local population size , migration rates , or both . Here , we build on these previous works [9 , 33] by generating ~9 . 4 Mb of targeted exome sequence data from 303 chipmunks ( 194 T . alpinus , 100 T . speciosus , and 9 samples from 4 other species ) . We used these comparative population genomic data to quantify evolutionary responses over the past century at two scales ( Fig 1 ) . First , we sequenced 96 modern ( 48 T . alpinus and 48 T . speciosus collected between 2003–2008 ) and 108 historic ( 56 T . alpinus , 52 T . speciosus collected in 1915–1916 ) samples collected across geographic transects in YNP for the focal species . Second , we generated an independent geographic transect of 38 modern ( 2003–2012 ) and 52 historic ( 1911–1916 ) T . alpinus in the Southern Sierras ( SS ) , where this species also shows range contraction [20] , to test to what extent evolutionary responses across the range of this declining alpine specialist have been consistent . To analyze these temporal data , we developed a novel analytical framework based on Approximate Bayesian Computation ( ABC ) that allowed us to quantify changes in population sizes and migration rates in the context of demographic history , and then to localize positive selection on standing genetic variation at specific genes . Our results provide new insights into recent evolutionary responses in this system and demonstrate how a genomic time-series approach can be broadly applied to other archived specimens to improve understanding of evolutionary responses over the past few centuries .
High sequencing coverage is necessary to reliably genotype ancient DNA samples [34] due to extensive DNA degradation [35] . This persistent technical challenge makes whole genome resequencing of historic mammalian populations impractical , especially in species without high-quality reference genomes ( e . g . , Tamias ) . Therefore , we designed a custom targeted capture to enrich and sequence exons from over 10 , 107 protein-coding genes ( 9 . 4 Megabases or Mb total ) in 294 T . alpinus and T . speciosus samples ( S1 Table ) and nine samples from four other chipmunk species ( total n = 303 ) . We sampled geographic transects of modern and historic ( ~100 year-old ) populations in YNP for both species as well as an independent SS transect of T . alpinus , where this species has also contracted [20] . An average of 49 individuals were sequenced per population ( Fig 1 ) . This design allowed us to ( i ) compare stable and retracting species within the same montane mammal community ( YNP ) , and to ( ii ) determine to what extent the same evolutionary responses have occurred across two transects ( YNP and SS ) spanning the latitudinal range of the range-retracted species T . alpinus . Exome enrichment was highly specific ( 90–93% of cleaned reads on target ) and sensitive ( >92% of the target regions sequenced ) , resulting in high coverage of targeted regions ( 26–35× average individual coverage per population , S2 Table ) . Although historic DNA samples are notorious for poor technical performance , all targeted T . speciosus and T . alpinus individuals yielded moderate to high coverage data with similar capture success between modern and historic samples . Analysis of mitochondrial DNA indicated that empirical error rates were ~fourfold higher in historic ( 0 . 16% ) versus modern samples ( 0 . 04% ) , due primarily to DNA damage typical of century-old museum samples [9 , 36] . Although nucleotide misincorporations associated with deamination of methylated cytosine bases ( C-to-T and G-to-A ) were most common near the ends of DNA fragments , such changes remained elevated throughout the sequence ( S1 Fig ) . Therefore , we applied several quality filters to remove all putative misincorporations and to mitigate other common sources of genotyping error [9] ( S3 Table ) . All filters were uniformly applied to historic , modern , and simulated data ( see below ) to facilitate comparisons across time periods and species . After filtering , we identified 20 , 395 , 10 , 395 , and 10 , 954 high-quality single nucleotide polymorphisms ( SNPs ) in YNP T . speciosus , YNP T . alpinus , and SS T . alpinus , respectively . Upwards range contraction in montane environments can lead to increased population structure and reduced genetic diversity due to decreased population size . We observed a consistent trend towards relatively minor reductions in nucleotide diversity in modern versus historic samples of T . alpinus and T . speciosus estimated at two spatial scales: metapopulations ( e . g . , YNP or SS ) and demes of spatially clustered sampling localities ( θπ and θW; S2 Fig , S4 Table ) , consistent with prior results [33] . We then quantified the degree of population genetic structure within each species by estimating the global fixation index ( FST ) for historic and modern populations . Within YNP , population structure was relatively low overall but increased nearly two-fold in modern T . alpinus ( FST , historic = 0 . 032 , FST , modern = 0 . 058 ) , consistent with increased population fragmentation as this species has contracted upwards [33] . By contrast , population structure in T . speciosus increased only slightly over time ( FST , historic = 0 . 027 , FST , modern = 0 . 030 ) . We detected very little uncertainty in our estimates of the site frequency spectrum ( SFS ) , and estimates of global FST showed non-overlapping 0 . 95 bootstrapped confidences intervals for all pairwise temporal contrasts ( S4 Table ) . Collectively , these patterns suggest that the stronger increase in FST observed for YNP T . alpinus is not simply explained by reductions in per deme nucleotide diversity , which are of similar magnitude across YNP and SS T . alpinus and YNP T . speciosus ( S2 Fig ) . To further evaluate changes in the spatial patterning of population genetic structure , we used NGSadmix to conduct a maximum likelihood ( ML ) analysis of historic and modern population structures [37] . We detected a substantial increase from two to six genetic clusters for YNP T . alpinus , compared to stable overall population structure ( K = 2 across both time points ) within T . speciosus ( Fig 2A; S5 Table ) . Principal component analyses of these data also indicate less genetic similarity among modern YNP T . alpinus samples ( Fig 2B ) . These genome-wide estimates incorporate genotype uncertainty to provide an accurate overview of changes in genetic diversity and structure over recent timescales . Although consistent in the direction of change , the increases in YNP T . alpinus population genetic structure appear even more striking than previously detected using lower resolution data [33 , 38] . By contrast , there was minimal structure in T . speciosus within the same ecosystem whether considering a few microsatellite loci [33] or thousands of SNPs ( Fig 2A ) . Consistent with patterns of increased population genetic structure , spatial models of occurrence indicate that range contraction has reduced local connectivity between suitable habitat patches for YNP T . alpinus [33] . Given that range contraction has also been detected at the southern limit of the range of T . alpinus [20] , we next tested to what extent similar temporal signatures of increased genetic structure were apparent in SS populations ( Fig 1 ) . Population structure in the SS transect has also increased ( FST , historic = 0 . 034 , FST , modern = 0 . 044 ) , but to a lesser extent than observed in YNP , and with no overall increase in the number of distinct genetic groups ( modern and historic K = 2; Fig 2A ) . This suggests some variation in the local genetic consequences of seemingly synchronous , range-wide contractions . However , we note that our power to detect changes in overall population structure may have been limited by the fact that historic and modern sampling localities were more spatially clustered in SS when compared to YNP ( Fig 1 ) . Rapid range shifts could also increase the likelihood of hybridization if there are changes in the degree of sympatry or in relative densities among closely related species [39 , 40] . Occasional hybridization appears to be common in western chipmunks [41] , including evidence for ancient mitochondrial introgression from the broadly distributed least chipmunk , T . minimus , into T . speciosus [42] . The alpine chipmunk is very closely related to T . minimus with evidence of historical gene flow [38] , raising the possibility that recent range fluctuations have induced hybridization between T . alpinus and either T . minimus or T . speciosus . However , we found no evidence for recent appreciable nuclear gene flow between T . alpinus and adjacent populations of T . speciosus or with neighboring ( lower elevation ) populations of T . minimus ( S3 Fig ) . Thus , recent range collapse does not appear to have led to the breakdown of reproductive barriers between this high elevation endemic and other co-distributed species [38] . We did , however , detect nuclear introgression from T . speciosus into at least one T . minimus sample ( S3 Fig ) , suggesting that reproductive isolation remains incomplete between these species despite strong ecological partitioning [26] . These basic descriptions of genetic variation expand the scope and resolution of previous analyses of a few microsatellite loci genotyped in T . alpinus , T . minimus , and T . speciosus [33 , 38] , and from more limited exon capture data from T . alpinus [9] . However , disentangling changes in migration versus local effective population size , and identifying genes under positive selection in the context of recent demographic change , requires more comprehensive analyses . Therefore , we developed a novel analytical framework to fully exploit our temporal dataset . Traditional population genetic analyses often assume even sampling across space and time , yet studies using museum specimens are typically imbalanced because of limited availability of samples . ABC is well suited for demographic inference under such circumstances because biased temporal sampling and sample processing can be simulated for populations that have experienced complex demographic histories . Accordingly , simulations can be filtered in the same way as observed data ( e . g . , removal of SNPs associated with errors in historic DNA ) , permitting meaningful comparisons between expected and observed results . A major difficulty of ABC is the choice of statistics that sufficiently describe demographic parameters of interest . Multiple , jointly informative , summary statistics are often used to sufficiently estimate parameters while reducing the risk of any particular statistic biasing the results [43] . The site frequency spectrum ( SFS ) is often an optimal choice for fitting demographic histories since many commonly used summary statistics can be derived from it . In practice , high dimensionality and low count categories of joint site frequency spectra make them difficult to fit . Consequently , we developed an effective means of fitting binned two-dimensional SFS ( 2D-SFS ) using an ABC framework designed to infer population histories from serially sampled metapopulations ( S4 Fig ) . We constructed 2D-SFS for each of the three pairwise temporal contrasts by pooling individuals across sampling localities within YNP or SS into a single metapopulation per time period ( S5 Fig ) . While allele frequencies should be highly correlated over such short time scales , the overall shape of the joint spectrum should change in predictable ways in response to various population-level processes . We fitted multiple demographic models to the 2D-SFS ( S6 Fig; S7 Fig ) describing population size ( constant size , bottlenecks , expansions ) and connectivity ( migration ) among subpopulations or demes representing spatially clustered localities ( see S6–S9 Tables and S1 Text for details on model selection , evaluation , and inference ) . The best fitting population history for YNP T . alpinus was characterized by relatively small but constant deme effective sizes through time ( ~1 , 350 individuals ) and an approximately threefold decrease in migration within the past ~90 years ( Fig 3 ) . This fitted demographic model fits well with our observation of increased population structure in YNP T . alpinus ( Fig 2 ) despite only minor reductions in nucleotide diversity per deme through time ( S2 Fig , S4 Table ) . In contrast , both SS T . alpinus and YNP T . speciosus were found to have much larger effective deme sizes ( ~4 , 600 and ~4 , 560 individuals respectively ) and higher migration rates overall . Consistent with the observed increase in FST , SS T . alpinus showed some evidence for a recent , very small decline in effective size ( Fig 3 ) . YNP T . speciosus was the only population that showed a clear signature of size change , albeit related to a historic ( pre 20th C ) population expansion ( Fig 3 ) . We found that modern samples tended overall to be less genetically similar to each other than did historic samples in all three comparisons ( Fig 2B; S8 Fig ) . Consistent with this observation , the fitted demographic histories for each species and transect support a recent ( < 90 years ago ) decrease in migration among demes ( Fig 3; see S1 Text , S6 Table ) . Decreased migration is expected if climate change is broadly affecting the amount of connectivity between suitable habitat patches available to species within this montane community [33] . However , as noted above , substantially increased genetic structuring was most evident in YNP T . alpinus ( Fig 2A ) . Long-term changes in gene flow should ultimately affect the genetic composition of metapopulations across these landscapes . It is possible that higher overall migration rates and larger effective population sizes have so far buffered the population genetic effects in T . speciosus and SS T . alpinus . However , genetic structure could accumulate over time according to the inferred histories . These results emphasize the utility of high-resolution demographic inference from genomic data not only for reconstructing population histories , but also as a potentially powerful conservation management tool [18 , 44 , 45] . An ABC framework is generalizable to other temporally sampled genetic datasets , allowing high-resolution inference into demographic histories over shallow evolutionary timescales that are relevant to recent anthropogenic climate change . In species of conservation concern , signatures of reduced migration could be used to motivate introductions between populations prior to significant genetic erosion , buffering against the future loss of genetic diversity and the accumulation of deleterious variation [46] . The benefits of such proactive efforts would have to be weighed carefully relative to the potential risks of introducing locally maladaptive variation [47] . Connections between broad ecological patterns and the genetic structure of populations are often intuitive and predictable . Upwards range contraction in T . alpinus is associated with reduced connectivity between suitable montane habitats [33] , which we infer has reduced migration between patches and increased genetic drift . However , a priori expectations for patterns of recent adaptive evolution are far less predictable in these species . Recent range shifts [20] and temporal changes in diet and skull morphologies [27 , 30] are both consistent with stronger directional selection gradients in YNP and SS T . alpinus relative to T . speciosus . On the other hand , lower effective population sizes and reduced migration ( at least in YNP ) should make selection relatively less effective in T . alpinus . Likewise , more effective adaptive responses could explain why larger and more connected populations of T . speciosus have remained stable in the face of common environmental stressors . To begin to tease these issues apart , we tested for specific genetic changes that might underlie recent adaptive responses in these species by directly comparing genetic differences between historic and modern populations . All three temporal population pairs are very closely related ( Fig 2 ) , however , they are also separated by changes in population structure and sizes ( Fig 3 ) that may confound standard signatures of positive selection [48] . Therefore , we tested for individual SNPs that had undergone large frequency shifts between historic and modern populations using an approach that is robust to the confounding influence of complex population histories on the genomic distribution of FST [49–51] . We found no significant allele frequency shifts over time in YNP T . speciosus or SS T . alpinus . In contrast , we identified five outlier SNPs in YNP T . alpinus populations ( false discovery rate [FDR] q-value < 0 . 01 ) relative to the inferred null distribution of per-site , genome-wide FST between the temporally sampled populations ( genome-wide temporal FST = 0 . 012; Fig 4A ) . To verify the inference of positive selection on these SNPs , we compared the observed FST values to null distributions simulated under the best ABC-fitted demographic history for YNP T . alpinus ( Fig 3 ) . Our simulated FST distributions were in close agreement with the overall observed FST values . Thus , it is very unlikely that demography alone could produce the extreme changes in allele frequencies that we observed at the outlier loci ( p-value < 3e-7; S9 Fig ) . Derived allele frequencies at all five differentiated SNP positions increased ~threefold in the modern populations ( average frequencies of 0 . 22 historic versus 0 . 65 modern; Fig 4B ) and all were located in the protein-coding gene , Arachidonate 15-Lipoxygenase ( Alox15 ) ( Fig 4D ) . Alox15 is a broadly expressed lipoxygenase involved in the resolution of acute inflammation through the generation of lipid-derived signaling molecules known as resolvins [52–54] . Alox15 expression has been associated with cardiovascular disease , oxidative stress , and response to hypoxia [55–57] as part of the Hypoxia-inducible factor-1α ( HIF-1α ) regulation pathway [58] . Two of the outliers represent synonymous changes in non-adjacent exons ( positions a , b; Fig 4D ) while the three other SNPs ( positions c-e ) were at non-coding positions within the same intron . All five positions were in strong linkage disequilibrium ( historic r2 = 0 . 86; modern r2 = 0 . 93 ) in YNP T . alpinus but invariant in all other populations except for one site ( b ) that was at similar frequency across the SS T . alpinus temporal contrast ( historic = 0 . 13 , modern = 0 . 2 ) . Given an ~500m contraction of the low elevation range limit in YNP T . alpinus over the last century [20] , the temporal allele frequency shifts at Alox15 could simply reflect non-sampling of extinct low elevation populations . To test this , we first estimated Alox15 allele frequencies as a function of elevation by pooling individuals into discrete 100-meter elevation bands . We observed the largest increases in derived allele frequencies at low to mid-elevation localities ( Fig 4C ) , and mean derived allele frequencies were not significantly correlated with elevation in either of the temporal samples ( historical R2 = 0 . 46 , p-value = 0 . 14; modern R2 = 0 . 34 , p-value = 0 . 42 ) . Furthermore , all five positions remained strong outliers in our temporal FST contrasts when we excluded low elevation sampling localities that were present only in the historic YNP T . alpinus transect ( OutFLANK FDR q < 0 . 05 , ABC-fitted FST distribution p-value = 4e-7 ) . Thus , evolutionary responses at Alox15 are consistent with in situ evolutionary change primarily among remnant demes below the upper bound of the modern YNP T . alpinus range ( <3200 meters elevation ) . In principle , the large shift in Alox15 allele frequencies observed between historic and modern samples could be driven by changes in habitats , food availability , or some other non-climate related environmental factor . However , based on previous modeling of changes in the elevational range [28] and the function of Alox15 , we suggest that physiological response to warming is the strongest current hypothesis . Winter temperature appears to be a primary limiting factor in the distribution of T . alpinus , with range contractions strongly tracking upslope shifts in minimum winter temperatures [28] . Increases in minimum winter temperatures at mid-elevations are resulting in reduced YNP snowpacks [59 , 60] , which Rubidge and colleagues suggested might reduce over-winter survival of T . alpinus through loss of critical thermal insulation of hibernacula [28] . Interestingly , arousal from hibernation has been shown to induce oxidative stress and hypoxia [61] and Alox15 shows strong seasonal induction in other species of hibernating squirrels [62] . A potential link between the intensity of selection on variation at Alox15 and changes in winter snowpack could also explain why we did not detect selection at the same gene in the SS transect . Tamias alpinus populations in the Southern Sierra are fixed for ancestral alleles at all but one of the outlier YNP SNPs , suggesting that these populations may lack genetic variation at Alox15 that is putatively adaptive in YNP . Moreover , SS T . alpinus populations are currently found above ~3200 meters—above the elevation range showing the largest allele frequency shifts in YNP—and overall snowpack has been more stable in the southern Sierra during the last century [59] . Though speculative , these scenarios help illustrate how evolutionary responses among populations may depend on both adaptive potential ( i . e . , standing genetic variation ) and local environmental conditions .
Temporal sampling of genomic data has the potential to provide powerful insights into the evolutionary effects of rapid environmental change [18] . Here we built on previous works [1 , 9 , 33] by generating targeted genome-wide sequence data from 294 chipmunks spanning a century of climate change . By integrating high throughput sequencing , cost and time-effective targeted enrichment technologies , and sophisticated inference methods , we provide powerful comparative insights into demographic and evolutionary responses of two montane species experiencing rapid environmental change . Our genomic time-series approach demonstrates one way that historical archives can be used to study biological responses during recent environmental change [9 , 11 , 18] . Temporal genomic data can provide a means to understand the current state of populations and their potential evolutionary trajectories , providing powerful tools to inform the conservation of populations experiencing changing environments . The identification of targets of positive selection during the recent upslope range contraction in T . alpinus points to a candidate gene and potential phenotypes associated with physiological stress that warrant further study . We caution that further evidence , such as differences in over-winter survival across genotypes or other functional studies , are necessary to demonstrate a causal relationship between Alox15 and response to climatic-induced stress . Further , our capture experiment only covered a subset of protein coding genes ( ~50% ) and did not include extensive coverage of regulatory regions that may often modulate rapid evolutionary responses [63] . That said , alpine chipmunks also show greater stress response to changes in external conditions [64] , a narrower range of activity patterns [31] , and more pronounced shifts in diet and functional aspects of cranial morphology when compared to T . speciosus over the past century [27 , 30] . Thus , the combination of phenotypic , behavioral , and now genetic evidence points to some component of physiological stress as a key factor in the greater sensitivity of T . alpinus to environmental change . Even in the absence of links to specific phenotypes or fitness , the identification of evolutionary responses at specific genes should help inform future on-ground studies focused on identifying the proximate causes of warming-related population declines across the range of this or other affected species [32] . Indeed , the potential for adaptive evolution to rescue populations in decline has emerged as an important concept in conservation biology [65] , with increasing efforts to directly incorporate evolutionary principles into conservation planning [66] . As a cautionary note , our results suggest that putatively adaptive responses in T . alpinus at Alox15 ( Fig 4 ) , as well as rapid shifts in functional morphology and diet [27 , 30] , have nonetheless been insufficient to prevent extensive extirpation of lower elevation populations of this alpine specialist . Comparative analyses of species range shifts over the past century have provided powerful insights into the ecological impacts of and biological responses to rapid environmental changes [1 , 19 , 21–23] . Here we have begun to extend these ideas to a comparative population genomic framework . Moving forward , we suggest that the true power of analyzing genomes and phenotypes of historical museum archives lies in the potential to extend across species [9 , 11] . Though the occurrence of museum records tend to be highly punctuated through space and time for a given species , historic collection efforts , such as those led by Joseph Grinnell and other early naturalists , usually surveyed many co-distributed species . With comparable contemporary sampling efforts , these invaluable archives will enable comparative community level insights into the impacts of and evolutionary responses to rapidly changing environments .
All animals sampled in the modern era were collected following procedures approved by the University of California , Berkeley Animal Care and Use Committee ( Permit number R278–0315 ) . Permits were provided by Yosemite National Park and Sequoia-Kings Canyon National Park . Tamias speciosus and T . alpinus surveyed in this study were collected from montane transects in Yosemite National Park ( YNP ) and the Southern Sierras ( SS ) . Historic samples were collected by Joseph Grinnell and his colleagues from 1911 to 1916 , and are preserved as dried skins in the Museum of Vertebrate Zoology ( MVZ ) , at the University of California , Berkeley . Modern samples were collected from the same sites by the ‘Grinnell Resurvey’ team led by MVZ researchers and collaborators from 2003 to 2012 ( Fig 1; S1 Table ) . We examined 100 YNP T . speciosus ( 52 historic , 48 modern ) , 104 YNP T . alpinus ( 56 historic , 48 modern ) , and 90 SS T . alpinus ( 52 historic , 38 modern ) from each transect . We also sampled six T . minimus ( the Least chipmunk ) collected east of YNP , which were used to test for potential hybridization between T . alpinus and T . minimus [38] . Furthermore , we included one sample each of three other species ( T . striatus , T . ruficaudus , and T . amoenus ) in order to polarize SNPs identified in our focal populations . Historic DNA was extracted from toe pad tissue ( ~3 x 3 mm ) in a separate dedicated laboratory using a previously described protocol [9] . DNA was extracted from modern samples using Qiagen DNeasy Blood and Tissue kits following the manufacturer’s protocol . Genomic libraries for all samples were constructed following Meyer and Kircher [67] with slight modifications [9] . We used RNA-seq [68] to sequence and assemble [69] transcriptomes for multiple tissues sampled from a single modern SS T . alpinus to serve as a reference for exome capture probe design . We targeted exonic regions ( 6 . 9 Mb , including flanking introns and intergenic regions ) corresponding to 8 , 053 T . alpinus genes targeted by our previous array-based capture experiments in chipmunks [9 , 70 , 71] . In addition , we extracted a broad set of candidate genes from the AmiGO and NCBI protein databases with functional annotations that were potentially relevant to environmental stress responses ( e . g . , HSP/HSF , hemoglobin , cytokines , apoptosis , immunity , oxidative stress , oxidative phosphorylation ) . We then used a BLASTx search against these genes to locate 2 , 054 orthologous transcripts ( 2 . 4 Mb ) from the Tamias transcriptome and included these transcripts in our capture . We also targeted the complete mitochondrial genome ( ~16 . 4 Kb ) to assess empirical error rates and five previously sequenced nuclear genes [42 , 72] to use as positive controls in post-capture qPCR assays of global enrichment efficiency . Probes were designed and manufactured by NimbleGen ( SeqCap EZ Developer kits ) . Barcoded genomic libraries were pooled together and hybridized in seven independent reactions with Tamias Cot-1 DNA and barcode-specific blocking oligonucleotides . Six hybridization experiments were used for the focal species ( one per time point for each of the three temporal contrasts ) and one additional capture was performed on pooled libraries from six T . minimus and three outgroup samples ( T . striatus , T . ruficaudus , and T . amoenus ) . After hybridization , each of the enriched genomic libraries were amplified using PCR and sequenced using one lane of Illumina HiSeq2000 per capture ( 100-bp paired-end ) . Bioinformatic processing of exon capture data followed our previous protocols [9 , 70] . All raw sequencing reads were treated to remove adapters , exact duplicates , low complexity ( i . e . , runs of ambiguous or mononucleotide sequence ) , and reads sourced from bacteria and human contamination . Overlapping paired reads were merged to avoid inflated estimates of coverage and biased genotype likelihoods . We used filtered sequencing reads ( 28 . 9 Gb ) from 48 modern YNP T . alpinus samples to generate de novo assemblies with ABySS [73] that were then merged using Blat [74] , CD-HIT [75] , and CAP3 [76] to remove redundancies . This total assembly was then compared to the original targets to construct a non-redundant target reference of 21 , 128 assembled contiguous sequences ( contigs ) totaling 20 . 8 Mb , and error-corrected following Bi and colleagues [9] . We then aligned cleaned reads from T . alpinus , T . speciosus , and T . minimus samples to the T . alpinus reference using Novoalign ( http://www . novocraft . com ) . Nucleotide positions were filtered at individual , contiguous sequence , and position levels of quality control following our previously described methods [9] ( S3 Table ) using the script snpCleaner ( https://github . com/tplinderoth/ngsQC/tree/master/snpCleaner ) . For each of the three temporal transects , we retained the intersection of filtered contigs between all historic and modern populations . As a result , 2 , 569 , 2 , 451 , and 2 , 738 contigs ( 11 . 6–13% of the total ) were eliminated from YNP T . speciosus , YNP T . alpinus , and SS T . alpinus datasets , respectively . At the site level , we removed sites showing unusually high or low coverage , excessive strand bias , end distance bias , base quality bias , and map quality bias . We also filtered out sites with extensive missing data among samples within each population . We were particularly attentive to errors associated with long-term DNA degradation . Postmortem nucleotide damage from hydrolytic deamination causes conversion from cytosine ( C ) to uracil ( U ) residues resulting in misincorporation of thymine ( T ) during PCR amplification [34 , 35 , 77] . We conservatively removed all C-to-T and G-to-A ( i . e . , the reverse complement of the C-to-T change with respect to the original PCR template molecules ) SNP positions from the datasets to avoid inaccurate population genetic inferences stemming from base misincorporation . In total , 9 . 0 , 9 . 3 , and 8 . 5 Mb of data from YNP T . speciosus , YNP T . alpinus , and SS T . alpinus passed all quality controls and were used in subsequent analyses . We used probabilistic methods for variant discovery and allele frequency estimation as implemented within ANGSD [78] . Using a population-specific SFS estimated from allele frequency likelihoods as a prior , we obtained allele frequency posterior probabilities and called SNPs using a 0 . 95 probability cutoff of being variable . The realSFS function was used to generate 1 , 000 bootstrap replicates of the folded site frequency spectrum ( SFS ) for each metapopulation by resampling per site allele frequency likelihoods . We then used ANGSD to estimate the number of segregating sites ( S ) , Watterson's theta ( θW ) , pairwise nucleotide diversity ( θπ ) , and Tajima's D in the historic and modern T . alpinus and T . speciosus metapopulations . For each metapopulation , we generated 100 estimates of θW and θπ using randomly chosen SFS bootstrap replicates as priors to evaluate sensitivity of these point estimates on the SFS prior . Additionally we estimated diversity statistics for demes within metapopulations , the former representing spatially clustered sampling localities . For each transect , population differentiation within and between the modern and historic populations was determined using probabilistic methods for estimating FST [79] and individual covariance matrices for principal component analysis ( PCA ) as implemented in ngsTools [80] . Confidence intervals ( 0 . 95 ) for global FST were generated from 1 , 000 bootstrap replicates of per-site FST values . To compare allele frequencies over time , we estimated the 2D-SFS between the pooled modern and pooled historic demes of each transect ( i . e . , three 2D-SFS comparisons ) . SNPs identified in T . speciosus and T . alpinus were polarized relative to T . striatus , T . ruficaudus , and T . amoenus . We further examined population genetic structure using NGSadmix [37] , which estimates admixture proportions from genotype likelihoods . We ran 10 replicates for K ( number of clusters ) ranging from 1–10 and summarized results ( S5 Table ) across runs to determine the best K [81] . To test for hybridization between T . alpinus and T . minimus samples , we used the program ADMIXTURE [82] to estimate individual ancestries using one randomly sampled SNP per contig . Next we developed an ABC framework for fitting binned 2D-SFS from serially sampled populations or metapopulations ( S4 Fig ) and used this approach to test hypotheses about the demographic histories of the sampled chipmunk populations . Additional details on demographic model construction , simulations , model selection , and inference are provided in S1 Text . Briefly , we fitted 5–9 explicit demographic models ( S6 Fig ) characterized by possible changes in migration and population size to each of the temporal contrasts . We performed 25 , 000 simulations per model , drawing parameter values from uniform or log-uniform prior distributions and then simulating ~20 . 2 Mb of sequence data for each individual under the specified history using the coalescent simulator fastsimcoal [83] . Lineages from the different demes were sampled at the present ( modern sample ) and 90 generations in the past ( historic sample ) according to the actual number of sampled individuals . Then all samples within a respective time period were pooled and the historic versus modern 2D-SFS was calculated . A custom script was used to calculate diagonal and off diagonal bins of the joint SFS ( S4 Fig ) , which served as our ABC summary statistic . We used the R package 'abc' [84] to calculate model posterior probabilities and to evaluate the reliability of our model selection procedure . We considered the best fitting models for each species/transect to be those with the highest posterior probabilities ( S7 Table ) and we used a cross validation procedure to determine error rates associated with model choice ( S8 Table ) . To aid model choice , we also considered the fit of the maximum likelihood ( ML ) estimate for each model to our observed data . We evaluated goodness-of-fit for the selected models by comparing the Euclidean distance between our observed and simulated 2D-SFS bins ( S9 Table ) . We considered SNPs with large allele frequency shifts between the modern and historic time periods that could not be attributed to demography as evidence for positive selection . We used the program OutFLANK [50] to detect FST outlier SNPs ( FDR q-value < 0 . 01 ) , empirically adjusting the degrees of freedom of χ2-distributed FST values to account for the influence of demography . We then compared the observed SNP FST values to null exome-wide and per-site FST distributions generated by performing 1 , 500 neutral simulations under the best fitting population history for YNP T . alpinus . | Museum specimens represent an irreplaceable archive that can be used to understand how species respond to rapid environmental change . We recovered genomic data from archived samples spanning a century of climate change in co-distributed declining versus stable species of montane chipmunks . Applying novel statistical methods , we find evidence for strong positive selection on a physiologically relevant gene despite increased population fragmentation in the declining species . Our results reveal rapid evolutionary responses , but also highlight that genetic adaptation has been insufficient to prevent range collapse in this endemic alpine species . These findings illustrate how biological archives can be used to pinpoint genetic responses through time to better understand how species are responding to rapidly changing environments . | [
"Abstract",
"Introduction",
"Results",
"and",
"discussion",
"Conclusions",
"Methods"
] | [
"genome",
"evolution",
"population",
"genetics",
"vertebrates",
"social",
"sciences",
"animals",
"mammals",
"animal",
"behavior",
"evolutionary",
"adaptation",
"population",
"biology",
"zoology",
"genomics",
"genetic",
"polymorphism",
"behavior",
"molecular",
"evolution",
"chipmunks",
"comparative",
"genomics",
"evolutionary",
"genetics",
"rodents",
"psychology",
"eukaryota",
"animal",
"migration",
"genetics",
"biology",
"and",
"life",
"sciences",
"computational",
"biology",
"evolutionary",
"biology",
"amniotes",
"evolutionary",
"processes",
"organisms"
] | 2019 | Temporal genomic contrasts reveal rapid evolutionary responses in an alpine mammal during recent climate change |
Among the different faces of immune reconstitution inflammatory syndrome ( IRIS ) developing in HIV-patients , no clinical definition has been reported for Schistosomiasis-IRIS ( Schisto-IRIS ) . Although Schisto-IRIS remains largely uninvestigated , matrix metalloproteinases ( MMP ) and tissue inhibitors of metalloproteinases ( TIMP ) have previously been associated with S . mansoni infection and tuberculosis-IRIS . Here , we aimed to investigate the relevance of these markers in Schisto-IRIS . Patients were diagnosed with IRIS related to S . mansoni within a cohort of patients with Schistosomiasis-HIV co-infection , using a clinical working definition of Schisto-IRIS . We compared 9 patients who developed Schisto-IRIS to 9 Schisto+HIV+ controls who did not , and 9 Schisto-HIV+ controls . Plasma levels of MMP-1 , MMP-7 , MMP-10 , TIMP-1 , TIMP-2 , sCD14 , intestinal fatty-acid binding protein , C-reactive protein , and 8 anti-nuclear antibodies ( ANA ) were analyzed prior to and during 3 months of ART . Although no differences were observed for MMP-1 and -7 , MMP-10 levels decreased significantly in Schisto+HIV+ controls during 3 months of ART ( p = 0 . 005 ) while persisting in Schisto-IRIS patients at significantly higher levels compared to Schisto-HIV+ controls ( p≤0 . 030 ) . In contrast TIMP-1 levels only decreased significantly in Schisto-IRIS patients ( p = 0 . 012 ) , while TIMP-2 levels were lower compared to Schisto+HIV+ controls at 2 weeks ( p = 0 . 007 ) , 1 month ( p = 0 . 005 ) and 3 months ( p = 0 . 031 ) of ART . Five out of 8 ANAs studied decreased significantly in Schisto-IRIS patients after 1 month of ART ( p≤0 . 039 ) , whereas only 1 ANA decreased for Schisto+HIV+ controls ( p = 0 . 027 ) . In this study , we propose a working definition for the diagnosis of Schisto-IRIS in resource limited settings . We report persistent plasma levels of MMP-10 , along with a more pronounced decrease in TIMP-1 and ANA-levels , and low levels of TIMP-2 during 3 months of ART . Corresponding to the clinical symptoms , these data suggest that Schisto-IRIS is marked by unbalanced MMP/TIMP dynamics which favor inflammation .
HIV-patients initiating antiretroviral therapy ( ART ) while dealing with a co-infection are at risk of developing immune reconstitution inflammatory syndrome ( IRIS ) . IRIS is described as a clinical deterioration of HIV patients in the first weeks or months after starting ART , often marked by tissue-destructive inflammation [1–3] . At least three conditions need to be present for an HIV patient to be at risk of IRIS; severe immune suppression , a treated ( paradoxical IRIS ) -or- undiagnosed ( unmasking IRIS ) opportunistic infection , and initiation of ART as the trigger . Despite these shared features , IRIS embodies a heterogeneous collection of clinical manifestations [1 , 2] , associated with a plethora of pathogens [3 , 4] . Among these pathogens , IRIS associated with Schistosoma mansoni ( Schisto-IRIS ) has received only limited attention in research , with few cases reported prior to 2010 [5–7] . Nonetheless , HIV and Schistosomiasis are highly co-endemic in Sub-Saharan Africa [8] . This is especially true in regions with frequent human-water interaction [9 , 10] , e . g . fishing villages along the shores of Lake Victoria . Areas such as these could form focal-points of IRIS development , as demonstrated previously in a Kenyan cohort with ~36% of Schistosomiasis-HIV patients on ART developing worsening schistosomiasis symptoms consistent with IRIS [11] . Thus , while perhaps not as prevalent as IRIS associated with tuberculosis ( TB-IRIS ) [3 , 12] , the clinical burden of Schisto-IRIS in the field should not be underestimated . Little is known on the immune dysregulation in patients who develop Schisto-IRIS , nor has the relevance of findings in TB-IRIS for Schisto-IRIS been studied before . Most findings in TB-IRIS can be directly linked to an acute inflammatory response to TB-antigens , coinciding with monocyte activation and a cytokine storm [13–15] . Elevated levels of matrix metalloproteinases ( MMPs ) have also been reported , with MMP-1 , -3 , -7 , -8 , and -10 being of particular interest [16–18] . MMPs hydrolyze various components of the extracellular matrix , allowing classification as collagenases ( e . g . MMP-1 ) , gelatinases ( e . g . MMP-2 ) , matrilysins ( e . g . MMP-7 ) or stromelysins ( e . g . MMP-10 ) [16] . Along with tissue inhibitors of metalloproteinases ( TIMPs ) , MMPs are functionally involved in inflammation , granuloma formation and tissue remodeling , thus illustrating their involvement in TB lung pathology and symptoms seen in TB-IRIS [17 , 18] . Cells of the innate immune system have the capacity to produce MMPs and TIMPs [19 , 20] , further suggesting a major contribution of the innate immune system to TB-IRIS development . In schistosomiasis , MMP-1 and -2 and TIMP-1 and -2 have been linked to active periovular granulomas in humans [21] , whereas increased levels of MMP-8 and -10 , and TIMP-1 and -2 were observed in mice [22] . Conversely , mice treated with praziquantel ( PZQ ) showed a decrease in MMP and TIMP levels , although MMP-10 levels increased consistent with ongoing resorption of fibrous tissue [23] . PZQ treatment has been reported to influence schistosome-specific immune responses [24 , 25] . Indeed , a pro-inflammatory shift in schistosome-specific cytokine production has been observed in patients treated with PZQ , which may contribute to resistance to re-infection [26] . Moreover , one study previously described an inverse relationship between S . haematobium infection intensity and anti-nuclear antibody ( ANA ) levels , which rise upon PZQ treatment [27] . Together , these findings highlight protective , but potentially pathogenic immune responses to schistosome antigens following PZQ treatment . Drawing parallels with TB-IRIS , we hypothesized that Schisto-IRIS is characterized by an over-representation of pro-inflammatory factors after PZQ treatment that might otherwise improve resistance to re-infection . Schisto-IRIS could thus be associated with; increased plasma levels of MMPs or ANAs; decreased plasma levels of TIMPs; and increased monocyte activation ( measured by soluble CD14 ) . In addition , we previously reported lower levels of intestinal fatty-acid binding protein ( I-FABP ) in TB-IRIS patients [13] . Since I-FABP is used as a marker for damage to the intestinal epithelium , we further hypothesized that persisting levels of I-FABP could be associated with intestinal symptoms associated with Schisto-IRIS . Using samples collected from Schistosoma mansoni-HIV co-infected fishermen starting ART in Kenya [11] , we conducted a nested 3-month case-control analysis of plasma MMPs , TIMPs , ANAs , sCD14 , and I-FABP among those who developed Schisto-IRIS .
Patients from a prospective case-control study at the fishing community in Uyoma , Rarieda District , Kenya , were studied as described previously [10 , 11] . The study focused on a group of permanent residents that are occupationally-exposed to water infested with the infective stage of the Schistosoma mansoni parasite ( S1 Fig ) . All participating individuals were screened for schistosomiasis and underwent voluntary counseling and testing ( VCT ) for HIV . All HIV-patients were given lamivudine , stavudine and nevirapine based combination ART shortly after screening , according to Kenya national guidelines at the time . When eggs were identified in stool samples , co-infected patients were also given a single dose of 40mg/kg PZQ treatment according to standard local clinical practice . Seventy-one ART naïve HIV-patients with a history of treated schistosomiasis infection ( Schisto+HIV+ ) were included in the study , of whom 26 developed Schisto-IRIS . In addition , a group of ART naïve HIV-patients without a history of schistosomiasis infection ( Schisto-HIV+ ) were recruited as controls ( Fig 1 ) . Patients were followed up for 3 months and blood-plasma was collected prior to initiating ART ( baseline ) , at 2 weeks , at 1 and 3 months after starting ART . ART adherence and efficacy were monitored by oral questioning , CD4 counts and viral load ( VL ) measurements at each time point ( Fig 2 ) . In the current study , only patients with available clinical data were included , who had samples of sufficient quality available at all 4 time points to allow longitudinal analysis . Diagnosis of schistosomiasis was performed at baseline and all subsequent time points by Kato Katz thick stool smear as described previously [28] . At each timepoint , 1 stool sample was taken , and 2 smears were analyzed by experienced lab technicians . S . Haematobium infection was excluded using a parallel urine filtration method [29] . Patients were further excluded when presenting with eggs belonging to other helminth infections; Ascaris lumbricoides , Trichuris trichuria and hookworm . Other exclusion criteria included; being <18 years of age and having any of the most common co-infections such as malaria , tuberculosis or hepatitis B , respectively detected by microscopy , skin test or serology . For the purpose of the current study , a working case definition of Schisto-IRIS was refined to match consensus definitions used for the diagnosis of other forms of IRIS ( Table 1 ) [2 , 30] . Antecedent requirements for paradoxical Schisto-IRIS were; diagnosis of S . mansoni , followed by successful PZQ treatment in close proximity to ART initiation ( and eggs were no longer detectable during the next visit ) [7] . Based on a clinical Schisto-IRIS description used previously [11] , symptoms were re-evaluated according to frequency and relevance , and classified as “major” or “minor” symptoms . Schisto+HIV+ patients were diagnosed with paradoxical Schisto-IRIS when presenting with at least 2 major , or 1 major and 3 minor re-emerging symptoms of otherwise successfully treated schistosomiasis during ART . Major symptoms include; new or worsening bloody diarrhea , new or worsening hematuria , and new or worsening hepatomegaly ( HMG ) , splenomegaly ( SMG ) , or portal vein enlargement ( PVE ) within 3 months of starting ART . Minor symptoms include; constitutional symptoms ( fever , night sweats ) , focal inflammation ( skin infection , skin rash , swollen glands , mouth sores ) , algetic symptoms ( chest pain , joint pain ) , watery diarrhea , and new eggs in stool ( whose presence could not be explained by other possible causes ) . Finally , other possible explanations , such as re-infection , other co-infections and treatment failure were excluded . In accordance with previous IRIS studies , we propose to use the term “ART-associated schistosomiasis” to refer to PZQ naïve individuals who start de novo production of schistosome eggs during ART . Diagnosis of unmasking Schisto-IRIS in patients with ART-associated schistosomiasis required; heightened intensity of clinical manifestations and a clinical course consistent with paradoxical Schisto-IRIS once PZQ treatment was initiated ( Table 2 ) . Venous blood was drawn into EDTA tubes and plasma was separated from cells by centrifugation and stored at -80°C . Plasma concentrations of 3 MMPs and 2 TIMPs were determined in duplicate using Bio-Plex human MMP and TIMP assay kits ( Bio-Rad Laboratories NV-SA , Nazareth , Belgium ) according to the manufacturer’s instructions . We thus measured plasma levels of MMP-1 , -7 , -10 and TIMP-1 & -2 . The Bio-Plex array reader and Manager 5 . 0 software were used to analyse concentrations using a weighted five-parameter logistic curve-fitting method . In addition , plasma concentrations of sCD14 and intestinal fatty-acid-binding protein ( I-FABP ) were determined by ELISA ( HyCult biotechnology BV , Uden , The Netherlands ) , and C-reactive protein ( CRP ) was determined using VITROS Chemistry Products CRP Slides ( Ortho-Clinical Diagnostics , NY , USA ) . In addition , a semi-quantitive determination of 8 ANAs ( directed against; Smith antigen ( Sm ) , U1 small nuclear ribonucleoprotein ( U1 snRNP ) , snRNP/Sm complex , Sjögren’s-syndrome-related antigen A & B ( SS-A & SS-B ) , topoisomerase I ( Scl-70 ) , centromere protein B ( CenpB ) and Jo-1 ) was performed by ELISA using index values , calculated by ratio of sample to cut-off calibrator ( AESKU . DIAGNOSTICS GmbH & Co . , Wendelsheim , Germany ) . The study was approved by the Scientific Steering Committee ( SSC number 1763 ) and Ethical Review Committee ( ERC ) at the Kenya Medical Research Institute ( KEMRI ) and written informed consent was obtained from all study participants . The use of plasma samples in the current study was approved by the institutional review board of the Institute of Tropical Medicine of Antwerp . Statistics were performed using SPSS software ( version 17 . 0 ) or GraphPad Prism ( version 7 ) with significance level set at p < 0 . 05 . Differences between patient groups were analyzed using Mann-Whitney U tests , or Pearson Chi-square tests for dichotomous values . The significant change over time of variables for each patient group was calculated using the Friedman test ( p-values shown in graphs ) . When Friedman tests showed global significance , Dunn’s multiple comparison post-hoc tests and multiplicity adjusted p-values were used to indicate differences between specific time points , ( indicated in graphs by horizontal bars with an asterisk ) . Significant differences between 2 individual time points ( baseline vs . month 1 for ANAs ) within groups were determined using a Wilcoxon signed-rank test . Correlations were performed using Spearman's rank-order correlation . Because of the hypothesis driven nature of this study , no other correction for multiple testing was applied [31 , 32] .
A subset of plasma samples collected within a prospective case-control study at the fishing community in Uyoma , Kenya [11] were selected for further analysis . Plasma samples from a total of 9 Schisto-IRIS patients , 9 Schisto+HIV+ and 9 Schisto-HIV+ patients were thus analyzed . All 3 groups did not differ significantly in age , gender or baseline viral load ( Table 3 ) . No significant differences could be observed between any of the groups for CD4 counts across all 4 time points , except for Schisto-HIV+ patients who had lower counts compared to Schisto-IRIS patients after 3 months on ART ( p = 0 . 047 ) . Time analysis showed a significant increase in CD4 counts for Schisto-IRIS patients ( p = 0 . 008 ) and Schisto+HIV+ patients ( p = 0 . 007 ) during 3 months of ART , whereas Schisto-HIV+ participants did not have a significant increase in CD4 counts ( p = 0 . 214 ) ( Fig 3 ) . Dunn’s post-hoc test subsequently highlighted a significant increase from baseline to month 3 in Schisto-IRIS patients ( p = 0 . 004 ) and Schisto+HIV+ patients ( p = 0 . 003 ) . Schisto-IRIS patients did not differ from Schisto+HIV+ patients for baseline egg-count [eggs per gram ( EGP ) ] and EPG were reduced in both groups after PZQ treatment . Due to patients being referred to government facilities for ART , treatment intervals between PZQ and ART varied from patient to patient ( Fig 4 ) . Overall , the treatment interval between PZQ and ART did not differ between selected Schisto-IRIS and Schisto+HIV+ patients ( p = 0 . 287 ) . All patients received PZQ before the standard planned visit at week 2 , except for 2 Schisto+HIV+ patients ( who received PZQ 35 and 81 days after ART initiation ) . All co-infected patients reported in this study experienced clinical signs consistent with S . mansoni infection prior to starting PZQ and ART including; hepato-/splenomegaly , bloody/watery diarrhea , abdominal pains , etc . ( S1 Table ) . Patients were retrospectively classified as Schisto-IRIS or Schisto+HIV+ patients , according to symptoms presented during ART . Both groups showed similar distribution of abdominal symptoms at baseline , while neither group experienced PVE at this time . SMG at baseline was diagnosed in 6/9 ( 67% ) Schisto-IRIS patients but only in 1/9 ( 11% ) Schisto+HIV+ patients , though 2/9 ( 22% ) Schisto+HIV+ patients experienced HMG . Following PZQ/ART treatment , EPG declined in all patients . Symptoms subsided within 3 months of ART in patients who were not diagnosed with IRIS . In contrast , 6/9 ( 67% ) Schisto-IRIS patients developed new or worsening bloody diarrhea , 3/9 ( 33% ) developed new HMG or SMG and 5/9 ( 56% ) developed new PVE ( Table 4 ) . All Schisto-IRIS patients developed at least 3 minor symptoms during ART , except for one who developed 2 ( in addition to 2 major symptoms ) . This included 7 patients with paradoxical IRIS and 2 patients with unmasking IRIS , whom did not differ in baseline characteristics or CD4 count . Unmasking Schisto-IRIS patients showed zero EPG prior to ART , but experienced de novo egg production at 2 weeks after starting ART . These patients initiated PZQ treatment only after egg production was diagnosed ( at 13 and 15 days post ART ) . Since patients were followed up at pre-determined visits , IRIS was diagnosed at the closest visit . The median time to IRIS diagnosis was 29 days ( IQR 14–91 ) for paradoxical Schisto-IRIS , and 126 days ( IQR 119–132 ) for unmasking Schisto-IRIS ( p = 0 . 053 ) . We observed a significant correlation between treatment interval and time to IRIS diagnosis ( R = 0 . 759; p = 0 . 021 ) ( Fig 4 ) . Increased plasma levels of MMPs have previously been associated with Schistosomiasis , TB infection and TB-IRIS [22 , 23 , 33–35] . In order to investigate the role of MMPs in Schisto-IRIS , we evaluated 3 different MMPs classified as either collagenase ( MMP-1 ) , matrilysin ( MMP-7 ) or stromelysin ( MMP-10 ) [16] . We thus evaluated plasma MMP levels in 3 patient groups at baseline and after 2 weeks , 1 month and 3 months on ART ( Table 5 ) . We could not observe significantly different MMP-1 , MMP-7 or MMP-10 levels in Schisto-IRIS patients when directly compared to Schisto+HIV+ patients . However , Schisto-IRIS patients showed significantly higher MMP-10 levels during ART compared to Schisto-HIV+ patients ( p ≤ 0 . 030 ) . Conversely , MMP-10 levels were similar between Schisto+HIV+ and Schisto-HIV+ patients during ART , except for month 1 ( p = 0 . 041 ) . Subsequent analysis over time revealed a significant overall change in plasma MMP-10 levels for Schisto+HIV+ patients ( p = 0 . 005 ) , with Dunn’s post-hoc test highlighting a significant decrease from baseline to month 3 specifically ( p = 0 . 006 ) . In contrast , Schisto-IRIS and Schisto-HIV+ patients showed no significant change in MMP-10 levels ( Fig 5A–5C ) . We next investigated plasma levels of TIMP-1 and TIMP-2 to evaluate their role in Schisto-IRIS ( Table 5 ) . TIMP-1 levels did not differ significantly between the 3 patient groups at any time point . Although TIMP-1 levels declined over time in each group , this change over time only reached significance in Schisto-IRIS patients ( p = 0 . 012 , Fig 5D and 5E ) . Dunn’s post-hoc test subsequently highlighted a significant decline from baseline to month 3 in these patients ( p = 0 . 012 ) . Conversely , Schisto-IRIS patients had markedly lower TIMP-2 levels across the board compared to Schisto+HIV+ patients , reaching significance at week 2 ( p = 0 . 007 ) , month 1 ( p = 0 . 005 ) and month 3 ( p = 0 . 031 ) . No differences over time could be observed in TIMP2 levels within the Schisto+HIV+ group . In order to evaluate the balance between MMPs and TIMPs , we next analyzed ratios of MMP-1 , MMP-7 and MMP-10 to TIMP-1 and TIMP-2 in patients and controls ( S2 Fig ) . The ratio of MMP-10/TIMP-2 decreased over time in Schisto+HIV+ patients ( p = 0 . 012 ) , whereas the ratio of TIMP-1/TIMP-2 decreased in Schisto-IRIS patients only ( p = 0 . 028 ) . I-FABP is released into the bloodstream upon damage to the intestinal epithelium , sCD14 is shed upon monocyte activation , and CRP is a well-known acute phase protein . In theory , these markers could therefore be used for monitoring tissue damage and inflammation in Schistosomiasis and/or Schisto-IRIS . We thus evaluated plasma levels of these markers in all 3 patient groups ( Table 6 & Fig 6 ) . However , all patients showed comparable I-FABP & sCD14 levels at every time point . No differences were observed for CRP levels between Schisto-IRIS and Schisto+HIV+ patients either . Compared to Schisto-HIV+ patients , Schisto-IRIS and Schisto+HIV+ patients showed higher CRP levels after 1 month ( p = 0 . 021 ) and 2 weeks ( p = 0 . 012 ) of ART respectively . Only Schisto+HIV+ patients showed significant overall variation in CRP over time ( p = 0 . 028 ) , while Dunn’s post-hoc test showed no significance . Anti-nuclear antibodies are associated with auto-immune diseases and are reported to rise in S . haematobium patients following PZQ treatment [27] . We thus performed a semi-quantitative analysis of 8 ANAs in Schisto-IRIS and Schisto+HIV+ patients at baseline and after 1 month of ART ( Fig 7 ) . No significant differences could be observed between patient groups at either time point . Nonetheless , Schisto-IRIS showed an overall decrease in plasma levels of 5 ANAs after 1 month of ART; U1-RNP ( p = 0 . 012 ) , SnRNP/Sm ( p = 0 . 027 ) , Sm ( p = 0 . 020 ) , Scl-70 ( p = 0 . 039 ) , SS-A ( p = 0 . 020 ) , SS-B ( p = 0 . 074 ) , CenpB ( p = 0 . 570 ) , Jo-1 ( p = 0 . 313 ) , whereas Schisto+HIV+ patients only showed decreased levels of Scl-70 ( p = 0 . 027 ) . Comparison of change over time ( delta-values calculated by subtracting baseline from month 1 ) showed a significantly stronger decline of SnRNP/Sm ( p = 0 . 046 ) and SS-A ( p = 0 . 008 ) in Schisto-IRIS patients compared to Schisto+HIV+ patients . Since the number of CD4+ T cells directly influence immunological processes , we next correlated CD4 counts to our observations . Schisto+HIV+ patients showed a significant negative correlation between CD4 counts and MMP-1 , pre-ART ( R = -0 , 683; p = 0 . 042 ) and at week 2 ( R = -0 , 717; p = 0 . 030 ) , whereas Schisto-IRIS and Schisto-HIV+ patients did not . Next , we evaluated whether excretion of S . mansoni eggs could be associated with damage to the intestinal epithelium . However , I-FABP levels showed no correlation with EPG at any time point in any group . To evaluate whether MMP-10 levels could be maintained by inflammatory factors , we then performed a correlation with CRP levels . However , no correlations were observed between CRP and MMP-10 levels at any time point for any group ( S3 Fig ) . We next correlated CRP levels with ANA levels to evaluate a potential link between ANA-levels and systemic inflammation ( S4 Fig and S5 Fig ) . Schisto-IRIS patients showed moderate to strong correlations at baseline between CRP levels and levels of U1-RNP , SS-A , Jo-1 , and SS-B ( R ≥ 0 . 669; p ≤ 0 . 043 ) . Except for U1-RNP , these correlations were preserved at month 1 ( R ≥ 0 . 695 , p ≤ 0 . 043 ) . Schisto+HIV+ patients showed a similar pattern of correlations at baseline , although CenpB correlated to CRP instead of SS-B ( R = 0 . 783 , p = 0 . 017 ) . At month 1 , CRP correlated with SS-A , SS-B , Sm , and CenpB ( R ≥ 0 . 722 , p ≤ 0 . 031 ) in these patients . Nonetheless , CPR and ANA levels did not show a similar change over time , as no correlation could be observed between delta values ( baseline subtracted from month 1 ) of CRP and any of the ANAs determined . Finally , we evaluated a potential link between the decreasing TIMP-1 and ANA levels in Schisto-IRIS patients ( S6 Fig and S7 Fig ) . However , only Schisto+HIV+ patients showed correlations between TIMP-1 and 6 of the 8 ANAs at both time points ( R ≥ 0 . 700; p ≤ 0 . 043 ) .
Among the different faces of IRIS developing in HIV-patients , Schistosomiasis-associated IRIS remains largely uninvestigated despite both infections being highly co-endemic [8] . Unlike clinical definitions used for TB-IRIS [2] and Cryptococcal IRIS [30] , no consensus definition for Schisto-IRIS exists . Within a previously described cohort of HIV-patients with Schistosomiasis co-infection , 36 . 6% developed Schisto-IRIS [11] . Here , we report clinical characteristics of patients who developed Schisto-IRIS related to S . mansoni and propose a working definition for the diagnosis of paradoxical and unmasking Schisto-IRIS in resource limited settings . In addition , we explored the immunopathogenesis of Schisto-IRIS by measuring plasma levels of 3 MMPs , 2 TIMPs , sCD14 , and 8 anti-nuclear antibodies , which were previously suggested to have a role in TB-IRIS and/or schistosomiasis [22 , 23 , 27 , 33–37] . We hypothesized that Schisto-IRIS could be associated with increased plasma levels of MMPs , sCD14 , and ANAs; or decreased plasma levels of TIMPs . To that end , we compared plasma levels of these markers during 3 months of ART between Schisto-IRIS patients and Schisto+HIV+ , as well as Schisto-HIV+ controls . In addition , we explored I-FABP levels as a marker of intestinal damage . We report a significant decline in MMP-10 levels following ART initiation in Schisto+HIV+ controls , but not in Schisto-IRIS patients . Conversely , plasma levels of TIMP-1 decreased in Schisto-IRIS patients , and TIMP-2 levels were significantly lower shortly after starting ART . In line with our hypothesis , these findings suggest that Schisto-IRIS patients in our cohort experience a MMP/TIMP profile that favors inflammation and tissue damage . Contrary to our hypothesis , plasma levels of ANAs decreased upon ART in Schisto-IRIS patients , possibly reflecting the release of S . mansoni antigens upon PZQ treatment [27] . The functions ascribed to MMPs range from tissue remodeling and angiogenesis to regulation of immune responses and inflammation [38] . In Schistosomiasis , MMPs regulate the granulomatous response to schistosome-eggs [22] . Upon praziquantel treatment , MMP levels decline in parallel with diminishing inflammatory responses [23] . In line with this , Schistosomiasis-HIV patients in our cohort who did not develop Schisto-IRIS showed a significant decline in MMP-10 levels during ART . Schisto-IRIS patients , however , experienced persistent levels of MMP-10 throughout follow-up . Although we observed no significant differences in MMP-1 , -7 or -10 in direct comparison between these 2 groups , Schisto-IRIS patients retained significantly higher MMP-10 levels compared to Schisto-HIV+ controls , whereas Schisto+HIV+ controls did not . Overall , these findings indicate that Schisto-IRIS patients do not readily normalize MMP-10 levels within the first months of ART . The spread in data likely masked this effect in direct comparison between groups . Interestingly , MMP-10 gene expression has previously been observed to be paradoxically elevated in praziquantel-treated mice , matching declining collagen gene expression [23] . The persistent levels of MMP-10 in Schisto-IRIS patients observed here could therefore indicate an ongoing immune response to S . mansoni antigens [39] , or a continuous resorption of fibrous tissue surrounding schistosome-eggs [23] . Alternatively , MMP-10 levels could have been maintained by inflammatory factors such as CRP , interferon-gamma , interleukin-6 , and/or monocyte activation [40 , 41] , which have previously been associated with TB-IRIS [13 , 42] . However , we could not observe significant correlations between CRP levels and MMP-10 in our current study . Moreover , no differences were observed for sCD14 , which is consistent with our previous observations on TB-IRIS [13] . Counterbalancing the effects of MMPs , TIMPs are natural inhibitors of MMP activity [43] . The balance between MMPs and TIMPs thus influences the level of tissue degradation and inflammation [38] . As expected after praziquantel treatment [23] , TIMP-1 levels declined during ART in both Schistosoma-infected groups . However , this decline was much more pronounced in Schisto-IRIS patients . Conversely , TIMP-2 levels remained stable after ART initiation in all groups , but were significantly lower in Schisto-IRIS patients compared to Schisto+HIV+ patients . These findings suggest that Schisto-IRIS patients have a lowered capacity to compensate for the inflammatory effects of MMP-10 which seem to persist during ART . Interestingly , Schisto-HIV+ controls showed similarly low levels of TIMP-2 during ART as Schisto-IRIS patients . However , MMP-10 levels were lower still , leading to significantly decreased MMP-10/TIMP-2 ratios at week 2 compared to Schisto-IRIS patients . Although the difference in MMP-10/TIMP-2 ratios between Schisto-IRIS and Schisto+HIV+ patients did not reach significance , only Schisto+HIV+ patients showed a significant decrease of this ratio during ART . Overall , these findings suggest diverging TIMP & MMP dynamics in Schistosomiasis-HIV patients who develop IRIS , favoring inflammation and/or tissue damage which becomes clinically apparent with the onset of IRIS symptoms . All co-infected patients reported in this study experienced clinical signs consistent with S . mansoni infection prior to starting PZQ and ART [44 , 45] . Following effective PZQ/ART treatment , most symptoms gradually subsided within 3 months of ART in patients who did not develop IRIS , in line with decreasing levels of MMP-10 . Using the working definition proposed in this study , Schisto-IRIS patients could be distinguished from Schisto+HIV+ controls , as the clinical condition worsened during ART . Compared to baseline , all 9 Schisto-IRIS patients in our study developed new or worsening symptoms of schistosomiasis during this period , despite successful PZQ treatment . In line with persistent MMP-10 levels , a steeper decline in TIMP-1 and consistently low TIMP-2 levels suggest that these patients experience a more vigorous and extended period of tissue damage and/or inflammation . Indeed , the majority of Schisto-IRIS patients developed bloody diarrhea ( >50% ) , new PVE ( >50% ) , and/or new HMG/SMG ( >30% ) during ART , often accompanied by minor symptoms such as fever , etc . Nonetheless , this did not coincide with altered damage to the intestinal epithelium , as we observed no differences in I-FABP . Overall , our observations on MMP/TIMP dynamics during ART thus correspond to the clinical spectrum of Schisto-IRIS patients as defined here , and may drive or be driven by the ‘major’ symptoms which occurred more rarely in Schisto+HIV+ controls [46–48] . Nonetheless , the distribution of symptoms among IRIS patients during ART remains somewhat heterogeneous , as is to be expected in IRIS [2 , 30] . Moreover , the interactions of MMPs and TIMPs are very complex . Consequently , it is difficult to predict the specific roles these factors may have , and what other factors ( e . g . cytokines ) may be involved . The immune responses and cytokine profiles of Schisto-IRIS patients should therefore be further explored . Despite the heterogeneity that surrounds IRIS , a high pre-ART antigen burden is commonly recognized as a risk factor , as is a short interval between treatment for the opportunistic infection and ART [49 , 50] . Although the PZQ-ART interval was similar between our patients and controls , we observed a correlation between PZQ-ART interval and onset of IRIS symptoms . Patients who received PZQ treatment sooner relative to ART ( i . e . before ART ) showed earlier onset of IRIS symptoms than patients who received PZQ later ( i . e . during the first 2 weeks of ART ) , suggesting a time-dependent association between PZQ treatment , ART initiation , and Schisto-IRIS . Interestingly , studies in HIV-negative patients have demonstrated a rise in pro-inflammatory responses to S . haematobium [26 , 51] and S . mansoni [52 , 53] antigens within the first months following PZQ treatment . These responses may reflect either the removal of worm-induced immunosuppression or the release of adult worm antigens as a result of PZQ treatment [51] . Although our study did not directly determine antigen levels , we determined relative plasma levels of 8 anti-nuclear antibodies , which have previously been reported to correlate inversely with intensity of S . haematobium infection and rise during PZQ treatment [27] . In contrast to Schisto+HIV+ controls , our Schisto-IRIS patients showed a significant decrease in plasma levels of several ANAs after 1 month of ART . Given the inverse relationship of ANA-levels with infection intensity , one could argue that this decrease reflects a sudden release of antigens . Alternatively , a stronger S . mansoni specific immune response to these antigens could be present which indirectly downregulates ANA levels . However , this decrease was not mirrored in CRP levels , since no correlation could be observed in change over time ( delta value ) with ANAs . Considering the time-dependent association with PZQ treatment , it is thus plausible that Schisto-IRIS manifests itself as an aggravated reaction to antigens released by worms that were killed as a result of PZQ treatment . Still , additional studies are needed to fully explore the relationship between ANAs and S . mansoni antigen loads in Schisto-IRIS . Although this work is nested in one of the first Schisto-IRIS specific studies to date , our strict selection of patients with complete samples and follow-up data lead to a relatively small population . As we did not include a Schisto+HIV- population , our study also cannot provide additional insights in the effects of PZQ without ART . Since this study was focused on S . mansoni co-infection , the occurrence of IRIS with other species ( e . g . S . haematobium ) has also not been investigated . The results reported here should thus inspire larger studies to fully explore MMP profiles in Schisto-IRIS , associated with different Schistosoma species . Our study accounted for re-infection with S . mansoni by monitoring patients at intervals shorter than 6 weeks . Although exposure between study visits cannot be fully accounted for , all Schisto-IRIS patients developed symptoms before any de novo eggs were documented in stool . We identified onset of IRIS symptoms at the nearest pre-planned visit , with 5 patients showing onset of IRIS symptoms earlier than 45 days on ART , and 4 later than 89 days on ART . Thus , our sample collection did not include IRIS-specific time points , but instead spans the timeframe in which IRIS developed . Nonetheless , as our observations persist across multiple time points , the lack of an IRIS time point did not affect our conclusions . Moreover , no differences could be observed when comparing early- and late-onset IRIS cases for any of the parameters analyzed . Finally , our patient selection included 2 unmasking IRIS patients , making our population more heterogeneous . These patients showed no significant variation compared to paradoxical IRIS patients for any of the clinical variables tested ( apart from EPG ) . However , unmasking Schisto-IRIS patients showed some modest differences in laboratory markers at month 3 . Since both cases initiated PZQ treatment after starting ART , this timing could have altered MMP/TIMP dynamics we observed at month 3 . Nonetheless , these patients also developed IRIS symptoms at this time point , which likely explains these differences . A larger unmasking Schisto-IRIS population is required to fully explore differences with paradoxical Schisto-IRIS . In conclusion , we describe characteristics of patients who developed IRIS related to S . mansoni in one of the first Schisto-IRIS cohorts to date . Therefore , we propose a refined working definition for the diagnosis of paradoxical and unmasking Schisto-IRIS in resource limited settings . Schisto-IRIS patients in our study showed persistent plasma levels of MMP-10 , along with a steep decline in TIMP-1 and low levels of TIMP-2 during 3 months of ART . Consistent with the IRIS symptoms reported in the definition , these aberrant MMP and TIMP dynamics suggest the presence of prolonged inflammation and/or tissue damage . Although further research is required , decreasing levels of anti-nuclear antibodies in Schisto-IRIS patients following PZQ/ART may reflect a PZQ-induced release of S . mansoni antigens , which in turn may drive Schisto-IRIS inflammation . Elucidating the immune pathogenesis behind this complication could lead to treatment strategies for Schisto-IRIS , as well as provide insight in the heterogeneous disease that is IRIS in general . | A subset of HIV-patients starting antiretroviral therapy are at risk of developing immune-driven worsening symptoms of a previously treated opportunistic infection . This paradoxical immune reconstitution inflammatory syndrome ( IRIS ) has been abundantly described in common co-infections such as M . tuberculosis ( TB-IRIS ) , whereas IRIS associated with Schistosoma mansoni ( Schisto-IRIS ) is less well studied . Nonetheless , HIV and S . mansoni are highly co-endemic in Sub-Saharan Africa and the considerable clinical burden of Schisto-IRIS in the field should not be underestimated . Moreover , no clinical definition exists to help diagnose this complication . Although little is known about the immune dysregulation in Schisto-IRIS , matrix metalloproteinases ( MMPs ) and tissue inhibitor of metalloproteinases ( TIMPs ) have been linked to schistosomiasis and TB-IRIS on account of their role in tissue-destructive inflammation . The current study is nested within a three-month case-control study in schistosomiasis/HIV co-infected fishermen starting ART in Kenya . We propose a clinical working definition for Schisto-IRIS , based on critical evaluation of symptoms developing during ART . Our study now links aberrant dynamics of MMPs and TIMPs to Schisto-IRIS as well . Given the role of MMPs and TIMPs in tissue remodeling and inflammation , our findings suggest that Schisto-IRIS is marked by unbalanced MMP/TIMP dynamics that favor inflammation . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"schistosoma",
"invertebrates",
"schistosoma",
"mansoni",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"helminths",
"antiviral",
"therapy",
"enzymes",
"immunology",
"tropical",
"diseases",
"enzymology",
"metalloproteases",
"parasitic",
"diseases",
"animals",
"diarrhea",
"preventive",
"medicine",
"signs",
"and",
"symptoms",
"antiretroviral",
"therapy",
"gastroenterology",
"and",
"hepatology",
"neglected",
"tropical",
"diseases",
"vaccination",
"and",
"immunization",
"digestive",
"system",
"public",
"and",
"occupational",
"health",
"inflammation",
"proteins",
"immune",
"response",
"gastrointestinal",
"tract",
"biochemistry",
"helminth",
"infections",
"schistosomiasis",
"eukaryota",
"diagnostic",
"medicine",
"anatomy",
"fevers",
"biology",
"and",
"life",
"sciences",
"proteases",
"organisms"
] | 2018 | Aberrant plasma MMP and TIMP dynamics in Schistosoma - Immune reconstitution inflammatory syndrome (IRIS) |
The detailed positions of nucleosomes profoundly impact gene regulation and are partly encoded by the genomic DNA sequence . However , less is known about the functional consequences of this encoding . Here , we address this question using a genome-wide map of ∼380 , 000 yeast nucleosomes that we sequenced in their entirety . Utilizing the high resolution of our map , we refine our understanding of how nucleosome organizations are encoded by the DNA sequence and demonstrate that the genomic sequence is highly predictive of the in vivo nucleosome organization , even across new nucleosome-bound sequences that we isolated from fly and human . We find that Poly ( dA:dT ) tracts are an important component of these nucleosome positioning signals and that their nucleosome-disfavoring action results in large nucleosome depletion over them and over their flanking regions and enhances the accessibility of transcription factors to their cognate sites . Our results suggest that the yeast genome may utilize these nucleosome positioning signals to regulate gene expression with different transcriptional noise and activation kinetics and DNA replication with different origin efficiency . These distinct functions may be achieved by encoding both relatively closed ( nucleosome-covered ) chromatin organizations over some factor binding sites , where factors must compete with nucleosomes for DNA access , and relatively open ( nucleosome-depleted ) organizations over other factor sites , where factors bind without competition .
DNA in eukaryotes is highly packaged into nucleosome arrays , which together compact ∼75–90% of the genome [1] . Because most DNA is wrapped in nucleosomes , and nucleosomes occlude their DNA from access to most other DNA binding proteins , revealing the detailed organization of nucleosomes across genomes and understanding the mechanisms that control their positioning is critical for understanding transcription factor binding and thus transcriptional regulation . Several studies predicted in vivo nucleosome positions directly from the DNA sequence [2]–[6] , suggesting that nucleosome organizations are partly encoded in the genomic sequence itself , through the nucleosomes' intrinsic DNA sequence preferences , which vary greatly between differing DNA sequences [7] , [8] . However , an intriguing and less explored question concerns the functional roles that this encoding may have . Studying this question requires detailed measurements of nucleosome organizations and availability of large-scale functional genomic data with which to compare these measurements . We thus focused on yeast , where many dynamic aspects of transcriptional regulation have been experimentally measured genome-wide , and where nucleosome occupancy have been measured using DNA microarrays [5] , [9] , [10] . To improve the resolution of the measured nucleosome organization , we used a parallel sequencing technology whose reads are longer than one nucleosome length , and obtained ∼380 , 000 fully sequenced yeast nucleosomes , resulting in a genome-wide map of nucleosome occupancy with high accuracy and dynamic range . While this manuscript was in review , two other studies that used parallel sequencing to map nucleosomes were published [11] , [12] . Here , we first use our map to better understand how nucleosome organizations are encoded by intrinsic signals in genomic DNA , and find that the genomic sequence is highly predictive of nucleosome organizations in yeast . By isolating nucleosome-bound sequences from fly and human , we further show that the key positioning signals in yeast are also predictive of nucleosome organizations in higher eukaryotes . Our results suggest that the yeast genome utilizes these intrinsic nucleosome positioning signals to encode both relatively open ( nucleosome-depleted ) and relatively closed ( nucleosome-covered ) chromatin organizations , resulting in two distinct modes of regulation by chromatin with different biological functions . In promoters that encode relatively open chromatin architectures , transcription factors can access their sites more freely , resulting in a homogeneous cell population with relatively low cell-to-cell expression variability , or transcriptional noise . Genes associated with these promoters are enriched in essential genes and in ribosomal protein genes . In contrast , in promoters that encode relatively closed chromatin architectures , factors compete with nucleosomes for access to the DNA , resulting in a heterogeneous cell population with higher transcriptional noise . Genes associated with these promoters are enriched in non-essential genes and in genes that are active only in specific biological conditions . Finally , we provide evidence that the encoding of relatively open and closed chromatin architectures may also play a role in DNA replication , such that replication origins that encode open chromatin organizations initiate replication with higher efficiency . Taken together , our results reveal new insights into the mechanisms by which the genomic DNA sequence dictates the nucleosome organization , and by which genomically encoded nucleosome organizations may influence chromosome functions .
To obtain a single molecule map of yeast nucleosomes , we isolated mononucleosomes from eight independent biological replicates and fully sequenced ∼503 , 000 of the nucleosome DNAs , using a parallel sequencing technology whose sequence reads are ∼200 bp long . Thus , aside from the limitations imposed by using micrococcal nuclease to isolate nucleosomes , our approach is optimal for mapping nucleosomes , since it extracts only the DNA segments of interest with little flanking DNA , and then reads them in full . Such full length nucleosome reads allow us to map the nucleosome organization with potentially greater resolution compared to approaches that map only one nucleosome end , because the kinetics of nuclease digestion result in nucleosomal DNA fragments that vary in length relative to the canonical 147 bp nucleosome , and thus , mapping only one end leaves considerable uncertainty regarding the location of the other end , for any given nucleosome DNA molecule . In addition , the sequencing method affords a large dynamic range , limited only by the number of sequence reads obtained . Compared to using microarrays as the readout of nucleosome occupancy , a sequencing-based approach provides an experimental decomposition of the average nucleosome occupancy , such as that measured by microarrays , into discrete nucleosome configurations . After excluding nucleosomes that map to repetitive regions , we obtained ∼380 , 000 uniquely mapped nucleosomes such that on average , every basepair is covered by five nucleosome reads ( Figure 1 ) . To validate our nucleosome map , we compared it to ∼100 nucleosome positions mapped using conventional sequencing [2] , three large collections of generic nucleosomes mapped using microarrays [5] , [9] , [10] , and two collections of generic [11] and H2A . Z [13] nucleosomes mapped by sequencing one end of each nucleosome . Our map shows significant correspondence with all existing maps but differs in both the detailed locations and occupancy of many measured nucleosomes ( Figure S1 ) . Before exploring the functional consequences of the intrinsically encoded nucleosome organization , we used the high resolution of the sequence-based nucleosome map to refine our understanding of how nucleosome organizations are encoded by the genomic sequence . Several models for predicting nucleosome positions from DNA sequence were recently constructed [2]–[6] . Our motivation for constructing a new model was twofold . First , none of these models were constructed from a genome-wide map of nucleosome positions based on direct sequencing , and we thus sought to utilize the high resolution and accuracy of such a map for constructing a model . Our second motivation was to combine into one model , two primary components that were each , separately , the basis of the previously published models . One of these components consists of periodicities of specific dinucleotides along the nucleosome length , on which earlier models were based [2] , [3] . The other component includes sequences that are generally disfavored by nucleosomes , regardless of their position along the nucleosome length , whose incorporation was shown to increase the predictive power [4]–[6] . Regarding the periodic component , several studies [2] , [3] , [14] , [15] characterized the nucleosomes' intrinsic sequence preferences primarily by ∼10 bp periodicities of specific dinucleotides along the nucleosome length , thought to facilitate the sharp bending of DNA around the nucleosome [16] . We find similar periodicities in our new large nucleosome collection , demonstrating that these periodic dinucleotides are important genome-wide ( Figure 2A and Figure S2 ) . These same periodicities also arise in H2A . Z-containing nucleosomes [13] , and in every in vivo and in vitro nucleosome collection obtained by direct sequencing from any organism [2] , [11] , [15] , [17]–[19] . Moreover , these periodicities are also present in yeast transcription start sites ( Figure 3 ) , worm introns , 5′ and 3′ UTRs [20] , human CpG dinucleotides not in CpG islands [21] , and HIV integration sites in human [22] . Other studies [4]–[6] focused on sequences that are generally disfavored by nucleosomes , regardless of their detailed position along the nucleosome . We thus used our map to systematically identify sequences that are generally disfavored by nucleosomes , by extracting from our map contiguous regions not covered by any nucleosome , and comparing the frequencies of 5-mers in these linker DNA regions to their frequencies in the nucleosome-bound sequences . Indeed , we find that many 5-mers are enriched in linkers , including AAAAA as the most dominant signal , as well as all other 5-mers composed exclusively of A/T nucleotides , and the repetitive sequence CGCGC , shown to disfavor nucleosome formation [23] ( Figure 2B ) . Notably , these same 5-mers are enriched in nucleosome-depleted regions from human [24] , further suggesting that they represent nucleosome-disfavoring elements , and that such disfavoring elements may be universal . We thus constructed a probabilistic nucleosome–DNA interaction model that integrates both the ( nucleosome-favorable ) position-specific periodic component and the ( nucleosome-disfavoring ) position-independent 5-mer component , and scores the nucleosome formation potential of every 147 bp sequence as the ratio between these components ( Figure 2C ) . In our model , the periodic component dictates the high-resolution positioning of nucleosomes ( known as the rotational setting ) , because its ∼10 bp periodicity results in strongly correlated scores between genomic positions separated by 10 bp , and strongly anti-correlated scores between positions separated by 5 bp . In contrast , the 5-mer nucleosome-disfavoring component scores each 147 bp sequence based on the set of its constituent 5-mers without regard to their exact position within the 147 bp sequence . Thus , scores of the 5-mer component primarily vary over longer genomic distances and hence this component dictates the absolute level of nucleosome occupancy of a region ( known as the translational setting ) . To validate our new model , we tested whether its predictions agree with the in vivo nucleosome map at the scale of individual nucleosomes . Specifically , we defined linkers as contiguous regions of lengths 50–500 bp that are not covered by any nucleosome , and evaluated the model's ability to separate these linkers from sets of nucleosomes with various levels of occupancy ( 1 , 2 , 4 , 8 , and 16 ) , where the occupancy of a nucleosome is defined by the number of nucleosome reads whose center is within 20 bp of its own center . We then scored each of the resulting linkers and nucleosomes as the mean score ( identical results were obtained by selecting the max score ) that our model assigns to the region that is 20 bp from the center of the linker or nucleosome , respectively . We used a cross validation scheme , in which model predictions on any given chromosome are computed from a model whose parameters were estimated only from the data of all other chromosomes . This way we can generate genome-wide nucleosome occupancy predictions at each chromosome , where the predictions on each chromosome were computed from models that were trained on other chromosomes . We use these cross-validation predictions in all of the following validation analyses . If the model were fully predictive of our in vivo map , then the model score of every nucleosomal region would be higher than that of every linker region . A standard quantification of this predictive power is the receiver operating characteristic ( ROC ) curve , whose area under the curve ( AUC ) is 1 for perfect performance and 0 . 5 for random guessing . We found a near-perfect AUC performance of 0 . 97 in separating ∼8 , 000 linkers from ∼12 , 000 regions that contain nucleosomes with a high occupancy of at least 8 nucleosome reads , and an AUC of 0 . 89 for separating these ∼8 , 000 linkers from ∼84 , 000 regions that contain nucleosomes with the minimal possible occupancy of one nucleosome read ( Figure 2D ) . For example , at the model score threshold in which 90% ( true positive rate ) of the nucleosomes of occupancy 8 are correctly predicted , less than 10% ( false positive rate ) of the linkers are incorrectly predicted as nucleosomes . The absolute performance in these tests is remarkable , and demonstrates that our model is highly predictive of nucleosome occupancy in yeast . We also find that the performance of the model in this cross validation scheme is nearly identical to its performance on the training data , suggesting that our model does not overfit the input data ( Figure S3 ) . The fact that the model performs better in classifying nucleosomes with higher occupancy indicates that the probability that a nucleosome will occupy a region within the genome is higher at regions that match the sequence preferences of nucleosomes , as represented by our model . Note that since our predictions are done in a cross validation scheme , this result is not a trivial consequence of our training procedure , since a trained model does not have access to the level of occupancy of the nucleosomes on which its predictions are tested . To calibrate the performance of our model , we compared it to the performance of previously published methods , and found that our model performs better than previous approaches when tested on our data ( Figure S3 ) . Similarly , we observed highly significant predictive power on two microarray-based nucleosome maps [5] , [10] ( Figure S3 ) . Here , three models achieved the best , equivalent performance [5] , [6] , and our model was among them . Despite the outcome of these comparisons , we note that it is difficult to conclude from these tests which model is best , since for such an objective evaluation , each model should be trained using exactly the same input data , and such a comparison is out of our current scope and objective . Nevertheless , the performance of all of these models strongly supports the overall conclusion that the genomic sequence is highly predictive of nucleosome organizations in yeast . Recent analyses of genome-wide nucleosome occupancy measurements in yeast identified different classes of nucleosome occupancy patterns in gene promoters , by clustering the nucleosome occupancy patterns [5] . Notably , we find that our model is also able to accurately predict the occupancy patterns of these different classes , suggesting that these differing nucleosome occupancy patterns are partly encoded in the DNA sequence , through the nucleosome sequence preferences ( Figure 4 ) . Taken together , we conclude , in accord with other recent studies [2]–[6] , that the genomic sequence is highly predictive of the nucleosome organization in yeast . Finally , we tested whether the nucleosome positioning signals of our model are also predictive of nucleosome occupancy in higher eukaryotes . To this end , we obtained nucleosome datasets from yeast [13] , worm [17] , and chicken [15] , and also isolated and sequenced two new independent nucleosome collections from fly and two from human . Since there is variability in the base composition of different regions in the human genome , in one of the human collections , we extracted nucleosome-bound sequences from regions of the human genome that are strongly enriched in G/C nucleotides ( 60% G/C , see Methods ) , allowing us to evaluate the model performance on regions with atypical base compositions . In addition , we isolated and sequenced nucleosomes reconstituted in vitro on human genomic DNA and also obtained a previous such in vitro-selected collection from yeast [2] , allowing us to test whether the model mainly captures nucleosome sequence preferences ( since the in vitro experiments are done with purified histone octamers assembled on purified genomic DNA ) . To test whether the nucleosome positioning signals that we find in yeast are also important in these in vitro collections and in the collections from higher eukaryotes , we evaluated the model's performance locally around the ∼200–2000 nucleosomes that were mapped in each collection . The idea behind this test is that relative to the genomic location of a given nucleosome-bound sequence , a predictive model should assign higher scores to the position of that sequence , compared , for example , to scores that it assigns to positions that are half a nucleosome away from that position . For all of the following tests , we used our above model , learned only from the nucleosome-bound sequences that we measured in yeast . Notably , in all of the above 12 nucleosome collections , our model assigns , on average , significantly higher scores around the center of the mapped nucleosome locations compared to scores that it assigns to nearby regions , suggesting that the nucleosome positioning signals of yeast are indeed predictive of nucleosome organizations in other eukaryotes ( Figure 5 ) . We also separately evaluated each of the two components of our model . We find that in all 10 collections obtained by direct sequencing , the periodic dinucleotide component alone predicts the correct rotational setting to within a 5 bp resolution , since on average , it assigns a higher score to the center of the nucleosome bound sequences in each collection compared to the score that it assigns to positions that are 5 bp away from that center ( Figure 5 ) . Similarly , in all 12 collections , the nucleosome disfavoring component of our model alone predicts the correct translational settings of the nucleosomes in each collection , since on average , it assigns a lower score to the center of the nucleosome bound sequences in each collection , compared to scores that it assigns in nearby regions ( Figure 5 ) . We also note that the 4th order Markov model alone ( this component is the constituent repeating component of the 147 bp nucleosome disfavoring component ) , readily reveals that its preferred and disfavored 5-mers , learned only from yeast , show similar preferences in these nucleosome collections from higher eukaryotes , such that linkers contain more nucleosome-disfavoring sequences ( Figure 5 ) . The success of our model , which is trained only on yeast nucleosomes , in predicting nucleosome locations across several eukaryotes , suggests that the key nucleosome positioning signals of our model , such as its periodic pattern and 5-mer sequence preferences ( and negative preferences ) , represent nucleosome sequence preferences , and are universal across eukaryotes . Clearly , although this result demonstrates that the nucleosome positioning signals of yeast apply to higher eukaryotes , it does not show that these positioning signals are the only ones that determine nucleosome positioning in higher eukaryotes , and it will be interesting to examine these questions using recent large-scale nucleosome maps in higher eukaryotes [25] , [26] . To better understand the effect of nucleosome-disfavoring sequences on the local depletion of nucleosomes , we focused on the association between nucleosome occupancy and homopolymeric tracts of A or T , termed Poly ( dA:dT ) elements , since in our data , AAAAA is the 5-mer with the strongest enrichment in linkers ( Figure 2B ) . Several studies examined this relationship [27]–[33] , and suggested that Poly ( dA:dT ) elements may be rigid in vitro [30] and in vivo [28] , resulting in a reduced affinity to nucleosomes [34] . These elements are enriched in eukaryotic , but not in prokaryotic , genomes [35] , and were shown to have important functions in vivo [27] , [29] , most likely mediated by their nucleosome disfavoring action [29] , [36] , [37] . Consistent with this hypothesis , microarray-based maps of yeast [5] , [9] and human [24] nucleosomes showed nucleosome depletion over Poly ( dA:dT ) elements . However , none of these studies focused specifically on quantifying the fold depletion over Poly ( dA:dT ) elements . To quantify the fold depletion over a set of Poly ( dA:dT ) elements of interest , we compare the observed and expected number of nucleosomes that cover these elements . For example , 100 Poly ( dA:dT ) elements whose combined length is 1 , 470 bp and that are collectively covered by only one nucleosome read are depleted by 50-fold , since according to the average genome-wide coverage of our map , which is 5 nucleosomes per basepair , we expect these regions to be covered by 50 nucleosome reads . Plotting these fold depletions over Poly ( dA:dT ) elements of varying lengths , we find large depletions over these elements , that steadily increase with their length ( Figure 2E and Figure S4 ) . For example , there is a 12-fold depletion of nucleosomes over the 225 Poly ( dA:dT ) elements in the yeast genome whose size is at least 17 bp . We found similarly large fold depletions over Poly ( dA:dT ) elements with several basepair substitutions and in clusters of short Poly ( dA:dT ) elements that alternate between strands ( Figure 2E and Figure S4 ) . The depletion over these imperfect elements also increases with their length . The large nucleosome fold-depletions over these sequence elements mean that these elements effectively create boundary zones , dividing the genome into discrete chromatin blocks; for simplicity , we henceforth refer to the sequence elements themselves as “boundaries” . The strength of a boundary , defined here as the fold depletion over all of its instances in the genome , can be estimated from DNA sequence alone , based on the length and perfection of its Poly ( dA:dT ) components . For example , Poly ( dA:dT ) elements of length 20 with two basepair substitutions have a 6-fold nucleosome depletion ( Figure 2E ) . We find 673 boundary elements in the yeast genome even at fold depletions of more than 10 , and these elements are primarily located in non-coding regions ( Figure S5 ) . A possible concern is that the nucleosome depletion that we observe over sequence boundaries results from artifacts in our experimental method . Two main concerns arise in this respect . First , the depletion over boundaries may result from biases in the sequencing technology that we employed . Arguing against this , however , are the facts that nucleosome depletion over Poly ( dA:dT ) elements was observed using the independent technologies of microarrays [5] , [9] , [10] , [24]; using alternative sequencing-based approaches that utilize short reads only and thus do not need to read through a Poly ( dA:dT ) element itself [13]; and that the effect we see is not restricted to perfect Poly ( dA:dT ) elements , which could conceivably be problematic [38] , but includes elements with many basepair substitutions ( Figure 2E ) and elements that alternate between Poly-A and Poly-T tracts on each strand ( Figure S4 ) . Together , these facts imply that the observed depletions do not result from an inability of our procedure to provide sequence reads from DNA fragments that contain Poly ( dA:dT ) elements . A second possible concern may arise from the use of micrococcal nuclease to isolate nucleosomes , since this enzyme was used in both our study and in all of the studies that used microarrays or alternative sequencing-based strategies to map nucleosomes . The concern is that if the sequence specificity of micrococcal nuclease was biased towards Poly ( dA:dT ) elements , then its use may select against nucleosome DNAs containing these sequence elements . However , such an effect is unlikely because stretches of pure Poly ( dA:dT ) do not match the known specificity of micrococcal nuclease [24] , [39] , and hybridizations of micrococcal nuclease-treated naked DNA show little correlation with measured nucleosome locations [9] . To confirm that nucleosome depletion over Poly ( dA:dT ) elements is not a result of the sequence specificity of micrococcal nuclease , we examined the ∼1 million cut sites of micrococcal nuclease provided by our data ( since we sequenced ∼500 , 000 individual nucleosomes altogether , and each nucleosome is sequenced in full , thereby providing two cut sites ) . By aligning all of these cut sites , we find that the sequence specificity in these cut sites is highly similar to that reported previously [39] , and that it has very little information content ( i . e . , the specificity of the nuclease is low , confined mainly to two basepairs ) . This means that a preferred sequence for micrococcal nuclease can be found in nearly every small stretch of DNA in the yeast genome ( Figure 6A ) . Moreover , ranking all of the 4096 possible 6-mers by their preference to be cut by micrococcal nuclease , defined as the ratio between the probability that they appear as a cut site and the probability that they appear in the yeast genome , we find that AAAAAA is ranked 1782 out of the 4096 possible 6-mers as a micrococcal nuclease cleavage site ( Figure 6B ) , while it ranks number 1 for its observed in vivo nucleosome depletion ( Figure 2A ) . In addition , plotting the distribution of Poly ( dA:dT ) elements as a function of their distance from all cut sites obtained in our data , we find that the most likely position for Poly ( dA:dT ) elements relative to cut sites is ∼50 bp from the cut site , which is consistent with the enrichment of Poly ( dA:dT ) elements in linker DNA regions , but not with the idea that Poly ( dA:dT ) elements are preferentially cut by micrococcal nuclease ( Figure 6C ) . Thus , the relative lack of nucleosome occupancy over Poly ( dA:dT ) elements in vivo is not attributable to these sites being preferentially degraded by the micrococcal nuclease . Taken together , the existing literature , the above analyses , and additional new experimental data that we present in a later section below , strongly suggest that the in vivo depletions that we observe over Poly ( dA:dT ) elements are not an artifact of our analysis , but a real phenomenon . What may cause the observed nucleosome depletion over boundaries ? One possible mechanism is through the action of DNA binding proteins that recognize and bind these elements . To date , a single protein in S . cerevisiae , called Datin ( Dat1p ) , that recognizes Poly ( dA:dT ) elements has been identified [40] . The binding specificity of Datin requires at least 9 basepairs of A or T nucleotides , and it appears to be the only DNA binding protein in S . cerevisiae that binds Poly ( dA:dT ) elements , since cell extracts of a Datin deletion yeast strain do not exhibit any detectable protein binding to Poly ( dA:dT ) elements [40] . However , Datin is unlikely to be the major cause of nucleosome depletion over boundaries , based on the sequence diversity of Poly ( dA:dT ) elements that we find to be depleted yet that do not match the binding specificities of Datin , on the steady increase of the depletion with the length of the Poly ( dA:dT ) elements ( Figure 2E ) , and on other studies that concluded that Datin is not important for the function of Poly ( dA:dT ) elements [28] , [29] , [36] , [37] , [41] , [42] . Another possibility is that the binding of transcription factors to sites near the boundaries causes nucleosome depletion over boundaries . Indeed , such an effect is to be expected on thermodynamic grounds; the question is the relative significance of this effect . To test this , we compared the nucleosome occupancy over boundaries that are near factor binding sites , to that over boundaries that are far from factor sites . We find strong nucleosome depletion over boundaries regardless of whether or not they are near factor sites ( Figure 7B ) . This result is not sensitive to binding site annotations , since we find a similar strong depletion over boundaries in intergenic regions that are not promoters , thought to be largely devoid of factor sites ( Figure 7B ) . These results suggest that transcription factor binding is not the main cause of nucleosome depletion over the boundary sequences . A remaining alternative is that Poly ( dA:dT ) elements themselves intrinsically disfavor nucleosome formation . This possibility was suggested previously , on the basis of studies done on a handful of genes [29]–[31] , [36] , [37] , though other single gene studies [43] , [44] concluded that nucleosome exclusion by Poly ( dA:dT ) elements cannot account for the full effect of Poly ( dA:dT ) elements . Studies of the structure and mechanics of Poly ( dA:dT ) elements further support that these elements act through nucleosome exclusion , since these tracts may be mechanically stiff and thus resist wrapping into nucleosomes [30] . If nucleosome exclusion is the primary mechanism by which Poly ( dA:dT ) elements exert their effect , then we might also expect these elements to show a reduced affinity for nucleosome formation in vitro . One study addressed this question , and demonstrated that incorporating a perfect Poly-A ( 16 ) element into a ( non-natural ) DNA sequence disfavors nucleosome formation , with an effect of about two-fold on DNA accessibility [31] . To examine whether natural boundary sequences also exhibit reduced nucleosome affinity in vitro , we selected three Poly ( dA:dT ) -containing regions from the yeast genome that each contain multiple Poly ( dA:dT ) elements and measured the relative affinities of these regions for nucleosome formation along with the relative affinities of four sequence variants that disrupt one of the Poly ( dA:dT ) elements in each sequence . Like many of the other Poly ( dA:dT ) elements in the genome , the Poly ( dA:dT ) elements that we selected exhibit nucleosome depletion in vivo ( Figure 8A–C ) . Consistent with earlier measurements [31] , we find that all seven Poly ( dA:dT ) -containing sequences have significantly reduced affinities , comparable to affinities of DNA sequences that were selected for their ability to resist nucleosome formation [45] ( Figure 8D and 8E ) . These relative affinity measurements for nucleosome formation were performed as previously described [2] , [7] . We also examined a systematic study of in vitro nucleosome reconstitution on ∼10 kbp from the β-Lactoglobulin locus of sheep [8] , and found strong nucleosome depletion over the one Poly-T ( 13 ) element in the locus ( Figure S6 ) . Taken together , these results demonstrate that sequence boundaries have an intrinsically reduced affinity for nucleosome formation . Thus , our new in vitro measurements of nucleosome formation on boundaries , combined with the conclusions reached by previous studies , and with the conclusion that Poly ( dA:dT ) -binding proteins and transcription factors cannot account for the in vivo depletion over Poly ( dA:dT ) elements , strongly suggest that the large in vivo depletion over Poly ( dA:dT ) elements is the consequence of a nucleosome-disfavoring character of these elements . What may be the function of sequence boundaries ? In the extreme case , a strong boundary that cannot be occupied by a nucleosome creates , on average , a nucleosome-depleted region centered on but larger than the boundary itself , simply because there are a smaller number of different nucleosome configurations in which basepairs that are close to the boundary can be occupied by a nucleosome , compared to basepairs located further away from the boundary . For example , a basepair immediately flanking the boundary can only be occupied by the one configuration in which a nucleosome is placed immediately adjacent to the boundary , whereas a basepair located 5 bp from the boundary can be occupied by any of 5 different nucleosome configurations ( Figure 7A ) . Ignoring the nucleosome sequence preferences for a moment , and assuming for simplicity that all allowed nucleosome positions are equally likely , then , in the above example , the basepair immediately flanking the boundary is 5-times less likely to be occupied by a nucleosome , compared to the basepair located 5 bp away from the boundary . Thus , the mere presence of a boundary acts as a force that , on average , creates a nucleosome-depleted region extending into the adjacent DNA [46] . Based on the above reasoning , we hypothesized that the flanking regions of our above Poly ( dA:dT ) boundaries will be depleted of nucleosomes , and we expect the strength of the effect to increase with the strength of the boundary . Indeed , examining the nucleosome occupancy in the vicinity of boundaries , we find large levels of nucleosome depletion even 50 bp away from a boundary , regardless of whether or not the boundary is located close to a transcription factor binding site , and whether or not the boundary is located in a promoter region or in intergenic regions that are not promoters ( Figure 7B ) . Moreover , examining the distribution of boundaries around transcription start sites where previous studies [5] , [9] , [13] found a stereotyped nucleosome depleted region , and around translation end sites where similar depletions were observed [11] , [12] , [26] , we find that both the depletion level and length of these depleted regions strongly correlate with the boundary strength ( Figure 9A and 9B ) . As expected , these differing nucleosome organizations around both transcription start sites and translation end sites are accurately predicted by our sequence-based model for nucleosome positioning ( Figure 9C and 9D ) . These results are consistent with the theoretical analysis of Kornberg and Stryer [46] , although , their boundary constraint was thought to be due to transcription factors , whereas we show that a boundary constraint arises also simply from the presence of Poly ( dA:dT ) -based sequence elements , through their reduced affinity for nucleosome formation . Our results thus suggest that relatively large open chromatin regions can be accurately predicted simply by the presence of Poly ( dA:dT ) elements , consistent with the suggestion that boundaries such as Poly ( dA:dT ) elements account for many aspects of the in vivo nucleosome organization [9] , [12] . If boundaries indeed cause nucleosome depletion at their flanking regions , then boundaries may enhance the accessibility of transcription factors to binding sites that are located close to the boundary . Indeed , we find strong nucleosome depletion over factor sites that are near boundaries , compared to a much weaker depletion over factor sites that are far from boundaries ( Figure 7C ) , suggesting that nucleosome depletion over many factor sites is partly encoded through the sequence preferences of nucleosomes , by the nucleosome-disfavoring action of Poly ( dA:dT ) elements . These results are consistent with studies done at a few loci , which suggested that Poly ( dA:dT ) elements may generally function to enhance the accessibility of transcription factors to their cognate sites [27] , [29] . We next asked whether nucleosome depletion over factor sites depends on the boundary strength and factor-boundary distances . Notably , the level of nucleosome depletion over factor sites increases significantly with both the strength of the boundary and its proximity to factor sites ( Figure 10A ) . Specifically , for 50 of 51 factors for which more than 10 sites are annotated [47] , we find stronger nucleosome depletion at the subset of its sites that are near boundaries compared to its other sites ( Figure 10B ) . The only exception is Reb1 , a highly abundant factor that possesses ATP-independent chromatin remodeling activity [48] . Taken together , our results demonstrate that boundaries enhance the accessibility of transcription factors to their cognate sites , by depleting nucleosomes from the adjacent DNA , with the magnitude of such depletion increasing with both the strength of the boundary and its proximity to the factor site . We hypothesized that since factor binding sites near boundaries are depleted of nucleosomes , factors could bind such sites in promoters with little or no competition with nucleosomes , leading to a homogeneous cell population with relatively low cell-to-cell expression variability , or transcriptional noise . In contrast , since steric hindrance may not permit simultaneous binding by factors and nucleosomes , factors that bind sites that are far from boundaries may need to compete with nucleosomes for access to the DNA . Such a competition may result in a mixed population comprising both cells in which a nucleosome covers the factor's site and the promoter is inactive , and cells in which that nucleosome is displaced and the promoter is active . To test this hypothesis , we utilized a dataset [49] , which for the majority of the genes in yeast , used a GFP-tagged strain to measure their protein expression variability in single-cells . Since they are easier to obtain , such measurements of variability at the protein level are typically used as a proxy for variability measurements at the RNA level [49]–[51] . This approach is justified by the experimental observation that variability in protein expression is dominated by variability in RNA levels [49] . Using these data , we compared the noise of promoters in which the sites [47] are covered by nucleosomes , to the noise of promoters in which the sites are not covered . Indeed , the former promoter set exhibits significantly more noise ( P<10−5 , Kolmogorov-Smirnov test ) . A similar model , in which high noise promoters are those where nucleosomes compete successfully with transcription factors , was suggested and validated for the Pho5 gene [51] . That model further suggested that the presence of TATA sequences should confer even more noise , presumably through facilitation of transcription re-initiation [51] , [52] . Thus , under this noise model , we expect , and indeed find , that within each of our two promoter sets above , the presence of TATA [53] elements further increases transcriptional noise ( Figure 11A ) . We further examined those promoters having TATA elements and nucleosome-covered factor binding sites , and those promoters lacking TATA elements and having nucleosome-depleted factor binding sites , since these promoter sets are the most and least noisy promoters , respectively ( Figure 11A ) , and they each have more genes than would be expected ( Figure 11B ) . Intriguingly , in addition to their differential noise , we also find distinct promoter architectures and nucleosome dynamics in these two promoter types . Type I promoters , which contain TATA elements and whose sites are nucleosome-covered , have many factor sites spread across the promoter region , a weaker signal of nucleosome depletion at the typical nucleosome depleted region ( NDR ) , and are enriched in targets of condition-specific factors and non-essential genes ( Figure 12A and 12B and Figure S7 ) . These promoters are targets of chromatin remodeling complexes [54] and their rate of histone turnover [55] is significantly high ( Figure 11C ) , consistent with an ongoing dynamic competition between nucleosome assembly and factor binding . In contrast , type II promoters , which are TATA-less and whose sites are nucleosome-depleted , have strong nucleosome depletion , many boundary elements at the typical NDR , low histone turnover , and an overall smaller number of factor sites but with a high preference for these sites to be located at the NDR ( Figure 12A ) . Type II promoters are enriched in essential genes and in ribosomal protein genes , the latter presumably owing to the fact that these proteins are highly expressed and are required stoichiometrically in a large complex , thereby conferring a benefit to regulation with low noise ( Figure 12B ) . While our paper was in review , analysis of nucleosome occupancy data resulted in a similar two-class partition of yeast promoters [56] . We find that our sequence-based nucleosome–DNA interaction model accurately predicts the different nucleosome organizations observed for each promoter type , suggesting that their distinct nucleosome architectures are partly encoded in the genome through the sequence preferences of nucleosomes ( Figure 12C ) . In fact , we can distinguish low- and high-noise promoters using only sequence information , by partitioning promoters according to the presence of our Poly ( dA:dT ) -boundaries and TATA elements ( Figure 11D ) . Taken together , our results point to a strong association between chromatin and transcriptional noise at the genome-wide level , as suggested on the basis of one gene [51] , and further uncover two distinct types of chromatin architectures by which high or low noise may be implemented in yeast promoters . Finally , analogous to the cell-to-cell variability observed in gene expression [49] , DNA replication origins also exhibit variability , with some origins initiating replication in most cell divisions and others initiating only occasionally . We examined whether this variability can be partly explained by differing nucleosome positioning signals in the two types of origins . In general , DNA replication origins are A/T- and Poly ( dA:dT ) -rich [57] , [58] and thus may disfavor nucleosome formation . Indeed , we find an overall ( both measured by our data and predicted by our model ) nucleosome depletion around replication origins in S . cerevisiae ( Figure 13A ) , and similar ( predicted ) depletion around origins in S . pombe ( Figure 13B ) . Consistent with the hypothesis that competition with nucleosomes may affect the efficacy of replication initiation [59] , a systematic sequence deletion study [60] around one replication origin in S . pombe found that deletion of a strong nucleosome-disfavoring element ( Poly-A ( 20 ) ) resulted in the largest reduction in replication efficiency ( Figure 14 ) . Similarly , for S . pombe , where data on efficiency of replication initiation are available [61] ( such data are not available for S . cerevisiae ) , we find on a genome-wide scale , that replication origins with lower ( predicted ) nucleosome occupancy initiate replication with higher efficiency ( P<10−6; Figure 13C and 13D ) .
For our data , model and genome-wide occupancy predictions in yeast , worm , fly , mouse , and human , and sequences provided by researchers , see http://genie . weizmann . ac . il/pubs/field08 . Our results are also viewable in Genomica ( http://Genomica . weizmann . ac . il ) .
Mono-nucleosomes were extracted from log-phase yeast ( Saccharomyces cerevisiae ) cells using standard methods . The DNA ( pooled together from eight independent biological replicates ) was extracted , and protected fragments of length ∼147 bp were sequenced using 454 pyrosequencing . Each of the resulting 503 , 264 sequence reads was mapped to the yeast genome using BLAST [62] requiring at least 95% identity . Sequences were further filtered by requiring that they: map to a unique genomic location; are of length 127–177 bp; do not overlap the ribosomal RNA locus ( chromosome 12: 451550–490540 bp ) . The resulting 378 , 686 nucleosomes constitute the nucleosome collection used in all of our analyses . We used a sequencing technology whose reads are ∼200 bp in length , and thus , each of the nucleosomal DNA fragments was read in full . These full sequence reads allow us to map both ends of each nucleosomal DNA fragment to the genome , without having to infer its other end , as is the case when using sequencing technologies with shorter reads that map only one nucleosome end . Two fly and one human in vivo nucleosome collections were obtained from fly ( Drosophila Melanogaster , S2 cells ) and human ( HeLa cells ) . Nuclei were prepared using standard methods , and the chromatin digested to primarily mononucleosomes using micrococcal nuclease . The DNA was extracted , and protected fragments of length ∼147 bp were cloned and sequenced as described [2] . An additional human in vivo nucleosome collection that is strongly enriched in G/C nucleotides ( 60% G/C ) was obtained by digesting the isolated human mononucleosomal DNA with two restriction enzymes: Mse I , and Tsp509 , with specificities of TTAA and AATT , respectively . DNA fragments remaining ∼147 bp in length following these digestions were gel purified , cloned , and sequenced . A collection of human in vitro nucleosome sequences was obtained as described previously from yeast [2] except using human genomic DNA instead of yeast DNA [7] . The resulting sequences from each experiment were mapped to their respective genomes using BLAST [62] requiring at least 97% identity . Sequences were further filtered by requiring that they map to a unique genomic location and have a length in the range 142–152 bp . The resulting sequences constitute the three fly and three human nucleosome collections used in our analyses and they have 99 ( fly 1; in vivo ) , 170 ( fly 2; in vivo ) , 329 ( human 1; in vivo ) , 208 ( human 2; in vivo G/C ) , and 176 ( human 3; in vitro ) sequences . The yeast genome sequence ( May 2006 build ) and gene and chromosome annotations were obtained from SGD [63] . Yeast transcription start sites were compiled from [64]–[66]: for each gene , the transcription start site was taken as that with the most sequence reads from [64] , [65] , or from [66] when no sequencing data was available . Functional transcription factor DNA binding sites in yeast , defined as sites that are bound by their cognate transcription factor were obtained from [47] , [67] . TATA elements in yeast were obtained from [53] . Functional annotations for yeast genes were downloaded from Gene Ontology [68] . Yeast genes bound by chromatin remodeling factors were obtained from [54] . Measurements of protein expression variability , referred to here as transcriptional noise , were obtained from [49] . Histone turnover rates at yeast promoters were obtained from [55] . Nucleosome-bound DNA sequences were obtained from: yeast [2] , [5] , [9] , [10] , worm [17] , chicken [15] . Microarray-based nucleosome maps of yeast ( 3 maps ) and human ( 1 map ) were obtained from [5] , [9] , [10] , [24] . The nucleosome fold depletion over a set of genomic regions of interest is defined as the ratio between their expected and actual nucleosome coverage . The expected coverage is equal to the average number of nucleosomes that cover a basepair in the genome , computed by dividing the total number of basepairs covered by our 378 , 686 nucleosome reads , with the total number of basepairs in the genome that are not in the ribosomal DNA locus or in repetitive regions . The actual nucleosome coverage over a set of genomic regions is computed as above , but only across the basepairs in the given set of genomic regions . In our data , the expected coverage is 5 . 27 . Thus , for example , a set of genomic elements whose actual average coverage per basepair is 0 . 1 , is depleted by 5 . 27/0 . 1 , or 52 . 7-fold . We use two sequence definitions for boundary elements . The first is based on single homopolymeric tracts of Poly-A or Poly-T ( Poly ( dA:dT ) elements ) , and the second on clusters of short Poly ( dA:dT ) elements . For the definition based on a single Poly ( dA:dT ) element , we iterate over allowed values k = 0 , 1 , 2 , … , 20 , for the number of mismatches relative to the Poly ( dA:dT ) tract . For each k , we then identify all maximal Poly ( dA:dT ) tracts in the genome with exactly k mismatches , where the mismatch cannot occur at the first or last basepair of the element . By maximal elements , we mean that if a Poly ( dA:dT ) element with exactly k mismatches is fully contained within a longer Poly ( dA:dT ) element with exactly k mismatches , then only the longer element is considered . For the definition based on clusters of short Poly ( dA:dT ) elements , we first define short Poly ( dA:dT ) elements as all Poly ( dA:dT ) elements with zero mismatches whose size is at least 5 bp . For each allowed value in the range k = 0 , 1 , 2 , … , 20 , representing the number of mismatches , we then identify maximal clusters of the above short Poly ( dA:dT ) tracts with exactly k mismatches . As with single Poly ( dA:dT ) elements , mismatches cannot occur at the first or last basepair of each cluster and maximal elements are defined similarly . Note that in the definition based on Poly ( dA:dT ) clusters , the resulting boundaries may contain Poly ( dA:dT ) elements that alternate between strands ( e . g . , AAAAATTTTTT ) . For various analyses , we partitioned boundaries into distinct groups based on their nucleosome fold depletion , which we refer to as their strength . To this end , we first compute the nucleosome fold depletion ( strength ) over the set of boundaries with exactly k mismatches and whose length is at least n , for k = 0 , 1 , 2 , … , 20 and all values of n for which elements of that size exist . This computation is performed separately for each of the two boundary definitions above ( single Poly ( dA:dT ) elements and clusters of Poly ( dA:dT ) elements ) . For a given requested partition of boundaries into strength groups , we then assign each set of boundaries with strength s to the strongest group among the groups whose strength is below s . Throughput this paper , we partitioned boundaries into groups of strength 2 , 5 , 10 , and 20 . Thus , for example , a set of boundaries whose strength is 30 will be assigned to the boundary group of strength 20 . In cases of overlap in the genomic coordinates of boundary elements assigned to the same group , we take only the boundary with the smaller number of mismatches; and if the number of mismatches of the overlapping boundaries is the same , we take only the longer boundary . Finally , since the actual fold depletion of a boundary group by this procedure may differ from its original requested fold depletion , we compute and use the actual fold depletion of the boundary group for the various graphs that show plots as a function of boundary strength . For the analyses of Figure 11 , we grouped promoters into four classes , based on the presence of TATA elements and on whether or not their binding sites are covered by nucleosomes or are nucleosome-depleted . TATA boxes are taken from [53] . We classify a promoter as having sites that are covered by nucleosomes if at least 80% of the total basepairs of its binding sites are covered by at least one nucleosome read from our data . We classify a promoter as having sites that are nucleosome-depleted if at most 20% of the total basepairs of its binding sites are covered by at least one nucleosome read . Binding sites are taken from [47] , [67] . We represent nucleosome sequence preferences using a probabilistic model that assigns a score to every 147 basepair ( nucleosome-length ) sequence . As discussed above , our model consists of two main components , each of which was separately and previously explored by published models . The first component , PN , represents the distribution over dinucleotides at each position along the nucleosome length , and thus captures the periodic signal of dinucleotides along the nucleosome . The second component , PL , represents the position-independent distribution over 5-mers at linker regions compared to nucleosomal DNA , and thus captures sequences that are generally favored or disfavored by nucleosomes regardless of their detailed position within the nucleosome . We chose to represent this component using 5-mers , since this is the highest order k-mer for which our data has sufficient statistics to robustly estimate each of the associated parameters , and the k-mer order that results in the highest AUC performance in a cross validation scheme . The final score that our model assigns to a 147 bp sequence S is then given by the log-ratio of these two model components: ( 1 ) where PN , i is the ith component of the dinucleotide model component and represents the conditional probability distribution over nucleotides at position i given the nucleotide that appeared at position ( i−1 ) , and Pl is the position-independent component of the second component of our model ( PL ) . Note that PN , 1 is represented by a mononucleotide model over the nucleotide at the first position . We now describe in detail how each of the two components of our model is derived . To compute the position-specific dinucleotide component of our model , PN , we start with a collection of nucleosome-bound sequences , and estimate PN from the 23 , 076 nucleosome sequence reads of length 146–148 . We restricted ourselves to this length range of nucleosomes , since the border of the nucleosome is the most likely cut site for the nuclease and thus these nucleosome reads are likely to be mapped with the highest accuracy . Indeed , these nucleosome reads exhibit clear periodicities of dinucleotides along the nucleosome length , similar to those reported previously [2] , [15] ( Figure S2 ) . For the estimation , we first align all sequences about their center , where each sequence is added twice to the alignment , once in its original form and once in its reverse complement form , to account for the 2-fold symmetry in the nucleosome structure [69] . Sequences of even length are treated as two sequences , each with a weight of 0 . 5 , once in a configuration that has one more base at the left side of the alignment , and once in a configuration that has one more base to the right of the alignment . This accounts for the uncertainty we have in the positioning of the even length sequences relative to the center . With each position i , we then associate a dinucleotide distribution , PN , i , which we estimate from the combined dinucleotide counts at alignment positions [i−2 , i−1] , [i−1 , i] , and [i , i+1] ( the two end positions of the nucleosome are averaged with less positions ) . Combining the dinucleotides at the two neighboring positions smoothes the resulting dinucleotide distribution at each position with a 3 basepair moving average , and is motivated by the experimental evidence that small ±1 basepair changes in spacing of key nucleosome DNA sequence motifs can occur with relatively small cost to the free energy of histone–DNA interactions [70] . To remove sequence composition biases from this component , we normalize the distribution , by dividing the final probability of every dinucleotide at each position by the probability of that dinucleotide across all positions , and finally normalize the resulting weights to a probability distribution . We used this estimation procedure in the 127 central positions of the nucleosome , and we force a uniform distribution over the 10 remaining positions at each end of the nucleosome profile . This was done to avoid biases in nucleotide distributions that may arise from the sequence specificity of the micrococcal nuclease used to isolate the nucleosome , since this way we do not include statistics that are taken from the cut site of the nuclease . Note that our above construction produces a reverse complement symmetric distribution , i . e . , the probability of a sequence and its reverse complement are equal . The position-independent component of our model , PL , whose purpose is to represent sequences that are generally favored or disfavored regardless of their position within the nucleosome , assigns a score to each 147 bp sequence , as the product of a position-independent Markov model , Pl , of order 4 . Thus , Pl defines a probability distribution over every one of the 1024 possible 5-mers . We chose to model the distribution over 5-mers , since this is the highest order in which our data still provides sufficient statistics to robustly estimate the value of each of the 1024 parameters . Given a collection of nucleosome-bound sequences , we set the weight of each 5-mer to the ratio between the frequency of that 5-mer in the linkers , and the frequency of that 5-mer in the nucleosome-bound sequences , where this ratio is then scaled to be a probability by dividing it by the sum of ratios across all 5-mers . As linkers , we take all 8022 contiguous non-repetitive regions of length 50–500 bp that are not covered by any nucleosome from the input collection . All 344 , 976 nucleosome-bound sequences of length greater than 146 are taken as the set of nucleosomes , and statistics are collected only from their central 127 bp to avoid alignment issues whereby the outermost regions of any given nucleosome may in fact be linkers . From the above linker DNAs , we ignored the statistics of the 5 basepairs at the end of each linker , to avoid biases that may be introduced from the sequence specificity of the micrococcal nuclease used in our experiments to isolate nucleosomes . Thus , this Markov model , Pl , includes contributions from both sequences that are disfavored by nucleosomes and sequences that are favored by nucleosomes , since it models the distribution over all 5-mers , with the disfavored sequences having a relatively high probability and the favored sequences having a relatively low probability . We note , that although we discuss each of the two model components separately , these components are in fact not independent , since each component captures some aspects of the other component . For example , the position-independent component , PL , may capture position-specific dependencies between nucleotides separated by four basepairs , and such dependencies are part of the position-specific component , and vice versa for the periodic PN component . The above probabilistic model assigns a nucleosome formation score to each sequence of ( nucleosome-length ) 147 bp . We then use the scores of this model to compute the genome-wide distribution over nucleosome positions , taking into account steric hindrance constraints between neighboring nucleosomes . To this end , we take the partition function to be the space of all legal configurations of nucleosomes on a sequence S , where a legal configuration specifies a set of 147 bp nucleosomes and a start position for each of these nucleosomes on S , such that no two nucleosomes overlap . A legal configuration thus respects a simple approximation of the detailed linker length-dependent steric hindrance constraints between nucleosomes . We score a sequence S for its apparent nucleosome binding affinity using the above formula for Score ( S ) . For each sequence S and legal configuration c with k nucleosomes positioned at c[1] , … , c[k] , we assign a statistical weight Wc[S] defined as:where τ represents an apparent nucleosome concentration , and β is an apparent inverse temperature parameter . Our default parameter settings are τ = 1 and β = 0 . 5 . In accord with the Boltzmann distribution and under the assumption of thermodynamic equilibrium , it follows that the probability of every configuration is then given by:where c′ goes over the space of all legal configurations C . A dynamic programming method [2] , [71] can efficiently compute the probability of placing a nucleosome that starts at each basepair in the genome . The underlying idea is that the probability of placing a nucleosome starting at a particular basepair i is equal to the sum of the statistical weights of all configurations in which a nucleosome starts at position i , divided by the sum of the statistical weights of all legal configurations . Both of these sums can be computed efficiently in three steps . The first is a forward step , in which we compute a set of variables F1 , …FN , where Fi represents the sum of the statistical weight of all legal configurations of the sub-sequence S1 , …Si , as follows: The second step is a reverse step , in which we compute a set of variables R1 , …RN , where Ri represents the statistical weight of all legal configurations of the sub-sequence Si , …SN , as follows: In the final step , we can directly compute the probability , P ( i ) , of placing a nucleosome that starts at each basepair i of S , where i≤N−146 , as follows: The probability that a basepair i in S is covered by any nucleosome , referred hereto after as the average nucleosome occupancy predicted by our model , is the sum of the probabilities of starting a nucleosome at any of the positions from i−146 to i , i . e . , . | The detailed positions of nucleosomes along genomes have critical roles in transcriptional regulation . Consequently , it is important to understand the principles that govern the organization of nucleosomes in vivo and the functional consequences of this organization . Here we report on progress in identifying the functional consequences of nucleosome organization , in understanding the way in which nucleosome organization is encoded in the DNA , and in linking the two , by suggesting that distinct transcriptional behaviors are encoded through the genome's intrinsic nucleosome organization . Our results thus provide insight on the broader question of understanding how transcriptional programs are encoded in the DNA sequence . These new insights were enabled by individually sequencing ∼380 , 000 nucleosomes from yeast in their entirety . Using this map , we refine our previous model for predicting nucleosome positions and demonstrate that our new model predicts nucleosome organizations in yeast with high accuracy and that its nucleosome positioning signals are predictive across eukaryotes . We show that the yeast genome may utilize these nucleosome positioning signals to encode regions with both relatively open ( nucleosome-depleted ) chromatin organizations and relatively closed ( nucleosome-covered ) chromatin organizations and that this encoding can partly explain aspects of transcription factor binding , gene expression , transcriptional noise , and DNA replication . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"genetics",
"and",
"genomics/epigenetics",
"genetics",
"and",
"genomics/bioinformatics",
"computational",
"biology/genomics",
"computational",
"biology/transcriptional",
"regulation"
] | 2008 | Distinct Modes of Regulation by Chromatin Encoded through Nucleosome Positioning Signals |
Dermal hyperpigmentation or Fibromelanosis ( FM ) is one of the few examples of skin pigmentation phenotypes in the chicken , where most other pigmentation variants influence feather color and patterning . The Silkie chicken is the most widespread and well-studied breed displaying this phenotype . The presence of the dominant FM allele results in extensive pigmentation of the dermal layer of skin and the majority of internal connective tissue . Here we identify the causal mutation of FM as an inverted duplication and junction of two genomic regions separated by more than 400 kb in wild-type individuals . One of these duplicated regions contains endothelin 3 ( EDN3 ) , a gene with a known role in promoting melanoblast proliferation . We show that EDN3 expression is increased in the developing Silkie embryo during the time in which melanoblasts are migrating , and elevated levels of expression are maintained in the adult skin tissue . We have examined four different chicken breeds from both Asia and Europe displaying dermal hyperpigmentation and conclude that the same structural variant underlies this phenotype in all chicken breeds . This complex genomic rearrangement causing a specific monogenic trait in the chicken illustrates how novel mutations with major phenotypic effects have been reused during breed formation in domestic animals .
Fibromelanosis ( FM ) is characterized by intense pigmentation of the dermal layer of skin across the entire body , which results in a dark blue appearance when viewed through the clear epidermis ( Figure 1 ) . The term Fibromelanosis was coined to denote the association of pigmentation with internal connective tissue [1] and can be readily seen in the trachea , pericardium , blood vessels , sheaths of muscles and nerves , gonads , mesenteries of the gut , and periosteum of bone [1]–[5] . The Silkie breed is the most widespread and well-studied breed displaying FM . Silkie chickens present a unique collection of interesting phenotypes; the namesake Silkie feathering trait , blue earlobes , polydactyly , walnut comb , crest , beard , vulture hock , and feathered legs , all of which may have contributed to the human fascination and subsequent global distribution of this breed seen today [1] , [4] , [6] . Silkies are very popular with exhibition and backyard poultry breeders in the USA and Europe and are also available in many Asian grocery stores within the USA . The Fibromelanosis ( FM ) or dermal hyperpigmentation phenotype of the Silkie chicken is one of only a few skin pigmentation mutants in the chicken and has been a subject of cultural importance and scientific interest for centuries . This breed is thought to originate in China and closely resembles fowl described in 16th century Chinese texts on traditional medicine , although the exact origin of the Silkie breed is unknown [7] . Marco Polo's description of chickens that “have hair like cats , are black , and lay the best of eggs” in 1298 or Aldrovandi's account of “wool-bearing” chickens with white feathers and five toes in 1600 may refer to the Silkie [8] , [9] , and there are numerous vague references to chickens with similar features to the Silkie in much older Chinese texts . Indeed , folklore describes the Silkie chicken as receiving healing properties after eating pills of immortality created by the deity Lu Dongbin at Tiger-Nose peak . Although the most common globally , the Silkie chicken is not the only breed with dermal hyperpigmentation . Other FM strains are found in India , Indonesia , Japan , Korea , Sweden and Vietnam with varying degrees of overall phenotypic similarity to the Silkie ( personal observations ) . One of the earliest studies of the Silkie dermal hyperpigmentation phenotype was by Bateson and Punnett in 1911 [10] which together with the work of Dunn and Jull [11] showed the autosomal dominant nature of the *FM allele in conjunction with the sex-linked Inhibitor of Dermal Melanin ( ID ) locus acting upstream of *FM; here we have adopted the currently recommended nomenclature system for the chicken where FM refers to the Fibromelanosis locus and *FM and *N refer to the dominant Fibromelanosis inducing allele and the recessive normally pigmented wild-type allele respectively . All birds expressing the FM phenotype are homozygous wild-type *N at the ID locus , or hemizygous in the case of females as ID is located on the Z chromosome . Melanocytes are derived from neural crest cells ( NCCs ) , a multi-potent population of cells emigrating from the dorsal neural tube . In the Silkie embryo melanoblasts , the NCC-derived precursors of pigment-producing melanocytes , enter a migratory pathway that is normally reserved for NCCs of the neuronal and glial cell lineages [12] . This results in colonization of target tissues which normally are not exposed to melanoblasts and would otherwise remain unpigmented . In addition to this abnormal choice of migratory pathway melanoblasts also accumulate in large numbers throughout the body plan of the Silkie embryo [2] . This suggests a two-fold molecular mechanism of ectopic migration and continued proliferation , which may correspond to the classically described ID and FM loci , respectively . Previous embryo grafting experiments have clearly shown the proliferative effect of the Silkie tissue environment on melanocyte behavior but have been unable to determine if Silkie melanocytes possess inherent differences in migratory ability or if this also is a non-cell autonomous attribute of the Silkie [13] . Here we show that FM is caused by an inverted duplication of two genomic regions , each greater than 100 kb , located on Gallus gallus autosome 20 , which results in increased expression of endothelin 3 ( EDN3 ) .
Using a backcross mapping population we previously identified a 2 . 8 Mb region of chromosome 20 that was completely associated with the dermal hyperpigmentation phenotype corresponding to the FM locus [6] . Using this same mapping population and additional markers we have now refined this region to 483 kb ( 10 , 518 , 217–11 , 000 , 943 bp ) of chromosome 20 which is completely associated with FM in 270 backcross individuals; all genome coordinates are respective to the May 2006 ( WUGSC 2 . 1/galGal3 ) assembly [14] . Identity by descent analysis in a diverse panel of chicken breeds identified a 75 kb haplotype ( 10 , 717 , 600–10 , 792 , 608 bp ) within this 483 kb region which contained five SNPs observed to be heterozygous in all *FM samples ( Figure S1 ) . Two of these five SNPs ( rs16172722 and rs16172768 ) were fixed for the reference allele in the diverse breed panel wild-type individuals . The other three SNPs ( GGaluGA180596 , rs16172794 and rs16172818 ) were segregating for both alleles to various degrees in wild-type individuals . The observation of fixed heterozygosity in *FM individuals prompted an analysis of copy number variation using the 60K Chicken iSelect chip Log R ratio data from the diverse breed panel . The GenePattern implementation of the Circular Binary Segmentation ( CBS ) algorithm [15] , [16] identified a region with elevated Log R ratio levels in *FM individuals indicative of a duplication , although not all *FM individuals surpassed the significance threshold ( Table 1 ) . This region overlapped the five previously described heterozygous SNPs . This analysis also suggested a second putative duplication event at 11 . 1–11 . 4 Mb on chromosome 20 in *FM individuals . To further define the exact boundaries of the putative duplications we performed a group-wise analysis by subtracting the average Log R ratio of known *N individuals from the average of *FM individuals on a single SNP basis . This method revealed a clearer picture of the two putative duplicated regions in *FM individuals from 10 , 717 , 600–10 , 842 , 919 bp and 11 , 264 , 226–11 , 432 , 336 bp ( Figure 2 ) . Quantitative PCR ( qPCR ) analysis confirmed the duplication of both genomic regions in FM birds with an estimated copy number of approximately 1 . 5–2× that of wild-type individuals , indicating that some FM birds were likely heterozygous for a mutant allele composed of a 2× duplication ( Figure 3 ) . Although the second duplicated region lies outside the 483 kb region we had identified in the mapping population , the presence of both duplicated regions in all *FM individuals from the diverse breed panel suggested that both regions were involved in a genomic rearrangement and duplication event associated with the FM locus ( Figure 4A ) . We investigated the structural arrangement of the putative duplicated regions by PCR between outward facing primers at each end of both putative duplicated regions . We first tested for the presence of a tandem duplication using primers Dup1_5'xDup1_3' and Dup2_5'xDup2_3' , however no amplification was detected . After testing all possible combinations of these four primers successful amplification was detected only for Dup1_5'xDup2_5' and Dup1_3'xDup2_3' , suggesting that each duplicated region was joined to the other in an inverted orientation ( Figure 4B ) . Sequencing of these PCR products revealed the exact coordinates of the first duplicated region to be 10 , 717 , 294–10 , 846 , 232 bp and the coordinates of the second duplicated region to be 11 , 262 , 904–11 , 435 , 256 bp . In conjunction with the successful amplification of the two PCR products described above , amplicons were successfully generated across the wild-type boundaries of both duplicated regions in several known *FM homozygotes . A three primer diagnostic test for each duplicated region was developed that is capable of amplifying both a wild-type and mutant allele in the same reaction ( Table 2 and Figure S2 ) . All samples from the diverse breed panel and other populations known to carry *FM tested positive for both duplicated region junctions while no *N individuals were found to carry either of these rearrangements ( Table 3 ) . In order to differentiate *FM/*FM from *FM/*N individuals a qPCR genomic copy number assay must be used due to the retention of all four duplication boundary wild-type sequences in the mutant allele . Our data suggest three possible spatial arrangements of the duplicated regions , which are indistinguishable from each other via conventional PCR assays across duplication boundaries ( Figure 4B ) . Our data favors rearrangement scenario *FM_2 given the detection of a single recombination event between the duplicated regions in our backcross population . This recombinant individual was phenotypically wild-type and had inherited a Silkie chromosome beginning in the 417 kb single copy region between the two duplications and continuing through the second duplicated region until the end of the chromosome ( Figure 4C ) . This effectively eliminates rearrangement scenario *FM_1 as recombination between inverted copies of the intervening single copy region would result in chromosomal loss . Scenario *FM_3 also can be eliminated given that in this arrangement recombination between the single copy region of the depicted *N and *FM_3 alleles would result in the retention of duplicated copies of both regions , which was not the case in this recombinant individual as estimated by the genomic qPCR assay . Sequencing of ∼800 bp at each of the duplication boundaries , in their wild-type arrangement , revealed a high degree of sequence variation between and within four different populations of Silkie chickens known to be homozygous *FM ( Tables S1 , S2 , S3 , and S4 ) . Conversely , at the junction of the 5′ and 3′ ends of each duplicated region , respectively , there was no sequence variation between any of these same *FM samples ( Tables S5 and S6 ) . The absence of sequence variation at both duplication junction points suggests that this haplotype has been highly conserved since the mutation occurred , but that with increasing distance from the duplication junction points recombination has occurred with other alleles . The single recombinant individual observed in the mapping population supports this view . We decided to take advantage of massively parallel whole genome sequencing data to verify the duplicated regions and their orientation . We compared the read depth between FM and wild-type DNA pools , 15–20 birds in each pool , sequenced on average to 30× coverage . In the FM pool there was approximately 2-fold higher coverage strictly within the duplicated regions we had previously identified ( Figure 5A–5C ) . A large number of mate-pair sequencing reads were detected that confirmed the inverted arrangement of the duplicated regions; one set of approximately 400 mate-pairs mapped to the 5′ area of each duplicated region and a second similar number of mate-pairs mapped to the 3′ area of each duplicated region . No other duplications , deletions or rearrangements in this genomic region were detected in this analysis ( Figure 5D ) . There are several known coding elements within the first duplicated region including ATP5E ( ATP synthase epsilon subunit ) , TUBB1 ( tubulin , beta 1 ) , SLMO2 ( slowmo homolog 2 ) and EDN3 ( endothelin 3 ) . EDN3 has a known role in melanocyte regulation [17]–[19] and was an obvious candidate gene for further analysis . The second duplicated region does not contain any known coding or regulatory elements but displays isolated pockets of elevated conservation scores across seven vertebrate species as calculated using phastCons [20] in the UCSC Genome Browser ( http://genome . ucsc . edu ) ( Figure 4A ) . The entire coding sequence of EDN3 lies approximately in the center of the first duplicated region . In the FM ( Silkie breed ) embryo EDN3 is significantly ( p<0 . 05 ) increased in expression during embryonic stages when melanoblasts are migrating and beginning to differentiate into melanocytes ( Figure 6A ) . The magnitude of increased expression of EDN3 in FM embryos appears to increase with developmental age and reaches a remarkably high level of differential expression ( about 10-fold ) in adult skin tissue ( Figure 6B ) . The expression of two other genes , SLMO2 and TUBB1 , located within the first duplicated region are also significantly increased in expression in both skin and muscle tissue from adult FM chickens ( Figure 6B ) . The expression of DDX27 ( DEAD ( Asp-Glu-Ala-Asp ) box polypeptide 27 ) , located outside but in close proximity to the first duplication is significantly differentially expressed in FM skin and muscle ( up and down , respectively ) , but the magnitude of the difference is minimal when compared to the genes within the duplication; EDN3 , SLMO2 and TUBB1 ( Figure 6B ) . We also examined the expression of two EDN3 receptors in adult chicken skin and muscle tissue; EDNRB ( endothelin receptor B ) which in the chicken is confined to non-melanocyte derivatives of the neural crest migrating through the dorsoventral pathway [21] and EDNRB2 ( endothelin receptor B subtype 2 ) which is expressed primarily by melanocytes and is involved in cell migration and differentiation [22] , [23] . The expression of TYRP2 ( tyrosinase-related protein 2 ) , which catalyzes the conversion of L-Dopa into DHICA [24] , was also assayed as an indication of the level of eumelanin biosynthesis occurring . In wild-type muscle tissue EDNRB2 and TYRP2 expression was below the detection level of the qPCR assay , while no significant difference in EDNRB expression was detected between FM and wild-type muscle or skin tissue ( Figure 6C ) . The expression of EDNRB2 and TYRP2 was significantly higher in FM skin tissue as compared to wild-type skin ( >88 and >2500 fold , respectively ) ( Figure 6C ) . Most notably , the expression of EDNRB2 and TYRP2 was 26-fold and 5-fold higher , respectively , in wild-type skin tissue as compared to FM muscle tissue ( Figure S3 ) . This reflects the absence of pigment producing melanocytes in FM muscle even though EDN3 expression is upregulated in both FM skin and muscle tissue . Note that while wild-type skin dermis is unpigmented , there are active melanocytes within the feather follicle . The exonic sequence of the EDN3 transcript was sequenced from genomic DNA ( data not shown ) , including across a gap in the current genome assembly that corresponds to a portion of the second coding exon . In these samples no non-synonymous sequence variants were detected between FM and wild-type individuals .
We have demonstrated that the FM phenotype shows complete concordance with a complex structural variant involving the duplication of two genomic regions , each larger than 100 kb and separated by 417 kb on wild-type chromosomes . The precise definition of the structural variant was facilitated by our use of a modified Log R ratio analysis of SNP data generated using the Illumina 60K Chicken iSelect chip . By comparing *FM and *N individuals in a group-wise manner , we were able to predict the duplication boundaries to an accuracy of 300–3 , 300 bp from the actual breakpoints , limited primarily by the spacing of markers on the chip . This novel structural variant was not found in samples of Red Junglefowl ( Gallus gallus ) or in a diverse panel of domestic chickens , representing 21 different breeds all of which do not display the FM character . However , the same complex rearrangement associated with FM was found in four different breeds representing four countries and two continents , Silkie from China , Ayam Cemani from Indonesia , Black H'Mong from Vietnam and Svarthöna from Sweden . The result is consistent with that of a single mutation event underlying FM in all breeds . Sequence analysis of ∼800 bp PCR fragments centered on all four duplication boundaries and both duplication junctions revealed substantial sequence polymorphisms between and also within *FM populations at the wild-type duplication boundaries , but complete sequence conservation at both duplication junctions . This supports the causal nature of the inverted duplications for the FM phenotype and suggests that the mutant allele has experienced recombination with wild-type alleles with increasing distance from the inverted duplication junction points . We have demonstrated that the structural variant underlying FM involves the joining of the 5′ ends of the two duplicated regions as well as the joining of the 3′ ends of the two duplicated regions , while still allowing for the successful PCR amplification of all four wild-type duplication boundary sequences . We searched for homologous sequences at the duplication junction boundaries in order to infer the mechanism of this genomic rearrangement event . We detected only a single base pair of homology at the immediate junction of the 3′ ends of both duplicated regions while no homology was detected at the junction of the 5′ ends of both duplicated regions ( Figure 7 ) . This lack of sequence homology suggests a mechanism such as non-homologous end joining ( NHEJ ) via double-strand break repair . However the complex nature of this rearrangement with two distinct duplicated regions joined in an inverted fashion with no overall gain or loss of sequence at the junction points is difficult to reconcile with the known mechanisms of NHEJ in which sequence is often added or deleted at the breakpoint [25] , [26] . Alternative mechanisms , which rely on a small amount of sequence homology , are fork stalling and template switching ( FoSTeS ) and micro-homology mediated break induced replication ( MMBIR ) [27] , [28] . The FoSTeS/MMBIR DNA replication based mechanisms have been proposed for many complex rearrangements similar to the one we have identified as causing FM but typically relies on 2–5 bp of micro-homology at the junction point , although examples relying on a single base pair have been described [27] . No previously annotated segmental duplications were found in proximity to any of the four duplication boundaries [29] . We cannot exclude the possibility of additional genetic elements being involved in the formation and current structure of this genomic rearrangement but the analysis of massively parallel sequencing data from the FM DNA pool suggests that we have identified all major aspects of this genomic rearrangement ( Figure 5 ) . A characteristic feature of duplicated sequences is that they are prone to copy number variation due to unequal crossing-over . This is well illustrated by the dominant white locus in pigs that shows extensive haplotype diversity in breeds with the dominant white color based on the presence of 1–3 copies of a 450 kb duplication encompassing the entire KIT gene [30] . The FM rearrangement appears to be stable , unequal crossing-over is probably suppressed by the presence of an inverted copy of Duplication 2 located between the two copies of Duplication 1 ( Figure 4 ) . Our whole genome resequencing of pooled samples and qPCR analysis of individual samples did not indicate the presence of FM chromosomes with more than two copies of the duplicated sequences ( Figure 3 and Figure 5 ) . We have shown that the structural variant underlying FM is associated with an increased expression of endothelin 3 ( EDN3 ) , located completely within the first duplicated region ( Figure 4A ) . EDN3 has been shown to have a mitogenic effect on melanocytes in vitro [17]–[19] , which is similar to the proliferative effect extracts from Silkie embryos have on NCCs [31] . Furthermore , a transgenic mouse model of ectopic EDN3 expression results in dermal hyperpigmentation [32] , [33] that is strikingly similar to that of the Silkie chicken . In chicken wild-type embryos , NCCs specified as melanoblasts begin to delaminate from the dorsal neural tube at stage 18 and migrate almost exclusively within the dorsolateral pathway . In the Silkie embryo the timing of melanoblast migration from the neural crest is slightly delayed but perhaps more interesting is the transit of melanoblasts through both the dorsoventral and dorsolateral migratory pathways . In addition to this abnormal choice of migratory pathway , melanoblasts appear to continue to proliferate in the Silkie embryo giving rise to a large number of pigmented melanocytes in ectopic positions of the embryo already at embryonic day 18 [2] . We show that in the developing embryo , when melanoblasts are in the process of migrating to their target tissues , there is already increased expression of EDN3 in the Silkie embryo that continues to increase in the magnitude of differential expression throughout development . It is quite remarkable that in the adult skin tissue of Silkie chickens EDN3 expression is increased about 10-fold ( Figure 6B ) . This suggests a possible role for EDN3 in melanoblast maintenance later in life given that continued expression of EDN3 in a transgenic mouse model was necessary for maintenance of the dermal hyperpigmentation phenotype after birth [33] . The expression of EDN3 was significantly increased in both FM skin and muscle tissue , but the expression of genes indicative of the presence of active melanocytes was lower in FM muscle than in wild-type skin tissue . This suggests that while EDN3 may be upregulated in both FM skin and muscle , few melanocytes in the muscle are present and/or able to respond and thus FM muscle remains largely unpigmented . The FM locus constitutes both a cis-acting and trans-acting eQTL . For instance , we observed a staggering >2500-fold upregulated expression of TYRP2 , a key gene in the melanin biosynthesis pathway , in FM skin compared with wild-type skin ( Figure 6C ) . This is partially explained by the dramatic increase in melanocyte numbers in FM skin and partially by downstream effects of increased EDN3 signaling on these melanocytes . Thus , a whole transcriptome comparison of FM and wild-type skin is expected to reveal significant differential expression of hundreds , if not thousands , of transcripts and it would be challenging to reveal the causative nature of these alterations without a genetic analysis like the present study . Previous experiments documenting the migration pattern of melanoblasts in the Silkie embryo utilized the White Leghorn and Light Brown Leghorn as the control breed [2] , [34] . Unfortunately , both of these control breeds possess opposing alleles at both the FM and ID ( Inhibitor of Dermal Melanin ) loci compared to the Silkie . This complicates the reconciliation of increased EDN3 signaling with the abnormal migration pattern and increase of melanoblasts previously documented in the Silkie , as these developmental differences could be due to either ID or FM . We have performed in situ hybridization analysis of EDN3 expression ( data not shown ) and did not observe any difference in spatial expression when compared to a FM*N control breed carrying the same ID allele ( *N ) as the Silkie . The elevated level of EDN3 expression detected via qPCR , but without any detectable difference in EDN3 spatial expression pattern from the in situs , is consistent with the perspective of ID controlling pathways of melanoblast migration and FM being responsible for melanoblast proliferation and maintenance . Even in the presence of *FM and the associated increased expression of EDN3 , individuals carrying the ID*ID allele lack visible pigmentation except for a small number of isolated patches most easily visible on the shank . In FM*N ID*N chickens pigmentation of the skin is only seen in the dermis of the shank , which could be explained by extensive migration of melanoblasts throughout the body plan as a result of ID*N , but lack of melanoblast maintenance and proliferation due to FM*N except for in the shank where expression of EDN3 may be under different regulatory influences . Further work directed at identifying the ID causal mutation located on the Z chromosome is necessary to further explore this hypothesis . The FM structural rearrangement results in the duplication of a key regulator of melanoblast and melanocyte proliferation and maintenance , EDN3 . Loss of function mutations in EDN3 are associated with pigmentation , auditory and gut innervation disorders in humans ( Waardenburg Syndrome and Hirschsprung Disease ) [35] and in the mouse ( lethal spotting ) [36] suggesting a crucial role for EDN3 in regulating NCC-derived lineages . The long standing view on the origin of melanocytes in all tissues except the retinal pigmented epithelium is that they are derived from NCCs which are fate specified as melanoblasts and exploit the dorsolateral migratory pathway [37] . However , recent evidence has shown that in the mouse and chicken melanocytes derived from Schwann cell precursors , which originally migrated through the dorsoventral pathway , make an equivalent contribution to mature melanocytes as do melanoblasts which have migrated through the canonical dorsolateral pathway [38] . In light of how EDN3 promotes the reversion of melanocytes and glia to a bipotent precursor [17] , [18] , [39] it is interesting to speculate on how increased EDN3 signaling may affect the relative contribution these two sources of melanocytes make in FM chickens . FM chickens do not appear to suffer from any gross neurological defects suggesting that increased EDN3 signaling does not have a major impact on the development of the completely neural crest-derived peripheral nervous system in this strain . There are several different varieties of the Silkie chicken breed as distinguished by feather color ( white , buff , red , blue , grey , partridge , and black ) , which indicates that FM does not affect feather pigmentation . Pigmentation of the feather is controlled by the activity of epidermal melanocytes and therefore this suggests non-overlapping mechanisms regulating the development of dermal and epidermal melanocytes . In the mouse ectopic expression of EDN3 results in hyperpigmentation of the dermis by melanocytes that are more similar to non-cutaneous melanocytes of the eye , inner ear , and harderian gland than they are to epidermal melanocytes in regards to responsiveness to EDN3 , hepatocyte growth factor ( HGF ) , and tyrosine kinase ( KIT ) ligand signaling [32] . These non-cutaneous like dermal melanocytes are incapable of contributing to epidermal hair follicle pigmentation further highlighting the functional differences between these two melanocyte populations [40] . This suggests a strong role for EDN3 in regulating non-cutaneous melanocytes and presents the Silkie chicken as a readily accessible naturally occurring mutant in which to further study this newly described population of melanocytes . These results provide an example of how a complex structural variation involving an inversion and duplication of two distinct genomic regions each greater than 100 kb can drive a monogenic trait in a domestic animal species . The specific cause of the increased EDN3 expression remains unclear given the complex nature and size of this structural variant . It is unlikely to be a simple dosage effect given the dominant nature of this mutation as well as our observation of a 10-fold increase in expression of EDN3 in adult Silkie chicken skin tissue . We propose that the most likely explanation is an altered constellation of long-range cis-regulatory elements affecting EDN3 expression , disruption of silencer activities and/or recruitment of new enhancer ( s ) via the physical reorganization of the locus . The observation of increased expression of two other genes within the first duplicated region is consistent with this view of a large-scale perturbation of transcriptional regulation caused by this genomic rearrangement . The increased expression of SLMO2 , TUBB1 , and possibly other genes within the duplicated region ( s ) raises the possibility that multiple genes within this locus are contributing to the dermal hyperpigmentation phenotype , or possibly other more subtle phenotypes not previously associated with FM . We cannot exclude this possibility , but based on the similar dermal pigmentation phenotype seen when EDN3 is ectopically expressed in the mouse [32] , [33] we would suggest that EDN3 upregulation is the primary driver of dermal hyperpigmentation in FM chickens . To the best of our knowledge there are no other phenotypes consistently observed in all FM breeds , however other researchers in possession of populations segregating at the FM locus should explore the possible effect of upregulated expression of SLMO2 and TUBB1 on other types of phenotypic variation . It is possible that one of the duplicated copies of EDN3 has accumulated a gain of function mutation in a regulatory element within the duplicated region , but this appears unlikely since this complex rearrangement was most likely selected by humans due to a striking phenotypic effect associated directly with the inversion and duplication mutation itself . Several examples of copy number variation linked to single gene traits have previously been described in the chicken [41] , [42] and together with the structural variant we have now described as causing FM suggests that structural changes have contributed significantly to the evolution of phenotypic diversity in domestic chickens . Recently the amount of structural variation present in the chicken genome was surveyed using massively parallel sequencing techniques [43] , [44] , however FM breeds were not included in these analyses . As the technology available to assay genomic structural variation improves through longer sequencing read lengths and the development of high throughput optical mapping [45] there will no doubt be many more cases in which genomic rearrangements are found in association with both monogenic and polygenic traits in many species . As these results have shown , this type of large-scale structural rearrangement of the genome has the capability to alter gene expression regulatory mechanisms at an extended range . This suggests that while this type of variation is inherited as a single locus it may have multiple downstream effects through perturbation of several different genetic pathways , possibly contributing to more subtle or complex phenotypes .
Fine mapping of FM was performed using a backcross population of chickens as previously described [6] . Silkie chicken breed DNA samples were obtained from several different populations from the USA ( North Carolina and Wisconsin ) , Sweden , Vietnam , and China . These Silkie sub-lines displayed subtle differences in minor breed characteristics reflective of local preferences but overall were representative of the Silkie breed . Samples obtained from other breeds known to display Fibromelanosis included a population of Ayam Cemani , Black H'Mong , and Svarthöna chickens . The Ayam Cemani is an Indonesian breed although the samples used in this study were collected in the USA . The Black H'Mong chicken breed is from the Ha Giang region of Vietnam . The Svarthöna ( full name Bohuslän-Dals Svarthöna ) is a Swedish breed believed to have been imported from Norway in the early 1900s . A panel of diverse chicken breeds known to be FM *N was collected from across North Carolina and Vietnam . A custom GoldenGate BeadXpress panel ( Illumina ) containing 36 SNPs spanning the 2 . 8 Mb region we previously reported as being highly associated with Fibromelanosis [6] was used for fine mapping in the backcross population . The 60K Chicken iSelect chip [46] ( Illumina ) was used to screen the founders of this backcross population in order to identify completely informative SNPs and was also used to genotype the diverse breed panel and USA Silkie and Ayam Cemani samples for identity by descent haplotype analysis . The GenomeStudio V2010 . 3 software package ( Illumina ) was used to obtain normalized total signal intensity , the Log R ratio , for all SNPs on the 60K Chicken iSelect chip according to the manufacturer's instructions and Peiffer et al . [47] . A continuous region of SNPs having a Log R ratio above/below zero indicates a possible duplication/deletion event . To identify such segments , the exported Log R ratio data from all individuals was analyzed using the Circular Binary Segmentation algorithm implemented in GenePattern [15] , [16] . Default parameters were used in the segmentation analysis , i . e . number of permutations used for p-value computation was 10 , 000 , significance level ( alpha ) for the test to accept change-points was 0 . 01 and the seed for the random number generator was 12345678 . The group-wise Log R ratio was calculated by subtracting the average Log R ratio from all FM *N individuals from the average of all FM *FM individuals . TaqMan primer and probe sets were designed using Primer3Plus [48] according to standard parameters . Target probes were 5′ labeled with 6-FAM and 3′ labeled with the minor groove binder ( MGB ) non-fluorescent quencher , the reference probe located in an exon of SOX5 was 5′ labeled with VIC and 3′ labeled with TAMRA ( ABI ) . See Table S7 for primer sequences . Reactions were performed using Gene Expression MasterMix ( ABI ) containing 10 ng of genomic DNA , 800 nM of each primer , 250 nM probe in a total volume of 20 µl . Quality control of all primer/probe sets was evaluated in both simplex and duplex with the control primer/probe set using a 7-point standard curve with a 5× dilution factor and criteria previously described by Ishii et al . [49]: slope −3 . 1035 to −3 . 7762 ( PCR Efficiency = 92–105% ) , R>0 . 995 , and y-intercept values differ by less than 1 . Genomic qPCR assays were performed using the diverse breed panel , USA Silkie , Chinese Silkie , Svarthöna and Ayam Cemani breed samples as well as several known heterozygotes from the mapping population . Data was analyzed using CopyCaller software ( ABI ) , which uses a ΔΔCt method to first normalize the target Ct value to the reference Ct value within sample , and subsequently normalize all samples to a known calibrator sample . The calibrator sample in this experiment was the wild-type founder of the mapping population , with an expected diploid copy number of 2 for each TaqMan qPCR amplicon . Error bars represent the minimum and maximum estimated copy number as calculated from technical replicates of each sample . DNA from 15 Silkie and 20 commercial Broiler chickens were pooled , resulting in one pool for each breed . SOLiD mate-pair sequencing libraries were generated from these DNA pools with a mean insert size of approximately 2 . 5 kb . The libraries were sequenced using a SOLiD v . 4 instrument ( Life Technologies , Carlsbad , U . S . A . ) according to the manufacturer's instructions , The reads were mapped to the chicken genome ( WUGSC 2 . 1/galGal3 ) reference assembly using the Lifescope software ( Life Technologies ) , resulting in average read depths of approximately 30× per library over the chicken genome . The mapping data were used to determine read depths in 1 kb windows for the two samples over the region of interest ( 10 . 218–11 . 935 Mb on chromosome 20 ) . The mapping distances between mate-pairs were used to detect structural variation in relation to the reference assembly . Outward facing primer sets from each of the four duplication boundary regions were used to determine the spatial arrangement and orientation of the two duplicated regions via PCR . This facilitated the design of primers suitable for sequencing each of the four duplication boundaries and the development of a three primer diagnostic test for the presence of each of the duplications . See Table S7 for primer sequences . The KAPA2G Robust HotStart PCR system ( Kapa Biosystems ) was used for all standard PCR , the specific parameters of the breakpoint diagnostic test are 1× KAPA2G GC Buffer , 0 . 2 mM dNTPs , 1 . 5 mM MgCl2 , 200 nM of each of the three primers , 0 . 8 U of KAPA2G Robust HotStart DNA Polymerase , and 50 ng of DNA in a total volume of 20 µl . A touchdown thermal cycling protocol was used for the diagnostic test of 95°C for 5 min , 16 cycles of 95°C , 68°C ( −1 . 0°C/cycle ) , and 72°C for 30 s each , followed by 24 cycles of 95°C , 52°C , and 72°C for 30 s each . Tissue was collected from Silkie ( *FM ) and New Hampshire ( *N ) breed embryos at the level of the wing bud at stages 22 , 26 , 28 , 30 , and 42 according to Hamburger and Hamilton guidelines [50] as well as from adult skin and muscle tissue . At least three biological replicates of each breed were collected at all time points . Tissue was homogenized in Mini-Beadbeater ( BioSpec ) tubes containing TriZol and 1 . 0 mm glass beads . RNA isolation was performed using the PureLink Micro-Midi kit with TRIzol ( Invitrogen ) and included an on column DNase treatment . GenBank accession numbers and associated primer sequences for all genes can be found in Table S8 . Reactions were performed using iScript One-Step RT-PCR Kit With SYBR Green ( Bio-Rad ) containing 50 ng of RNA and 600 nM of each primer in a total volume of 25 µl . Thermal cycling parameters were as described by the manufacturer . PCR products were subjected to melt curve analysis and sequencing to confirm amplification of the correct target . A standard curve was used to calculate the PCR efficiency of each target . At each time point biological replicates of n = 3 and technical replicates of n≥3 were used . Data was analyzed using qbasePLUS v2 . 1 ( Biogazelle ) software . A standard curve was used to estimate the PCR efficiency of each primer set and individual samples were normalized to GAPDH . For within tissue and across breed comparisons samples were normalized to *N expression levels and tested for significance using an unpaired t-test . For all across breed and across tissue comparisons all sample groups were normalized to the group with the lowest expression level within a single gene . For the across breed and across tissue comparisons a one-way ANOVA was used to calculate significance values . Error bars in all gene expression figures represent 95% confidence intervals . | The process of animal domestication has been a long and ongoing effort of the human race to cultivate beneficial traits in agriculturally productive or otherwise beneficial species . We are just now beginning to understand the effect this type of selection pressure has had on genetic variation and overall genome architecture using quickly advancing modern genetic and genomic technologies . Here we show how along the path of animal domestication a single large rearrangement involving a duplication and inversion of two distinct regions of the chicken genome occurred , likely disrupting long-range cis-regulatory elements of endothelin 3 ( EDN3 ) and resulting in a very extreme skin pigmentation phenotype . Dermal hyperpigmentation , or Fibromelanosis ( FM ) , is a defining characteristic of the Silkie chicken breed , which originates in China . Chickens very similar to the Silkie have been described in ancient Chinese texts on traditional medicine , illustrating how unique phenotypes in domesticated animals are incorporated into human culture and tradition that persists to this day . The presence of the same rearrangement in other FM chicken breeds found around the world highlights both the causality of this mutation as well as how humans serve to spread genetic variation linked to novel traits in domestic animals . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"animal",
"genetics",
"trait",
"locus",
"heredity",
"gene",
"expression",
"genetics",
"molecular",
"genetics",
"biology",
"genomics",
"structural",
"genomics",
"genetics",
"and",
"genomics"
] | 2011 | A Complex Genomic Rearrangement Involving the Endothelin 3 Locus Causes Dermal Hyperpigmentation in the Chicken |
Mouse papillomavirus type 1 ( MmuPV1 ) provides , for the first time , the opportunity to study infection and pathogenesis of papillomaviruses in the context of laboratory mice . In this report , we define the transcriptome of MmuPV1 genome present in papillomas arising in experimentally infected mice using a combination of RNA-seq , PacBio Iso-seq , 5’ RACE , 3’ RACE , primer-walking RT-PCR , RNase protection , Northern blot and in situ hybridization analyses . We demonstrate that the MmuPV1 genome is transcribed unidirectionally from five major promoters ( P ) or transcription start sites ( TSS ) and polyadenylates its transcripts at two major polyadenylation ( pA ) sites . We designate the P7503 , P360 and P859 as “early” promoters because they give rise to transcripts mostly utilizing the polyadenylation signal at nt 3844 and therefore can only encode early genes , and P7107 and P533 as “late” promoters because they give rise to transcripts utilizing polyadenylation signals at either nt 3844 or nt 7047 , the latter being able to encode late , capsid proteins . MmuPV1 genome contains five splice donor sites and three acceptor sites that produce thirty-six RNA isoforms deduced to express seven predicted early gene products ( E6 , E7 , E1 , E1^M1 , E1^M2 , E2 and E8^E2 ) and three predicted late gene products ( E1^E4 , L2 and L1 ) . The majority of the viral early transcripts are spliced once from nt 757 to 3139 , while viral late transcripts , which are predicted to encode L1 , are spliced twice , first from nt 7243 to either nt 3139 ( P7107 ) or nt 757 to 3139 ( P533 ) and second from nt 3431 to nt 5372 . Thirteen of these viral transcripts were detectable by Northern blot analysis , with the P533-derived late E1^E4 transcripts being the most abundant . The late transcripts could be detected in highly differentiated keratinocytes of MmuPV1-infected tissues as early as ten days after MmuPV1 inoculation and correlated with detection of L1 protein and viral DNA amplification . In mature warts , detection of L1 was also found in more poorly differentiated cells , as previously reported . Subclinical infections were also observed . The comprehensive transcription map of MmuPV1 generated in this study provides further evidence that MmuPV1 is similar to high-risk cutaneous beta human papillomaviruses . The knowledge revealed will facilitate the use of MmuPV1 as an animal virus model for understanding of human papillomavirus gene expression , pathogenesis and immunology .
Human papillomaviruses are a group of small , non-enveloped , epitheliotropic DNA tumor viruses whose infection can result in benign lesions ( called warts or papillomas ) and in some cases cause malignancies . Certain genotypes of HPVs , such as HPV-16 , HPV-18 , and HPV-31 , that infect mucosal epithelia , have been recognized as causative agents of anogenital cancers that include cervical and anal cancers , as well as a growing subset of head and neck cancers , particularly those arising in the oropharynx [1 , 2] . Papillomaviruses are species-specific . Although papillomavirus infection models in large animal species such as rabbits , dogs and cows have been used to study the molecular biology and pathogenesis of papillomavirus infections , a laboratory mouse model would greatly facilitate the study of papillomavirus-associated warts and cancers . The recent identification of the murine papillomavirus ( MmuPV1 ) that can infect laboratory strains of mice now provides us with such a tractable laboratory animal-based infection model system [3–6] . The MmuPV1 circular , double stranded DNA genome is 7510-bp in length and encodes at least seven translational open reading frames ( ORFs ) , designated E1 , E2 , E4 , E6 , E7 , L1 and L2 based upon their conserved position within the viral genome and length comparable to ORFs of other papillomaviruses [7 , 8] . To date , a transcription map of MmuPV1 has not been described . Such a map would greatly facilitate understanding the MmuPV1genome structure and viral gene expression capacities and thereby information on nature of viral factors that contribute to papillomavirus infection and associated pathogenesis . In this report , we describe a comprehensive map of MmuPV1 transcripts based upon a multi-pronged analysis of viral mRNAs isolated from tumor tissues derived from MmuPV1 infected mice . The viral transcription start sites ( TSS ) were mapped by 5’ rapid amplification of cDNA ends ( 5’-RACE ) [9 , 10] in combination with PacBio Iso-seq and confirmed by TA cloning and Sanger sequencing . The polyadenylation cleavage sites of viral early and late transcripts were mapped by 3’-RACE [9 , 10] . Viral genome expression and RNA splicing was profiled by RNA-seq and primer-walking RT-PCR [9–11] . Accordingly , we assigned the coding regions of multiple potential viral gene products , E1^E4 , E1^M1 , E1^M2 , E8^E2 , L2 and L1 based upon the deduced structure of mRNA transcripts , and performed in situ hybridization studies in which we could detect a subset of these viral early and late transcripts in MmuPV1 infection-derived wart tissues by RNA-ISH in correlation with the presence of viral proteins and viral genomic DNA .
Athymic FoxN1nu/nu mice were infected with MmuPV1 as follows: inner ( right ) ear , muzzle , and three spots on the tail following scarification ( Fig 1A ) . Resulting warts were harvested six months following infection and total RNA was isolated . Papillomatosis was confirmed by histopathological analysis ( S1A Fig ) . Papillomas exhibited fibrillary projections accompanied with hyperkeratosis and were exophytic in morphology . Koilocytes , considered as hallmarks of papillomavirus infection , were seen throughout the papillomas . Productive phase of viral life cycle was confirmed by L1 positivity by immunofluorescence and detection of amplified viral DNA by fluorescence in situ hybridization . L1 and amplified viral DNA were detected throughout the papilloma including the terminally differentiating epithelia ( S1A Fig ) . To elucidate the presence and relative abundance of RNA transcripts arising from the MmuPV1 genome from these wart tissues ( S1A Fig ) , ribosomal RNA-depleted total RNA isolated from each wart tissue at three anatomical sites of three infected animals was analyzed by RNA-seq . In addition , the uninfected ear from the same animal was used as an uninfected control . Approximately 100 million paired-end reads with high quality were obtained from each tissue sample ( Table 1 ) . By mapping the RNA-seq raw reads from each lesion to the newly arranged linear MmuPV1 genome starting from nt 7088 and ending at nt 7087 using RNA sequence aligner TopHat [12] , we obtained ~0 . 4–3 . 3 million viral reads for each wart sample ( GEO Accession No . GSE104118 ) , varying among lesions and animals , which accounts for ~0 . 4%-3 . 4% of total RNA reads obtained from each sample and with the muzzle tissues from animal #2 containing the most RNA reads ( Table 1 ) . Surprisingly , we also obtained many MmuPV1 reads from the uninfected ear tissues ( control ears ) in all three wart-bearing animals ( Table 1 ) , with the animal #1 uninfected ear ( left ear ) displaying the viral reads similar to that of the infected right ear while appearing normal by visual scoring . MmuPV1 L1 protein and DNA were detected in this tissue confirming the presence of subclinical infections in these control ear tissues ( S1B Fig ) . By uploading these uniquely mapped viral RNA reads obtained from individual samples to the Integrative Genomics Viewer ( IGV ) program to visualize reads coverage profile along with the MmuPV1 genome , we found three major coverage peaks , one in the E7 region , one in the E4 region and one between E4 and L2 , make the last two peaks as a V shape ( Fig 1B ) among all wart-tissues and sub-clinically infected ear tissues obtained from the MmuPV1 wart-bearing animals . The V shape was attributed to ( 1 ) fewer uniquely mapped viral RNA reads , but more host-viral chimeric reads being excluded from the mapping in this region , and ( 2 ) RNA splicing by using a 5’ splice site at nt 3431 ( see more detailed description later in this report ) . We also saw a small drop of viral RNA reads within the E4 ORF in particular in the control animal ears and this might be a result from increased E1^E4 splicing in proportion in the control animal ears when compared with the tumor ear RNA . By analyzing the orientation of the uniquely mapped viral RNA reads from the muzzle tissues obtained from mouse #2 which appeared the highest viral reads coverage , we conclude that the vast majority of the viral reads were of the sense strand ( 98 . 6% ) . Antisense-specific reads were of low abundance ( ~1 . 4% of all viral reads ) , although a few peaks at various points along with the viral genome were evident ( Fig 1C ) . We view these antisense reads being background noise from cDNA library construction , sequencing errors or mapping artifacts . A step-wise zoom-in view further showed that the viral sense transcripts derived from the LCR ( long control region ) , E6 , E1 , E2 , L2 and L1 regions were less abundant than the reads from the E7 and E4 regions , with the L1 reads a little more than the others ( Fig 1D ) . Considering majority of eukaryotic RNA has a 5’-end cap structure , the current RNA-seq protocol lacks a de-capping step and fragments RNA into 200–350 nt pieces for adaptor ligation and amplification . These methods create an unfavorable bias for adaptor ligation to the RNA 5’ end and deletion of the RNA 5’ end reads smaller than 200 nts , making the RNA-seq unattainable for mapping the RNA start site [13] . To map the transcription start sites ( TSS ) of MmuPV1 transcripts , 5′ RACE analyses were performed on total RNA isolated from MmuPV1-infected ear lesions using virus-specific antisense primer Pr7237 , Pr352 , Pr518 , Pr687 , Pr738 , Pr1123 , Pr3299 , or Pr5452 ( Fig 2A , S2 Fig and S1 Table ) . The 5′ RACE products were visualized either by a DNA Bioanalyzer ( Fig 2B ) or by agarose gel electrophoresis ( Fig 2E and S2 Fig ) . The 5’ RACE products derived from Pr3299 and Pr5452 were also subjected to long read single-molecular , real-time sequencing using PacBio ( Pacific Biosciences ) Iso-seq technology ( Fig 2C and 2D , S3A Fig ) , for detection of existing full-length transcripts and overcoming the RNA-seq unfavorable bias on the fragmented RNA 5’ end . In addition , all 5’ RACE products were gel-purified , cloned and sequenced ( Fig 2E , S2 and S3B Figs ) . From PacBio Iso-eq sequencing , approximate five thousands of the full-length viral transcripts were detected from each primer-derived 5’ RACE products ( Table 2 ) . By uploading these Iso-seq-identified individual viral transcripts to the IGV program to visualize their coverage profile along with MmuPV1 genome , we found that the Pr3299-derived 5’ RACE products were mainly transcribed from two TSS , one at nt 533 ( ~50% ) and the other at nt 7503 ( ~20% ) in the viral genome , although other scattered TSS positions were also identified immediately downstream of the nt 533 position ( Fig 2C and 2D , Table 2 ) . In contrast , the Pr5452-derived 5’ RACE products were predominantly transcribed from the nt 7107 position ( >55% ) and secondarily from the nt 533 position ( >30% ) on the viral genome ( Fig 2C and 2D , Table 2 ) . Similar to the Pr3299-derived products , the Pr5452-derived 5’ RACE products displayed scattered TSS positions ( mostly from the nt 576 to nt 607 ) immediately downstream of the nt 533 position ( Fig 2C ) . Both 5’ RACE products gave the same TSS products initiating at nt 859 ( Fig 2C and 2D ) , while the TSS at the nt 360 was most associated with Pr3299-derived products ( Fig 2C and 2D , S3A Fig and Table 2 ) . The TSS of viral transcripts was also assessed by TA cloning and Sanger sequencing of individual primer-derived 5’ RACE products . As shown in Fig 2E , S2 Fig and S2 Table , one 5’ RACE product obtained from the Pr7237 ( lane 1 ) was mapped primarily to the nt 7107 ( 6/9 colonies ) . Three RACE products were detected from the Pr352 ( lane 2 ) and they were defined to have TSS at nt 260 ( 7/14 colonies ) , nt 7503 ( 6/10 colonies ) and nt 7107 ( direct sequencing of PCR products ) . Pr518 exhibited a similar 5’ RACE product profile , 166-bp larger than that of Pr352 . Direct sequencing of four Pr518 5’ RACE products ( lane 3 ) identified TSS at nt 360 , 260 , 7503 and 7107 . The Pr687 and Pr738 also showed a similar 5’ RACE product profile , with the two Pr738-derived 5’ RACE products being 51-nts larger than that of Pr692 ( compare lane 4 to lane 5 ) . Direct sequencing showed the main product from Pr687 had a TSS at nt 533 ( lane 4 ) . Cloning and sequencing of the Pr738-derived products 1 or 2 indicated that product 1 had a TSS at nt 576/579/589 ( 7/20 colonies ) and product 2 had a TSS at nt 533 ( 6/21 colonies ) ( lane 5 ) . The Pr3299 generated two major spliced 5’ RACE products containing splice junctions at nts 757/3139 or nts 7243/3139 ( lane 6 ) , with TSS for product 1 mainly placed at nt 7107 ( 12/18 colonies ) and for product 2 at nts 533 or 576 ( 6/17 colonies ) . Pr5452 ( lane 7 ) had a similar 5’ RACE profile to that of Pr3299 , but gave three faster migrating faint bands . TA cloning and sequencing of the Pr5452 5’ RACE products revealed that its product 1 had a TSS mainly at nt 7107 ( 3/8 colonies ) and product 2 had a TSS between nt 533–589 ( 6/8 colonies ) . Both were double spliced products either from nt 7243 to 3139 or from nt 757 to 3139 and then from nt 3431 to 5372 . We did not clone and sequence the faster migrating faint bands which are most likely the products of Pr7107 ( 216 bp ) , Pr360 ( 477 bp ) and Pr533 ( 304 bp ) being spliced from nt 7243 or nt 757 to 5372 . Using the Pr1123 for 5’ RACE ( lane 8 ) , cloning and sequencing revealed two smaller products being , respectively , transcribed from the nt 859 ( 7/12 colonies ) and nt 760 ( 3/5 colonies ) . Together , these 5’ RACE experiments identified nt 7107 , 7503 , 360 , 533 and 859 as preferred TSS for MmuPV1 gene expression . The mapped TSS all started at a purine A or G , which is in agreement with conserved TSS having a purine in eukaryotes [14] . The TSS at nt 7503 drives MmuPV1 E6 transcription , the TSS at nt 360 drives MmuPV1 E7 transcription and the TSS at nt 859 drives E2 expression . The TSS at nts 7107 and 533 drive MmuPV1 late transcription . Hereafter , each of the preferred TSS is named as a promoter: P7107 , P7503 , P360 , P533 or P859 . Analyses of the region 5′ to each mapped TSS show that the P7503 has a TATA box ( a eukaryotic core promoter motif ) 25-bp upstream of its TSS , but other two promoters for viral early gene expression do not ( S3B Fig ) . Two viral late promoters either bear a TATA-like box for P7107 or a TATA box 110-bp upstream of the promoter P533 ( S3B Fig ) . These features of viral promoters perhaps account for the observed heterogeneity of their transcription initiation as seen in the expression of HPV-18 [10] and other eukaryotic genes [15] . The P7107 late promoter identified above by 5’ RACE was confirmed by RNase protection assays ( RPA ) performed on total RNA isolated from MmuPV1-infected lesions using a 32P-labeled antisense RNA probe from nt 6846 to 7237 that covers the mapped TSS around nt 7107 . The RPA products were analyzed by electrophoresis in a denaturing 8% PAGE gel , along with the DNA sequencing ladders generated from MmuPV1 genome by a 32P-labeled antisense primer Pr7237 . As shown in S4 Fig , the protected RPA products from the 6846–7237 probe showed two major bands , one ( arrow ) of 131 nts in length corresponds to the TSS at nt 7107 . The other product of 217 nts ( arrowhead ) corresponds to the L1 mRNA cleaved at nt 7063 for the late polyadenylation ( see below ) . Genome analyses of MmuPV1 suggest a putative early poly ( A ) signal ( PAS ) AAUAAA downstream of the viral E2 ORF at nt 3844 presumably for early polyadenylation and two putative late PAS , at nt 5609 and 7047 , responsible presumably for polyadenylation of viral L2 and L1 transcripts ( Fig 3A ) . To determine further the early polyadenylation cleavage sites ( CS ) , total RNA isolated from MmuPV1-infected lesions was analyzed by 3′ RACE using a MmuPV1-specific sense primer Pr3277 located within the E4 ORF . Following gel purification , cloning , and sequencing of a 3′ RACE product of ~750 -bp ( Fig 3B ) , we found that 11/23 sequenced bacterial colonies exhibited a product with a 3′-end at nt 3864 , 15 nt downstream of the putative nt 3844 PAS ( Fig 3B ) . Additional 3′ RACE with a MmuPV1-specific sense primer Pr116 and Pr522 , located within the E6 and E7 ORF , respectively , also determined the early cleavage site around nt 3864 , confirming that MmuPV1 early transcripts are cleaved at nt 3864 for RNA polyadenylation using the nt 3844 PAS AAUAAA although its 3′ downstream sequence has no U/GU motifs , the highly conserved recognition sites for CSF ( cleavage stimulation factor ) binding in context of the RNA polyadenylation [16 , 17] . To detect the late CS , we carried out a 3′ RACE with MmuPV1-specific sense primers Pr6846 or Pr5433 located in the L1 ORF . Following gel purification and cloning of the Pr6846-derived 3′ RACE products of ~310 bp , we sequenced 16 bacterial colonies and found the usage of multiple cleavage sites for polyadenylation of MmuPV1 late transcripts . Eight clones exhibited a product with a 3′end at nt 7063 , 11 nt downstream of the putative nt 7047 PAS AAUAAA motif ( Fig 3C ) . 3′ RACE with an additional MmuPV1-specific sense primer Pr5433 also determined the late CS primarily mapping to nt 7063 ( S5 Fig ) . Analysis of the region downstream of this cleavage site shows three overlapping U/GU motifs from nts 7109 to 7152 . RPA analysis of the total cell RNA isolated from MmuPV1-infected lesions using the 32P-labeled antisense RNA probe from nt 6846 to 7237 further confirmed the presence of the late CS around the nt 7063 ( S4 Fig ) . In addition , we identified infrequent usage of the nt 5609 PAS AAUAAA for RNA polyadenylation at nt 5627 from Pr5433-derived 3’ RACE products ( S5 Fig ) . This PAS could be useful for the expression of L2 , but would lead to produce a truncated L1 protein . Because of its low frequency of usage , we consider that it represents a cryptic polyadenylation site of unknown function . Given the presence of multiple splice sites in both early and late transcripts of papillomaviruses , we used a snout ( muzzle ) wart sample from the animal #2 that gave 3 . 28 million viral reads , the highest among all twelve samples , to elucidate all possible usage of viral splice sites in the MmuPV1 genome . STAR aligner program [18] was used to explore the potential exon-exon splicing junctions with a threshold of minimal overhang >30 nts for non-canonical junctions and >10 nts for canonical junctions . As shown in Fig 4A , we identified from this sample 324 , 535 junction reads ( 10% of total viral reads ) that defined nine splicing junctions , with frequency by numbers of the junction reads in the order of nts 757/3139 >3431/5372 >7243/3139>757/2493 >7243/5372 >757/5372 >7243/2493 >1194/3139 >1125/3139 ( S3 Table ) . The most common splice junction read , at nts 757/3139 , accounted for 90% of total junction reads . The least common reads were those for the 1125/3139 splice , accounting for only 0 . 02% of the total junction reads . The identified splice junctions of 7243/3139 , 757/3139 , and 3431/5372 ( Fig 4A ) confirmed the findings from the 5’ RACE analyses using Pr3299 and Pr5452 primers ( Fig 2E ) . IGV Sashimi plot visualized that the obtained viral junction reads were derived from five different 5’ splice sites ( donor sites ) to three separate 3’ splice sites ( acceptor sites ) in the MmuPV1 genome ( Fig 4A ) . Analysis of the intron sequences between the exon-exon junctions confirmed that all introns contain a consensus GU dinucleotide in the intron 5′ end and a consensus AG dinucleotide in the intron 3′ end . The similar frequency patterns of the detected splice junctions were observed in all remaining samples , except the splice junction 1125/3139 which was detected only in 4 out of 9 tumor samples ( S3 Table ) . We noticed that the 757/3139 junction reads were proportionally higher in the sub-clinically infected control ears than that seen in the tumor ears ( S3 Table ) . Primer walking RT-PCR using various combinations of primer pairs ( Fig 4B ) on total mRNA isolated from MmuPV1-infected lesions was used to further validate the exon-exon junctions identified by RNA-seq and by 5’ RACE for MmuPV1 early and late transcripts . Gel-purification and sequencing of each RT-PCR product confirmed all of the splice junction identified by RNA-seq . Using a forward primer of Pr7140 downstream of the late promoter P7107 in combination with a backward primer of Pr5452 ( L1 ) , Pr4647 ( L2 ) , Pr3299 ( E4 ) or Pr2978 ( E2 ) , as shown in Fig 4C , we detected two L1 products ( lane 3 ) and one each for L2 , E4 and E2 ( lanes 5 , 7 and 9 ) . Sequencing of the gel-purified RT-PCR products indicated that two L1 products ( 185-bp and 478-bp , lane 3 ) were the singly spliced ( at 7243/5372 ) and doubly spliced ( at 7243/3139 and 3431/5372 ) transcripts , respectively ( Fig 4D ) ; both an L2 product ( 1612-bp , Fig 4C , lane 5 ) and an E4 product ( 265-bp , Fig 4C , lane 7 ) were singly spliced ( 7243/3139 ) transcripts ( Fig 4D ) ; and an E2 product ( 590-bp , Fig 4C , lane 9 ) was a singly spliced ( 7243/2493 ) transcript ( Fig 4D ) . Using a forward primer of Pr665 within the E7 ORF in combination with the same sets of backward primers also showed two L1 products ( Fig 4C , lane 12 ) and one each for L2 , E4 and E2 ( Fig 4C , lanes 14 , 16 and 18 ) . L1 product 1 ( 174-bp , Fig 4C , lane 12 ) was a singly spliced ( 757/5372 ) transcript , whereas the L1 product 2 ( 467-bp , Fig 4C , lane 12 ) was a doubly spliced ( 757/3139 and 3431/5372 ) transcript ( Fig 4D ) ; both L2 ( 1602-bp ) and E4 ( 254-bp ) products ( Fig 4C , lanes 14 and 16 ) were singly spliced ( 757/3139 ) transcripts; the E2 product ( 579-bp , Fig 4C , lane 18 ) was a singly spliced 757/2493 transcript ( Fig 4D ) . Using a backward primer Pr3299 in combination of a forward primer Pr1031 or Pr1141 , we detected three RT-PCR products in sizes of 257-bp ( product 1 ) , 326-bp ( product 2 ) and 2269-bp ( product 3 ) for Pr1031 and two products of 214-bp ( product 1 ) and 2158-bp ( product 2 ) for Pr1141 ( Fig 4C , lanes 20 and 22 ) . We found both 257-bp and 326-bp products were the spliced products of 1125/3139 and 1194/3139 and the 214-bp product was also a spliced product of 1194/3139 ( Fig 4D ) . Both the 2269-bp and 2158-bp products were unspliced products from this region presumably for E1 expression . Intron retention is one of the common alternative splicing events during RNA splicing and is essential for the expression of E1 and L2 in all papillomaviruses . In high-risk HPVs , intron retention is also necessary for viral E6 expression . From RNA-seq analysis , a small fraction of viral reads spanned over the entire MmuPV1 E1 and L2 ORF regions ( Figs 1D and 4A ) . These were further confirmed by primer-walking RT-PCR with a series of combined primer pairs ( Fig 5A ) from total RNA extracted from the MmuPV1-infected muzzle wart tissues of animal #2 . As shown in Fig 5B , the primer-walking RT-PCR using a forward primer Pr7140 in combination with a reverse primer Pr135 , Pr352 , Pr738 , Pr827 or Pr1938 for detection of the P7107-derived late transcripts , which are predominantly spliced from the nt 7243 5’ss to nt 3139 3’ss , led to the amplification of transcripts with an intron retained from nt 7244 to 3138 ( lanes 2 , 4 , 6 , 8 and 10 ) . Using a forward primer Pr1263 in combination with a backward primer Pr4647 for L2 or Pr2978 for E2 ( lanes 13–14 ) , we obtained a 1715-bp product ( lane 14 ) from the Pr2978 , but none from the Pr4647 ( lane 13 ) , indicating retention of the intron within the E1 and E2 ORFs , but not together with retention of another intron in the L2 region . Using a forward primer of Pr1141 in combination with a backward primer of Pr3299 , we detected a 214-bp product spliced from nt 1194 to 3139 ( lane 18 ) . The other primer pairs used in the same detection did not give any obvious RT-PCR products ( lanes 16 , 17 , 19 and 20 ) . Next , we compared the primer-walking RT-PCR using fewer amplification cycles ( 25 cycles ) and using an exonic or intronic backward primer in combination with the same sets of forward primers ( Pr7140 , Pr116 , Pr522 or Pr665 ) to detect intron-containing viral transcripts derived from individual promoters in the same muzzle wart tissues . As shown in Fig 5C , we detected two major spliced products from the exonic backward primer Pr5452 ( L1 ) plus the forward primer Pr7140 , Pr522 or Pr665 ( lanes 2 , 6 and 8 ) , the larger amplicon arose from double splicing ( either 7243/3139 or 757/3139 and then 3431/5372 ) and the smaller amplicon arose from single splicing ( 7243/5372 or 757/5372 ) , depending on the promoter usage . In this case , only a very few products could be amplified from the primer pair of Pr5452 plus Pr116 ( lane 4 ) . However , the primer pair of Pr5452 plus Pr522 gave much less amount of the spliced L1 products ( lane 6 ) over the primer pair of Pr5452 plus Pr665 ( lane 8 ) , but relatively more than the primer pair of Pr5452 plus Pr116 ( lane 4 ) , the data suggest that L1 messages are the spliced product primarily transcribed from two promoters P7107 and P533 . But few L1 might be also derived from the promoter P360 . The identity of a faint band above 650-bp size from the primer pair of Pr5452 and Pr7140 ( lane 2 ) was unknown and might be nonspecific . When an intronic backward primer Pr4647 in the L2 region was used in combination with the same sets of the forward primers , we obtained only one major product predominantly transcribed from two promoters P7107 and P533 ( lanes 10 and 16 ) and spliced from 7243/3139 ( P7107 ) or 757/3139 ( P533 ) , but very little from the promoter P7503 or P360 ( lanes 12 and 14 ) , suggesting that these two promoter-derived transcripts are used for L2 production . To detect intron retention in the viral early region required for the expression of E1 , the primer-walking RT-PCR results from an exonic backward primer Pr3299 ( lanes 19–26 ) were compared with backward primers Pr2978 ( lanes 27–34 ) , Pr1938 ( lanes 36–43 ) or Pr827 ( lanes 44–51 ) in combination of the same sets of primers described above for the detection . We found that the Pr3299 amplified one single spliced RT-PCR product of 7243/3139 for P7107-derived transcripts ( lane 19 ) or 757/3139 for P7503- , P360- or P533-derived transcripts ( lanes 21 , 23 and 25 ) . The Pr2978 detected a single spliced , weak product of 7243/2493 for P7107-derived transcripts ( lane 27 ) or 757/2493 for P360- ( lane 31 ) , but a little more of 757/2493 for P533-derived transcripts ( lane 33 ) and none from the promoter P7503 ( lane 29 ) . The Pr1938 detected at low levels an unspliced E1 transcripts derived from the P360 promoter ( lane 40 ) , but more from the P533 promoter ( lane 42 ) and none from promoter P7107 ( lane 36 ) or P7503 ( lane 38 ) . In contrast , the primer Pr827 in combination of each forward primer led to detection of an E1 intron-containing product mostly from three promoters P7503 , P360 and P533 ( lanes 46 , 48 and 50 ) , but little from P7107 ( lane 44 ) . We also confirmed in an MmPV1-induced ear wart of animal #3 that the L1 messages were the spliced products primarily transcribed from the promoter P533 , but less from the promoter Pr360 ( compare lanes 2 and 4 in Fig 5D with lanes 6 and 8 in Fig 5C ) . Moreover , the intron-containing E1 transcripts were confirmed to be derived from the promoter Pr7107 and Pr7503 ( lanes 6 and 8 in Fig 5D and lanes 44 and 46 in Fig 5C ) . Further studies using an E1-specific primer Pr1263 for 3’ RACE demonstrated that the unspliced E1 transcripts are polyadenylated at nt 3864 position , an early CS site ( Fig 5E ) . Based on the mapped TSS , polyadenylation cleavage sites , and RNA splice sites of MmuPV1 early and late transcripts , we constructed a full transcription map from the MmuPV1 genome . As shown in Fig 6 . MmuPV1 expresses at least 36 RNA isoforms spanning the entire MmuPV1 genome , with the majority of them being polycistronic transcripts that can potentially translate multiple gene products . MmuPV1 early transcription mainly starts at nt 7503 for E6 polycistronic or nt 360 for E7 polycistronic RNAs . Both are polyadenylated at nt 3864 using a PAS at nt 3844 . MmuPV1 late transcription mainly starts either at nt 7107 or nt 533 and polyadenylates either at nt 3864 using the nt 3844 PAS for E1^E4 expression or at nt 7063 using a PAS at nt 7047 , the latter encoding L1 and L2 proteins . Most viral early transcripts contain two exons and one major intron spanning the entire E1 ORF and partial E2 ORF . Although the majority of the viral early transcripts have intron 1 spliced out , a small fraction of early transcripts retains this intron ( Fig 5B , lane 14 , Fig 5C , lanes 40 and 42 ) and could be further spliced ( Fig 4C , lanes 20 and 22; Fig 5B , lane 18 and Fig 5C , lanes 31 and 33 ) . Most viral late transcripts have three exons and two major introns and their 5’ portions overlap with the viral early transcripts . Depending on which late promoter drives transcription , the 5’ end of the first intron can start either at nt 7244 or 758 . If starting from nt 758 , the first intron of the late transcripts is the same as found in the viral early transcripts . The second intron of the late transcripts is invariably from nt 3432 to nt 5371 and covers the entire L2 ORF and the early PAS . Retention of the intron 2 ( Fig 6 , products C , D , G , H , Z ) and therefore capacity to encode L2 were found only in a small fraction of the late transcripts ( Fig 4C , lanes 5 and 14; Fig 5C , lanes 10 and 16 ) . The more abundant viral late transcripts are the ones that encode for E1^E4 and/or L1 ( Fig 6 , products A , X , AB ) . As shown in Fig 6 , both MmuPV1 early and late transcripts are alternatively spliced , and the coding capacities of each RNA species may be inferred from the ORF ( s ) included in the mRNA . We reassigned E4 as E1^E4 with an AUG codon starting from nt 742 position in the viral transcript exon 1 instead of the nt 3101 from a previous publication [4] and thus expression of the E4 ORF requires RNA splicing . We also reassigned the L2 ORF AUG start codon to be at nt 3745 instead of at nt 3735 in the prior publication [4] and the L1 ORF AUG start codon to be at nt 5372 instead of at nt 5291 in the prior publication [4] . To remove the intron 2 during RNA splicing , the viral late transcript exon 2 is spliced right to nt 5372 of the exon 3 and thus the 5291 AUG codon does not exist in the L1 mRNA after RNA splicing . Moreover , we identified the E8^E2 ORF as a spliced ORF ( nt 1094-1125/nt 3139–3685 ) and two small E1 ORF variants , E1^M1 ( nt742-1194/nt3139-3231 ) and E1^M2 ( nt742-1125/nt3139-3231 ) not previously described . A detailed analysis of the upstream sequences of these ORFs indicated that the first AUG codon of each contains a strong Kozak consensus sequence of either ANNaugN or GNNaugG [19 , 20] . As shown in Figs 1B and 2E , most RNA reads or transcripts appear to be derived from the promoter P533 . The relative expression levels of viral transcripts in the infected wart tissues were also measured semi-quantitatively by primer-walking RT-PCR with fewer cycle ( 25 cycles ) amplification ( Fig 5C ) and the data in this study also suggested that both L1 and L2 are transcribed from either promoter P7107 or P533 ( Fig 5C , lanes 2 , 8 , 10 and 16 ) , but most viral RNA transcripts are P533 transcripts ( Fig 5C , compare lane 25 to lanes 23 , 21 , and 19 ) . To quantify the expression levels of the existing viral transcripts better in the infected tissues , Northern blot analysis using 32P-labeled oligo probes , Pr7237 , Pr352 , Pr687 , Pr3299 , Pr3682 or Pr5452 ( Fig 7A ) , was further used to evaluate the quantitative levels of the existing viral RNA transcripts from individual promoters . In this study , the total RNA from mouse ears without MmuPV1 subclinical infection served as a MmuPV1-negative RNA control and was pooled RNA from ears of two naïve , freshly arrived female mice in ~4 months of age , with no detectable MmuPV1 reads by RNA-seq analysis . As shown in Fig 7B , we found that the late-transcript-specific probes Pr7237 ( lane 2 ) and Pr5452 ( lane 12 ) in Northern blotting displayed a comparable expression profile of the viral late region as did from the viral early region detected by the early transcript-specific probes Pr687 ( lane 6 ) and Pr3299 ( lane 8 ) , with the more hybridization signals seen from the Pr3299 and Pr5452 than the corresponding comparable probes . Size analyses of individual transcripts detected by each probe demonstrated that L1 transcripts of ~2 . 3-kb in size are predominantly transcribed from the promoter P7107 ( product A in lanes 2 , 8 and 12 ) or P533 ( product X in lanes 6 , 8 and 12 ) and are preferentially doubly spliced from nt 7243 to 3139 ( P7107 ) or from nt 757 to 3139 ( P533 ) and then from nt 3431 to 5372 , accompanied by a lower abundance of singly spliced L1 transcripts from nt 7243 ( P7017 , product B in lanes 2 and 12 ) or nt 757 ( P533 , product Y in lane 12 ) to 5372 . In principle , a minor form of the doubly spliced L1 transcript Q derived from promoter P360 also exist in these Northern blot assays ( lanes 6 , 8 and 12 ) . As expected , these doubly spliced L1 transcripts ( A/Q/X ) were not detectable by the probe Pr3682 ( lane 10 ) hybridizing to a downstream region of the 3431 5’ donor site . A small fraction of 4 . 2-kb products ( product C in lanes 2 , 8 , 10 and 12; product Z in lanes 8 , 10 and 12 ) are L2 mRNA arising from nt 7243 splicing to nt 3139 in the case of P7107-derived transcripts or from nt 757 splicing to nt 3139 in the case of P533-derived transcripts . The ~1-kb L1 products ( lanes 2 and 12 in a question mark ) appear to be a product of a cryptic PA usage in the L1 ORF ( S5 Fig ) . As predicted from RNA-seq analysis in Fig 4A , the majority of the P533-derived RNA transcripts are spliced from nt 757 to 3139 , therefore encoding E1^E4 , and utilize the CS at nt 3864 for RNA polyadenylation . As shown in Fig 7B , Northern blotting using 32P-labeled oligo probe Pr687 ( lane 6 ) or Pr3299 ( lane 8 ) was able to detect the abundant E1^E4 transcripts ( ~1 . 2-kb , products AB ) and the minor forms of E6 ( ~1 . 4-kb , product M ) , E7 ( ~1 . 1-kb , product T ) , E2 ( ~1 . 8-kb , products L/S ) , E1 ( ~4 . 2-kb , products K or K/P ) , L1 ( ~2 . 3-kb , products Q/X or A/Q/X ) and L2 ( ~4 . 3-kb , products C and Z ) . Oligo probe Pr3682 in this assay ( Lane 10 ) displayed the similar detection profile to these two probes except for the doubly spliced L1 which lacks the target sequence for the probe Pr3682 hybridization . Northern blotting using a 32P-labeled oligo probe Pr352 exhibited a predominant E6 RNA ( ~1 . 4-kb , product M ) transcribed from P7503 as detected by PacBio Iso-seq ( Fig 2C and 2D , Table 2 ) and spliced from nt 757 to 3139 in addition to the E1 RNA of ~3 . 9-kb ( products P ) from this promoter ( lane 4 ) . The ~0 . 8-kb products ( lanes 8 and 10 in a question mark ) appear to be the products of alternative late promoters around nt 576 to 607 ( Fig 2C ) , which are spliced from nt 757 to nt 3139 , but polyadenylated at nt 3864 or are doubly spliced and polyadenylated at a cryptic poly ( A ) site in the L1 ORF ( S5 Fig ) . Thus , our data from the Northern blotting are consistent with the conclusion from RNA-seq analysis ( Fig 4A ) and semi-quantitative RT-PCR ( Fig 5C ) that the majority of viral transcripts are spliced products of 757/3139 and are polyadenylated at nt 3864 , using an early PAS at nt 3844 for expression of the early region , and the fewer are polyadenylated at nt 7063 , using a late PAS at nt 7047 for expression of the late region . Subsequently , we also examined viral late transcripts in MuPV1-induced papillomas by RNA-ISH ( in situ hybridization ) . Using antisense probes to the E4 and L1 regions ( S6A Fig ) and the RNAscope methodology which is highly sensitive in detection of both viral RNA and DNA of MmuPV1 , we detected E4 and L1 signals primarily in the highly differentiated granular layers of the infected , hyperproliferative ear skin ( S6B Fig ) . These patterns of E1^E4 and L1 expression were also seen in the tail papillomas ( S6C Fig ) . Although these probes in RNAscope technology detect both viral RNA and DNA , pretreatment of tissue sections with DNase and/or RNase allowed us to distinguish between the DNA-derived and RNA-derived signals using this methodology ( S6D and S6E Fig ) . We found that the detected viral E1^E4 transcripts appeared more cytoplasmic distribution than the L1 transcripts did ( S6C Fig ) , particularly after removal of viral genomic DNA by DNase I treatment of the tissue sections ( S6D Fig ) . To understand the expression dynamics of viral L1 and MmuPV1 DNA over time , nude mice were infected ( three spots on the tail ) with equivalent amounts of MmuPV1 per site ( 108 VGE ) and tissue was harvested by sacrificing animals at different time points until papillomas were overtly seen ( 28 days post-infection ) . These tissues collected at each time point were analyzed for appearance of L1 viral protein and amplified viral DNA , hallmarks of the productive phase of the viral life cycle . As shown in Fig 8A and S7 Fig , no obvious L1 expression or viral DNA was detected in any of the infected sites ( 0/3 infected sites ) at 4 days post-infection . Hypertrophic scarring of terminal epithelia with intact basement membrane was observed . At day 10 , both L1 and MmuPV1 DNA were detected in one of the infected sites ( that closest to base of tail—shown in Fig 8A ) . L1 expression was found in the suprabasal layers along with terminally differentiating epithelia whereas MmuPV1 DNA appeared to be present only in suprabasal layers . Microscopically , this site had evidence for hyperplasia , koilocytes and some fibrillary projections; however , the overt appearance was that of a raised scar . At day 21 , again one site of infection ( shown in Fig 8A ) , that closest to the tail base , showed evidence of productive infection with an abundance of L1-positive and viral DNA-positive cells and more prominent fibrillary projections . The remaining 2 infected sites appeared mostly normal and no L1 or FISH positive nuclei were detected . At 28 days post infection , all three infected sites showed overt appearance of warts with L1 positivity ( S8 Fig ) . While sites were infected at the same time , papillomas grew asynchronously as is evident by variation in size of the papillomas . We consistently observe that papillomas near the base of the tail grow fastest and show the most robust features of a productive viral infection . The pattern of L1 expression and viral DNA amplification at day 28 was similar to that observed at day 21 ( Fig 8A ) . A similar timing and pattern of detection of viral RNA/DNA species was observed using the RNAscope in situ hybridization methodology with a probe to the L1 region ( Fig 8B ) . Next , we attempted to express each of the predicted MmuPV1 proteins based upon the transcript map in HEK293 and HeLa cells transiently transfected with individual FLAG-tagged viral ORF cDNA expression vectors , including MmuPV1 ORF E6 ( GenBank Accession #MF350298 ) , E7 ( GenBank Accession #MF350299 ) , E1 ( GenBank Accession #MF350300 ) , E2 ( GenBank Accession #MF350301 ) , L2 ( GenBank Accession #MF350302 ) , L1 ( GenBank Accession #MF350303 ) , E1^E4 ( GenBank Accession #MF350304 ) , E8^E2 ( GenBank Accession #MF350305 ) , E1^M1 ( GenBank Accession #MF350306 ) and E1^M2 ( GenBank Accession #MF350307 ) ( Fig 9A and 9B ) . As shown in Fig 9C , we were able to detect the expression of E2 ( lane 2 ) , E1^E4 ( lane 3 ) , E1^M1 ( lane 4 ) , E1^M2 ( lane 5 ) , E7 ( lane 9 ) and E8^E2 ( lane 10 ) in HEK293 cells , but were unable to detect E1 ( lane 1 ) , L1 ( lane 6 ) , L2 ( lane 7 ) and E6 ( lane 8 ) by FLAG-specific immunoblot analysis . However , both viral E6 and E7 could be better detected when HEK293 cells were treated with proteasome inhibitor MG132 ( Fig 9D ) , indicating they are likely degraded via the proteasome . We also failed to express E1 , L1 and L2 in mouse epithelial keratinocytes ( S9A Fig ) and HEK293TT or HEK293FT cells . By Northern blot analysis of the total RNA extracted from transfected HEK293 cells , we were unable to detect both L1 and L2 RNA , but identified two spliced E1 RNA in smaller sizes and all other expected sizes of viral ORF-derived RNA transcripts ( S9B Fig ) . Using FLAG-specific immunofluorescence ( Fig 9E ) , we detected the expression of E6 , E7 , E2 , and E8^E2 mainly in the nucleus of HeLa cells , E1^E4 as a filamentous protein in the cytoplasm , E1^M1 either in the cytoplasm or nucleus or both , and E1^M2 primarily in the cytoplasm . Distribution of viral E6 and E7 in the cells could be only slightly altered in the presence of MG132 ( Fig 9E ) .
Understanding of the structure and coding capacity of transcripts is critical for disclosing genome function and biology of any organism . In this report , we have utilized two cutting-edge technologies , RNA-seq and PacBio Iso-seq , in combination with various conventional technologies to analyze the structure and expression of MmuPV1 genome in MmuPV1-induced wart tissues with productive MmuPV1 infection . We have constructed the first full transcription map for MmuPV1 and demonstrated that MmuPV1 genome encodes ten ORFs and utilizes five major promoters , two polyadenylation sites and eight splice sites for its expression of thirty-six RNA isoforms during virus infection . Similar to other papillomaviruses [8 , 10 , 21–28] , this nature of the genome structure with alternative usage of promoters , pA sites and splice sites empowers a highly compact viral genome to express multiple gene products in a temporally and spatially organized manner within its viral life cycle . The delineated MmuPV1 genome structure and expression of MmuPV1 is more close to that of bovine papillomavirus type 1 ( BPV-1 ) , cottontail rabbit papillomavirus ( CRPV ) , cutaneous HPVs and some low-risk mucosotropic HPVs . Like all other papillomaviruses [7 , 8] , the MmuPV1 genome transcribes unidirectionally from one DNA strand , with ~99% of RNA-seq reads mapping to viral sense transcripts . Similar to cutaneous HPVs and low-risk mucosotropic HPVs [25 , 29 , 30] , MmuPV1 employs two separate early promoters for expression of viral E6 and E7 , the promoter P7503 for the E6 expression and the promoter P360 for the E7 expression . These two promoters could be also responsible for expression of other viral early proteins , including E1 , E1^M1 , E1^M2 , E2 and E8^E2 . This strategy for expression of MmuPV1 E6 and E7 is different from high-risk HPVs in that their expression of both viral E6 and E7 is exerted by a single early promoter upstream of the E6 ORF [10 , 31 , 32] and the expression of viral E7 requires RNA splicing of an E6 intron in this early transcript [19 , 33 , 34] . Also similar to low-risk HPVs [25 , 29 , 30] and other animal papillomaviruses [23 , 24] , MmuPV1 has no intron in the E6 ORF region . We found that the promoter P7503 is the only early promoter bearing a classical TATA box ( a eukaryotic core promoter motif for binding of RNA polymerase II ) 25-nts upstream of the promoter TSS . Based on our 5’ RACE and PacBio Iso-seq data , we conclude that the P7503 is stronger than the other two early promoters , P360 and P859 . The third early promoter P859 is a weak promoter most likely driving the expression of E2 and E8^E2 . Because of anticipated RNA-seq bias on eukaryotic RNA 5’ ends [13] , fewer RNA reads next to the P7503 and P360 promoters were noticed ( Fig 4A ) . MmuPV1 transcribes its late transcripts from two late promoters , P7107 in the LCR region downstream of L1 ORF and P533 in the E7 ORF region . Utilization of a late promoter in the LCR region for the expression of viral late gene L1 and L2 is a characteristic feature for BPV-1 [22] , CRPV [24] and some skin-tropic HPVs such as HPV-1 [35 , 36] and HPV-5 [29] , but not for Mastomys natalensis papillomavirus ( MnPV ) [23] and other HPVs [10 , 30–32] . In contrast , high-risk HPVs express their late genes mainly from a late promoter in the E7 ORF [10 , 31 , 32 , 37] . Similar to high-risk HPVs , the late promoter P533 in the MmuPV1 genome is most likely responsible for E1^E4 expression , but also for L1 and L2 expression . Although the transcripts derived from promoter P360 or P859 might have the potential to encode L1 , they were scarcely detectable from the infected tissues and could be negligible . We found that the P7107 has a TATA-like box 57-nts upstream of its TSS and the P533 bears a TATA box at 110-nts upstream of its TSS . Posttranscriptional RNA processing , including RNA capping , splicing , polyadenylation and export , provides multiple layers of regulation to guide efficient expression of eukaryotic genes [38 , 39] . By using 3′ RACE , we mapped the cleavage sites of both viral early and late transcripts for RNA polyadenylation and demonstrated that viral early transcripts are polyadenylated primarily at nt 3864 by using a PAS at nt 3844 and viral late transcripts are polyadenylated primarily at nt 7063 by using a PAS at nt 7047 . In addition , we have identified a few late transcripts being polyadenylated from nt 5627 by using a PAS at nt 5609 in the L1 ORF . Analyses of the sequences 3’ downstream of each mapped CS site for a highly conserved recognition site U/GU for CSF ( cleavage stimulation factor ) binding in RNA polyadenylation [16 , 17] showed three U/GU motifs in this region of the nt 7063 , but not so for the nt 3864 . Thus , what motif guides the polyadenylation cleavage of the nt 3864 remains unknown . Nevertheless , the mapped polyadenylation site usage for the expression of MmuPV1 early and late transcripts resembles to that of all papillomaviruses [8 , 10 , 22–24 , 29 , 30 , 32] . Extensive alternative RNA splicing contributes to the expression of multiple genes by papillomaviruses [8 , 10 , 22–24 , 29 , 30 , 32] . Our study revealed this feature in MmuPV1 gene expression by analyzing exon-exon splice-junction reads from RNA-seq and by primer-walking RT-PCR analyses of RNA extracted from MmuPV1-induced warts . We demonstrated that MmuPV1 employs five 5’ splice sites ( donor sites ) and three 3’ splice sites ( acceptor sites ) for expression of both viral early and late genes from its five promoters and produces at least thirty-six different RNA isoforms by alternative RNA splicing , of which thirteen were detectable from wart tissues by less sensitive Northern blotting . Thus , the primary MmuPV1 transcripts could have 2 or 3 exons and 1 or 2 introns , with viral E1 and L2 residing in the most commonly excised introns , as seen in other papillomaviruses . As with other papillomaviruses , a small fraction of MmuPV1 RNA transcripts does retain the capacity to express viral E1 and L2; the mechanism by which these RNA transcripts retain the introns for expression of the E1 or L2 should be an attractive area for future investigation . By construction of this full transcription map , we conclude that MmuPV1 has the potential to express 10 gene products: E6 , E7 , E1 , E1^M1 , E1^M2 , E1^E4 , E2 , E8^E2 , L2 and L1 . Like cutaneous HPVs [40] , MmuPV1 does not contain an E5 ORF and therefore no E5 gene product is predicted . The coding regions for the E1^E4 , L2 and L1 gene products have been reassigned from what was originally predicted based solely on viral DNA sequence information [4] . E1^M1 , E1^M2 and E8^E2 have not been described before for MmuPV1 . The E1^M1 and E1^M2 in MmuPV1 might be similar to the E1Ma and E1M in HPV-11 [25] . The MmuPV1 E8^E2 gene product is likely similar to the E8^E2 gene products characterized in BPV-1 [41–44] and HPVs [45 , 46] and predicted to be encoded by most papillomaviruses [47] . Although the first AUG codon for each of the predicted coding regions has a strong Kozak sequence [20] , we were unable to detect the expression of MmuPV1 E1 , L2 and L1 proteins from a common eukaryotic expression vector in HEK293 , MEK , 293FT or 293TT cells and unable to detect by Northern blotting a full-length RNA expressed from E1 , L2 and L1 vector in HEK293 cells . Several possible reasons for this , based upon studies of other papillomaviruses , include rare codon usage [48–50] , RNA splicing or instability [8 , 51 , 52] and protein stability [53–56] . For example , the E1 ORF in the expression vector contains three splice donor sites ( nt 757 , nt 1125 and nt 1194 splice donor sites ) and one acceptor site ( nt 2493 splice acceptor site ) . The RNA expressed from this vector in HEK293 cells that we could detect by Northern analysis was all spliced forms; no unspliced mRNA capable of expressing full-length E1 could be detected . Among seven viral proteins detected , MmuPV1 E6 and E7 were also found to be increased in their steady state by using a proteasome inhibitor , suggestive that they are normally subjected to proteasomal degradation as seen for HPV-16 [53] . Several lines of evidence indicate that MmuPV1 shares a number of genomic , molecular and pathological features with high-risk cutaneous , beta HPVs and thus could be a useful model to study these important human pathogens . These include our observation that MmuPV1 contains separate promoters for E6 and E7 , and the additional facts that MmuPV1 causes squamous cell carcinomas at cutaneous sites [57] , lacks an E5 ORF [4] and encodes an E6 protein that shares with HPV8 E6 the ability to bind MAML1 and SMAD2/SMAD3 but not E6AP and p53 [58] . In this study , we also performed a careful assessment of the timing of onset of viral gene expression and productive amplification of viral DNA in the context of emerging warts . These studies used a fixed amount of virus ( 108 VGE per infection site ) applied to three scarified sites on the tail of each nude mouse . We found that both viral late gene expression ( L1 RNA and capsid ) and viral DNA amplification could be observed in differentiated cell compartment as early as 10 days post infection , well before overt warts can be observed , which starts at 4 weeks with this dose of virus and this strain of mouse . We observed variability in the onset of detectable infection , be it at a microscopic or overt level , with the some infection sites not showing any microscopic evidence of infection until the 4-week time point when overt warts first appeared . Interestingly , we observed a reproducible spatial pattern in which sites of infection at the base of the tail gave rise to faster growing warts than sites infected at the tip of the tail . The reason for this is unknown , but some possibilities may include blood flow , temperature , and grooming behavior . Another feature of the time course study is that L1 positive cells were throughout the epithelium , including basal cells in mature warts harvested at 3 or 6 months post-infection , an uncommon feature of MmuPV1 observed by others [6] . However , the L1-positive cells were restricted to differentiated cells in the early time points , out to day 28 post-infection ( Fig 8A ) , suggesting that the complete nature of the viral life cycle is realized at times later than 4 weeks post-infection . Interestingly , canine oral papillomavirus ( COPV ) [59] is also found to be amplified in basal epithelial cells , though the timing at which this first appears is slightly different . COPV , while a mucosal papillomavirus , is closely related to cutaneous HPVs , HPV-1 and HPV-63 , that cause plantar warts [60 , 61] . Basal cells have been seen to support viral DNA amplification in lesions caused by HPV-1 and HPV-63 [62] . Lastly , we made the interesting observation that in experimentally infected nude mice that have developed warts at the sites of infection , other areas of the epidermis can show evidence for subclinical infections . These subclinical infections showed the same distribution pattern of viral RNA-seq reads as the experimentally infected sites with warts when the reads mapped to the reference MmuPV1 genome ( Fig 1B ) , and showed evidence for productive infection , based upon the detection of viral DNA amplification and L1 expression ( S1B Fig ) . This raises the intriguing possibility that subclinical infections may be common in immunodeficient or immunosuppressed contexts . In this regard , organ transplant patients are known to have an increased abundance of HPV DNA in randomly sampled hair follicles from clinically normal skin [63] . In conclusion , we observe a similar transcription pattern for MmuPV1 as observed with animal papillomaviruses and some HPVs . We believe this carefully mapped landscape of MmuPV1 transcription from MmuPV1-induced warts will provide a solid foundation for future understanding of MmuPV1 molecular biology , pathogenesis and immunology .
Immunodeficient athymic BALB/c FoxN1nu/nu used in this study were obtained from Harlan ( currently Envigo , Indianapolis , IN ) . All infected mice ( 6–8 weeks old at the time of infection ) were housed in aseptic conditions in micro-isolator cages . Animals were handled only by designated personnel and personal protection gear was changed between cages to prevent any virus cross-contamination . Experimental infection was performed using quasivirions containing MmuPV1 synthetic genome as described previously [57 , 64 , 65] . The synthetic MmuPV1 genome ( Gift from Dr . Chris Buck , NCI ) is identical to the original wild type genome and has been described previously [6] . Briefly , 293FT cells ( Thermo Fisher Scientific , Waltham , MA ) , a fast growing 293T cell line , were co-transfected with a codon optimized MmuPV1 capsid protein expression plasmid ( pMusSheLL , gift from Dr . Chris Buck , NCI ) [6 , 66] and recircularized MmuPV1 synthetic genome for encapsidation . The cells were harvested 48 h after cotranscfection and virions were purified using Optiprep ( Sigma-Aldrich , St . Louis , MO ) gradient centrifugation . The generated quasivirions were quantified by estimating viral genome equivalents ( VGE ) by comparing the amount of encapsidated viral DNA in the viral stock by Southern blot analysis using MmuPV1-specific probes , followed by quantification using ImageJ software as described previously [57] . BALB/c FoxN1nu/nu mice ( 6–8 weeks old ) were infected with 2×108 VGE MmuPV1 per site after scarifying skin of tail , ear or muzzle as described previously [57 , 58] . The wart tissues from three anatomical sites ( ear , tail and muzzle ) were collected from each animal 6 months post-infection and snap-frozen in liquid nitrogen for RNA isolation and a portion of the papillomas was excised , fixed in 10% neutral buffered formalin and embedded in paraffin . Serial sections ( 5 μm thick ) were stained with hematoxylin and eosin ( H&E ) and evaluated for histopathological features and processed for subsequent analyses . All animal experiments were performed in full compliance with standards outlined in the "Guide for the Care and Use of Laboratory Animals” by the Laboratory Animal Resources ( LAR ) as specified by the Animal Welfare Act ( AWA ) and Office of Laboratory Animal Welfare ( OLAW ) and approved by the Governing Board of the National Research Council ( NRC ) . Mice were housed at McArdle Laboratory Animal Care Unit in strict accordance with guidelines approved by the Association for Assessment of Laboratory Animal Care ( AALAC ) , at the University of Wisconsin Medical School . All protocols for animal work were approved by the University of Wisconsin Medical School Institutional Animal Care and Use Committee ( IACUC , Protocol number: M02478 ) . A home-based tyramide-based signal amplification ( TSA ) method was developed as described previously to detect MmuPV1 L1 [67 , 68] . Formalin-fixed paraffin embedded tissue slides were deparaffinized after 3 changes of xylene followed by rehydration in ethanol series ( 100% , 95% , 70% , 50% and finally double distilled water ) . Endogenous peroxidase activity was blocked using 0 . 3% hydrogen peroxide in methanol . Antigen retrieval was performed using antigen retrieval buffer ( pH = 9 . 0 , Abcam , Cambridge , MA , #ab93684 ) for 20 minutes in a microwave . Slides were cooled to room temperature and blocked for 1 h at room temperature in blocking buffer ( Perkin Elmer , Fermont , CA , #FP1012 ) . Rabbit sera against MmuPV1 L1 ( Gift from Dr . Chris Buck , NIH ) [6 , 69] was diluted at 1:5000 in blocking buffer and applied to sections overnight at 4°C . Samples were incubated with goat anti-rabbit-HRP secondary antibody ( at 1:500 dilution ) in blocking buffer for 1 h at room temperature . Subsequently , the secondary antibody was biotinylated by incubating with biotin-tyramide ( 10 μg/ml ) for ten minutes as described previously [68] . Slides were rinsed with PBS ( phosphate-buffered saline ) containing 0 . 1% Tween-20 and a cytokeratin cocktail containing equal amounts of anti-K14 ( BioLegend , San Diego , CA , #PRB-155P ) and anti-K10 ( BioLegend , #PRB-159P ) at 1:1000 dilution was applied at room temperature for 1 hour . Slides were rinsed with PBS containing 0 . 1% Tween-20 and incubated with secondary detection reagents as follows—anti-rabbit conjugated with Alexa Fluor488 ( Thermo Fisher Scientific , #A11008 ) at 1:500 to detect K10 and K14 , and Streptavidin-Alexa Fluor-594 ( Thermo Fisher Scientific #S-32356 ) at 1:500 to detect biotinylated L1 for 1 hour at room temperature . Tissues were counter stained with Hoechst for cellular DNA and coverslips were mounted using ProLong gold antifade ( Thermo Fisher Scientific , #P36930 ) . MmuPV1 DNA FISH was performed as described previously [57 , 70] . This protocol has been adapted from a DNA FISH protocol used to detect Epstein Barr Virus ( EBV ) DNA in monolayer cells and is described in detail at: https://mcardle . oncology . wisc . edu/sugden/protocols . html . Briefly , formalin-fixed paraffin embedded tissue slides were baked at 65°C overnight and deparaffinized using xylene followed by treatment with 100% ethanol . Slides were then boiled in 10 mM sodium citrate buffer ( pH = 6 . 0 ) for 30 minutes in a microwave . Slides were rinsed with PBS and completely dried before pre-hybridizing with 2 x SCC containing RNase A and 0 . 5% IPEGAL ( pH = 7 . 0 ) for 30 min at 37°C . Slides were dehydrated using a series of ice cold ethanol ( 70% , 80% , 95% ) for 2 min each . Slides were dried by placing them in an empty container at 50°C for 5 min and then placed in denaturation solution [28 ml formamide , 4 ml 20 x SSC ( pH = 5 . 3 ) and 8 ml water] at 72°C for 2 min . The ethanol series was repeated again , and after drying the sections , denatured probe was added to the slides . A biotin-16-dUTP ( Sigma-Aldrich , #11093070910 ) labeled probe was hybridized to tissue overnight at 37°C in a humidified chamber . To make the probe , nick translation was used to label the entire MmuPV1 plasmid DNA ( pMusPV [6] ) with biotin . Slides were then washed twice for 30 min with 2 x SSC and 50% formamide at 50°C followed by two washes for 30 min with 2 x SSC at 50°C . Signals were detected with streptavidin conjugated to Cyanine-3 ( Sigma-Aldrich , #S6402 ) at 1% by volume in STM solution ( 4 x SSC , 5% non-fat dried milk , 0 . 05% Tween-20 , 0 . 002% sodium azide ) for 30 min at 37°C . Nuclei were counterstained with Hoechst coverslips were mounted using ProLong gold antifade ( Thermo Fisher Scientific , #P36930 ) . High resolution wide-field fluorescent images were acquired by means of a super-resolution Leica SP8 STED confocal microscope equipped with a motorized stage . This microscope is equipped with PMT and HyD lasers . All images were taken by means of a 20X objective lens ( Specifications: HC PL APO 20x/0 . 75 CS2 , Dry ) . The images were acquired by tile-scanning by marking positions around the region of interest on the LAS-X suite ( version: 2 . 0 . 1 ) . The merged wide-field image was obtained by automatic stitching of individual styles by means of in-built auto stitching algorithm part of the LAS-X suite . All other images for tissue analyses were captured using a Zeiss AxioImager M2 microscope and AxioVision software version 4 . 8 . 2 ( Jena , Germany ) . The tissues were homogenized in TriPure reagent ( Roche , Indianapolis , IN , #11667165001 ) . Total RNA was extracted according to TriPure extraction protocol and treated with TURBO DNA-free Kit ( Thermo Fisher Scientific , Waltham , MA , AM1907 ) to eliminate all traces of viral DNA . The RNA concentration and integrity were assessed by Bioanalyzer 2100 ( Agilent , Santa Clara , CA ) . After removal of ribosomal RNA , the total RNA sequence libraries were prepared using Illumina Stranded Total RNA ( Illumina , RS-122-2201 , San Diego , CA ) protocol with TruSeq V4 chemistry and sequenced in the Sequencing Facility of NCI on Illumina HiSeq 2500 with 2×125 nts modality and depth of 100 million reads per sample . The obtained reads were trimmed of adapters and low-quality bases and aligned to MmuPV1 reference genome ( NC_014326; GI:301173443 ) with start site at nt 7088 using STAR aligner package [18 , 71] . This arrangement makes the linear MmuPV1 to end at nt 7087 , approximately 186 nts from L1 stop codon . Thus , all nucleotide positions described in this report refer to the reference genome sequence ( GenBank Acc . # NC_014326 ) [4] . Integrative Genomics Viewer ( IGV , Broad Institute ) program was used to visualize MmuPV1 reads coverage . The data discussed in this publication have been deposited in NCBI’s Gene Expression Omnibus and are accessible through GEO Series accession number GSE104118 ( www . ncbi . nlm . nih . gov/geo/query/acc . cgi ? acc=GSE104118 ) . Additional criteria were used to identify the real splice junctions extracted by STAR aligner: ( 1 ) a threshold of median spliced read alignment overhang >10 nt; ( 2 ) the number of uniquely mapping reads crossing the junction >50 to ensure filtering out sporadic false junctions; ( 3 ) entropy of overhang length distribution >1 to filter out junctions with unevenly distributed overhangs . A Sashimi plot for splice junction visualization was generated by IGV . The 5′ and 3′ RACE assays were carried out using a Smart RACE cDNA amplification kit ( Clontech , Mountain View , CA , #634858 ) according to the manufacturer’s instructions using 1 μg/reaction of total RNA as template [9] . The primers used in the assays are in S1 Table ( see supplemental materials ) . The final PCR products were gel purified , cloned into pCR2 . 1-TOPO vector ( Thermo Fisher Scientific ) and sequenced by Sanger sequencing ( Macrogen USA , Rockville , MD ) . To obtain the comprehensive coverage of viral TSS , the 5’ RACE products obtained by Pr3299 and Pr5452 were subjected to single molecule , real-time sequencing using PacBio Iso-seq technology ( Pacific Biosciences , Menlo Park , CA ) . The mouse papillomavirus amplicons produced using a 5’ RACE method were used as the input for PacBio Iso-seq sequencing . Using the SMRTbell Template Prep Kit 1 . 0 ( Pacific Biosciences ) , the cDNA underwent damage repair , A-tailing , and A/T hairpin adaptor ligation . Following adaptor ligation , the libraries were digested with Exonuclease III and VII to remove non-ligated and nicked SMRTbell molecules . AMPure PB beads ( Pacific Biosciences ) were used at a 0 . 6 x ratio to clean up all enzymatic reactions throughout library construction . The DNA/Polymerase Binding Kit P6 v2 ( Pacific Biosciences ) was used for annealing the sequencing primer and binding the polymerase to the final libraries . The MagBead loading Kit ( Pacific Biosciences ) was used for loading the polymerase-bound SMRTbell molecules onto the SMRT Cell . Each SMRT Cell was sequenced for six hours using the DNA Sequencing Reagent Kit 4 . 0 v2 ( Pacific Biosciences ) on a PacBio RS II sequencer . The obtained sequence reads were trimmed of adaptors and only the reads containing a specific adaptor at 5’ end introduced by 5’RACE were considered the full-length and used for mapping to MmuPV1 reference genome in further analysis and visualized in the IGV program . The position of 5’ends of the full-length reads were extracted and quantified in promotor usage analysis . PacBio SMRT Analysis Package ( smrtpipe . py v1 . 87 , with default settings ) was used to process raw data into circular consensus sequence ( CCS ) for further analyses . First , the raw data was processed into error corrected reads of insert ( ROI’s ) by RS_Read Of Interest ( ROI ) . 1 protocol provided in the SMRT Analysis Package . The ROI’s were then processed using the Classify module with default parameters to remove adapter sequences , poly ( A ) tails , artificial chimeras , and 3′ truncated transcript sequences which resulted into full-length non-chimeric ( FLNC ) reads by RS_IsoSeq . 1 protocol . For further analyses , we mapped the CCS reads and FLNC reads into the MmuPV1 genome and identified the transcription start sites by BLAT ( with-fine option otherwise default options ) [72] . Additional computational analyses were performed with Python ( version 3 . 5 , https://www . python . org/ ) . To remove contaminated genomic DNA , the total RNA was treated with TURBO DNA-free Kit ( Thermo Fisher Scientific ) . Reverse transcription ( RT ) was performed with the SuperScript II kit ( Thermo Fisher Scientific , #11904–018 ) . Amplification of reversed transcribed cDNA was performed by PCR using the Platinum SuperFi Taq Polymerase Kit ( Thermo Fisher Scientific , #12351–050 ) according to the manufacturer’s protocols . The MmuPV1-specific primers ( S1 Table , see supplemental material ) were used to detect viral transcripts . GAPDH RNA served as a sample loading control by using a mouse GAPDH-specific primer pair ( forward oMA1 , 5’-ATGTTCCAGTATGACTCCAC-3’ and backward oMA2 , 5’-TGACAATCTTGAGTGAGTTG-3’ ) . All PCR amplifications were performed under same conditions: on a primary denaturation step at 94°C for 2 min , followed by 25 or 35 cycles of 30 sec at 94°C , 45 sec at 55°C and 60 sec at 72°C , and final extension for 10 min at 72°C . Total RNA used for Northern blot analysis was isolated from mouse ears with or without MmuPV1 infection or extracted from regular HEK293 cells ( ATCC , Manassas , VA ) transfected with individual MmuPV1 ORF expression vectors . Total RNA from mouse ears without MmuPV1 subclinical infection served as a MmuPV1-negative RNA control and was pooled RNA isolated from ears of two naïve , freshly arrived female mice from Harlan lab in ~4 months of age , with no detectable MmuPV1 reads by RNA-seq analysis . In general , total 5 μg of RNA from each sample was mixed with NorthernMax Formaldehyde loading dye ( Thermo Fisher Scientific , #AM8552 ) ) and denatured at 75°C for 15 min . The RNA samples were then separated in 1% ( wt/vol ) formaldehyde-containing agarose gels in 1× morpholinepropanesulfonic acid ( MOPS ) running buffer , transferred onto a GeneScreen Plus hybridization transfer membrane ( Perkin Elmer , Waltham , MA , #NEF987001PK ) and UV light crosslinked and stained by ethidium bromide for 18S ribosome level as an internal loading control . The membrane was then prehybridized with PerfectHyb Plus hybridization buffer ( Sigma-Aldrich , #H7003 ) for 2 h at 42°C followed by overnight hybridization with a MmuPV1-specific oligo probe as described [73] . MmuPV1-specific oligo probes and a U6-specific probe ( oST 197 , 5’-AAAATATGGAACGCTTCACGA-3’ ) were prepared by end-labeling of antisense oligos ( S1 Table ) with γ-32P using T4 PNK ( Thermo Fisher Scientific , #18004–010 ) . After hybridization , the membrane was washed once with a 2× SSPE ( 1× SSPE: 0 . 18 M NaCl , 10 mM NaH2PO4 and 1 mM EDTA [pH 7 . 7] ) -0 . 1% SDS solution for 5 min at room temperature and twice with 0 . 1× SSPE-0 . 1% SDS for 15 min at 42° and then exposed to a PhosphorImager screen and X-ray film . The radioactive RNA probes were prepared by in vitro transcription in the presence of [α-32P]CTP with Riboprobe System-T7 ( Promega , Madison , WI , #P1446 ) , using PCR products with a built-in T7 promoter as DNA templates . The following primers were used for MmuPV-1 DNA template preparation: oXYX-12 and oXYX-23 ( S1 Table ) . The RNase protection assay ( RPA ) was performed with an RPA III kit ( Ambion , Austin , TX , #1414 ) according to the manufacturer’s instructions with minor modifications . Briefly , 4 ng of each probe ( specific activity , 35 , 000 cpm/ng ) was hybridized overnight at 50°C with 30 μg of total tissue RNA in hybridization buffer and then digested with an RNase A-T1 mixture for 30 min at 37°C . Five micrograms of yeast RNA was used as a negative control ( MmuPV1 - ) . Protected RNA fragments were separated in a denaturing 8% polyacrylamide gel containing 8 M urea . MmuPV-1 DNA sequencing ladders generated with the 32P-labeled Primer Pr7237 ( oXYX-28 ) ( S1 Table ) were used as size markers and run along with the RPA products as described [73] . Autoradiographic data were captured with a Typhoon Imaging System ( GE Healthcare Life Sciences , Pittsburgh , PA ) and analyzed with ImageQuant software ( GE Healthcare Life Sciences ) . For in situ detection of viral transcripts , the tissues were fixed in 10% neutral buffer formalin for 20 h at room temperature , dehydrated , and embedded in paraffin . The sections were cut into 5 μm slides and subjected to RNA-ISH using RNAscope technology ( Advanced Cell Diagnostics , Newark , CA ) as recommended by manufacturer . Two custom designed probes derived from MmuPV1 genome were used: E1^E4 ( nt 3139–3419 ) and L1 ( nt 5372–6901 ) . The signal was detected by colorimetric staining using RNAscope 2 . 5 HD Assay—BROWN followed by hematoxylin Gill’s No . 1 solution ( Sigma-Aldrich , #GHS116 ) counterstaining . The slides were dehydrated , mounted in Cytoseal XYL ( Thermo Scientific , #8312–4 ) , and scanned at 40× resolution using Aperio CS2 Digital Pathology Scanner ( Leica Biosystem , Buffalo Grove , IL ) . To distinguish viral RNA signal from viral genomic DNA signal , the MmuPV1-infected tissue sections with or without pre-treatment with DNase I or both DNase I and RNase A/T1 were compared in parallel in the RNAscope assays . To carry out DNase I or RNase treatment , all tissue sections after rehydration were digested first with RNAscope Protease Plus for 30 min and then followed by 20 units of DNase I ( Thermo Fisher Scientific , cat . No . #EN0521 ) diluted in 1 x reaction buffer with MgCl2 for 30 min at 40°C or by 20 units of DNase I and 500 ug of RNase A ( Qiagen , #1006657 ) plus 2000 units of RNase T1 ( Fermentas , Waltham , MA , #EN0542 ) diluted in 1 x reaction buffer with MgCl2 for 30 min at 40°C . To express viral proteins , the cDNAs of individual ORF under the optimized Kozak context were amplified by RT-PCR from total RNA isolated from infected tissues and cloned into pFLAG-CMV-5 . 1 ( Sigma-Aldrich ) vector in frame with a C-terminal FLAG tag . The obtained plasmid DNA ( 2 μg ) was utilized to transfect HEK293 cells ( 2 . 5 × 105 ) plated in a 12-well plate using LipoD293 transfection reagent ( SignaGen Laboratories , Rockville , MD , #SL100668 ) . In some cases the cells were treated 24 h after transfection with 10 μM proteasome inhibitor MG132 ( Sigma-Aldrich , #474790 ) for 6 h . The primary mouse keratinocytes were cultivated as described [74] in the presence of Rho kinase inhibitor Y-27632 ( Enzo Life Sciences , Farmingdale , NY , #ALX-270-333 ) . The mouse keratinocytes ( 1 × 105 ) were transfected with 1μg of plasmid DNA using Amaxa P3 Primary Cells 4D Nucleofector X Kit S ( Lonza , Walkersville , MD , #V4XP-3032 ) and program DS-138 as recommended by manufacturer . After transfection , the keratinocytes were plated into 24-well plate containing mitomycin-treated feeder 3T3 cells and incubated for 45 hours in the absence of Rho kinase inhibitor . Total protein extracts and total RNA were prepared 24 h after transfection of HEK293 cells and the expressed individual viral proteins were determined by Western blotting with a rabbit polyclonal anti-FLAG antibody ( Sigma-Aldrich , #F7425 ) . Total RNA was resolved on a 1% formaldehyde-agarose gel , stained by ethidium bromide for 18S ribosomal RNA as a sample loading control , and examined by Northern blotting for individual viral gene transcripts expressed from the transfected plasmid by a γ-32P-labeled oligo probe ( oVM79 , 5’-GGGCACTGGAGTGGCAAC-3’ ) which hybridizes to a common 3’ UTR region downstream of the FLAG-tag , but upstream of the poly ( A ) site . U6 snRNA served as a loading control and was detected using a γ-32P-labeled , U6-specific oligo probe oST197 . HeLa cells ( 2 . 5 × 105 , ATCC ) growing on the coverslips were transfected with each vector ( 0 . 5–1 μg ) encoding a FLAG-tagged viral protein by using LipoD293 transfection reagent ( SignaGen Laboratories ) . The cells at 24 h after transfection were fixed , permeabilized , and stained with a monoclonal anti-FLAG M2 antibody ( Sigma-Aldrich , # F1804 ) in combination with Alexa Flour488-labeled anti-mouse secondary antibody ( Thermo Fisher Scientific , #A11029 ) as described before [75 , 76] . The cell nuclei were counterstained by Hoechst 33342 dye ( Thermo Fisher Scientific , #H3570 ) . | Papillomavirus ( PV ) infections lead to development of both benign warts and cancers . Because PVs are epitheliotropic and species specific , it has been extremely challenging to study PV infection in the context of a naturally occurring infection in a tractable laboratory animal . The recent discovery of the papillomavirus , MmuPV1 , that infects laboratory mice , provides an important new animal model system for understanding the pathogenesis of papillomavirus-associated diseases . By using state of the art RNA-seq to provide deep sequencing analysis of what regions of the viral genome are transcribed and PacBio Iso-seq that produces longer reads to define the complete sequences of individual transcripts in combination with several conventional technologies to confirm transcription starts sites , splice sites , and polyadenylation sites , we provide the first detailed description of the MmuPV1 transcript map using RNA from MmuPV1-induced mouse warts . This study reveals the presence of mRNA transcripts capable of coding for ten protein products in the MmuPV1 genome and leads to correctly re-assigning the E1^E4 , L2 and L1 coding regions . We were able to detect individual transcripts from the infected wart tissues by RT-PCR , Northern blot and RNA ISH , to define the temporal onset of productive viral infection and to ectopically express a predicted viral protein for functional studies . The constructed MmuPV1 transcript map provides a foundation to advance our understanding of papillomavirus biology and pathogenesis . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"sequencing",
"techniques",
"dermatology",
"medicine",
"and",
"health",
"sciences",
"otology",
"ear",
"infections",
"genome",
"analysis",
"molecular",
"biology",
"techniques",
"mammalian",
"genomics",
"rna",
"sequencing",
"research",
"and",
"analysis",
"methods",
"genome",
"complexity",
"antisense",
"rna",
"gene",
"expression",
"warts",
"otorhinolaryngology",
"molecular",
"biology",
"animal",
"genomics",
"biochemistry",
"rna",
"polyadenylation",
"nucleic",
"acids",
"genetics",
"biology",
"and",
"life",
"sciences",
"genomics",
"gene",
"prediction",
"computational",
"biology",
"introns"
] | 2017 | The full transcription map of mouse papillomavirus type 1 (MmuPV1) in mouse wart tissues |
The potato late blight pathogen Phytophthora infestans secretes an array of effector proteins thought to act in its hosts by disarming defences and promoting pathogen colonisation . However , little is known about the host targets of these effectors and how they are manipulated by the pathogen . This work describes the identification of two putative membrane-associated NAC transcription factors ( TF ) as the host targets of the RxLR effector PITG_03192 ( Pi03192 ) . The effector interacts with NAC Targeted by Phytophthora ( NTP ) 1 and NTP2 at the endoplasmic reticulum ( ER ) membrane , where these proteins are localised . Transcripts of NTP1 and NTP2 rapidly accumulate following treatment with culture filtrate ( CF ) from in vitro grown P . infestans , which acts as a mixture of Phytophthora PAMPs and elicitors , but significantly decrease during P . infestans infection , indicating that pathogen activity may prevent their up-regulation . Silencing of NTP1 or NTP2 in the model host plant Nicotiana benthamiana increases susceptibility to P . infestans , whereas silencing of Pi03192 in P . infestans reduces pathogenicity . Transient expression of Pi03192 in planta restores pathogenicity of the Pi03192-silenced line . Moreover , colonisation by the Pi03192-silenced line is significantly enhanced on N . benthamiana plants in which either NTP1 or NTP2 have been silenced . StNTP1 and StNTP2 proteins are released from the ER membrane following treatment with P . infestans CF and accumulate in the nucleus , after which they are rapidly turned over by the 26S proteasome . In contrast , treatment with the defined PAMP flg22 fails to up-regulate NTP1 and NTP2 , or promote re-localisation of their protein products to the nucleus , indicating that these events follow perception of a component of CF that appears to be independent of the FLS2/flg22 pathway . Importantly , Pi03192 prevents CF-triggered re-localisation of StNTP1 and StNTP2 from the ER into the nucleus , revealing a novel effector mode-of-action to promote disease progression .
Lacking an adaptive immune system , plants have evolved a two-tier surveillance system to detect and deflect pathogen incursions . The first layer is triggered by receptor-like kinase pattern recognition receptors ( RLK-PRRs ) , which recognise conserved non-self molecules or pathogen associated molecular patterns ( PAMPs ) [1]–[3] . These PRRs enable plants to sense the proximity of potential pathogenic microbes and activate their defences accordingly , for example by production of reactive oxygen species , callose deposition and synthesis of antimicrobial compounds [4] . This is termed PAMP-triggered immunity ( PTI ) [5] and is thought to be effective in restricting the growth of the majority of would-be pathogens . However , dedicated plant pathogenic organisms have evolved the ability to secrete a range of effector molecules that suppress PTI . Some , such as bacterial type III effectors [6] and oomycete RxLR effectors [7]–[8] , are translocated inside plant cells and are believed often to attenuate PTI by manipulating host target proteins; this is termed effector triggered susceptibility ( ETS ) [5] . In response , a second layer of resistance has evolved in plants comprising resistance ( R ) proteins which detect effectors , or their activity , often resulting in a localised cell death or hypersensitive response ( HR ) . This is termed effector-triggered immunity ( ETI ) [5] , [9]–[11] . This model for plant-pathogen interactions is summarised for plant-oomycete interactions in Hein et al . [12] . However , to date , little is known about how oomycete effectors alter plant defence or metabolism to their own ends . Following the identification of oomycete RxLR effectors , and the demonstration that they are delivered inside plant cells [13]–[15] , effort has been directed at defining the repertoire of this class of effectors in the genomes of diverse oomycete species [16]–[18] . Yet little is known about the virulence function of these proteins within the plant host . Several relatively high throughput assays have been carried out to establish possible roles for RxLR effectors in the suppression of PTI and ETI [19]–[21] . Screens of Phytophthora sojae RxLRs found the majority of effectors were able to suppress cell death ( CD ) triggered by a range of CD-inducing elicitors in transient assays in Nicotiana benthamiana [20] . In addition , 70% of tested Hyaloperonospora arabidopsidis RxLRs were found to supress PTI by enhancing the growth of Pseudomonas in different ecotypes of Arabidopsis thaliana [21] . While such studies support the hypothesis that pathogens secrete effectors to manipulate plant defence , they do not shed light on the mechanisms behind this . Recently , large-scale matrix yeast-two-hybrid ( M2H ) screens were conducted with the aim of mapping plant protein-protein interaction networks and identifying the likely “hubs” or convergence points of these networks targeted for manipulation by plant pathogens [22]–[23] . Of >8000 screened Arabidopsis proteins 17 were found to interact with effectors from both Pseudomonas syringae and H . arabidopsidis , with 15 of these showing defence-associated phenotypes in plant knockout studies [23] , indicating that pathogens separated by 2 billion years of evolution have potentially developed similar strategies for manipulating their host . However , given that 165 putative host targets were identified in this screen [23] , the question of which interactions are relevant in planta , and the precise mode of action of these effectors remains open . The little that is known about the manipulation of plant targets by oomycete effectors has mainly concentrated on those RxLRs which are recognised by plant resistance proteins . One of the best studied oomycete RxLRs , AVR3a from Phytophthora infestans , was found to interact with and stabilise the host ubiquitin E3 ligase CMPG1 , interfering with its ability to mediate cell death in response to perception of pathogen elicitors at the host cell membrane [24]–[25] . Recently another P . infestans RxLR , AVR2 , was found to interact with the plant phosphatase BSL1 and this interaction appears to be recognised or “guarded” by the resistance protein R2 [26] . The RxLR effector Avrblb2 from P . infestans prevents secretion into the plant apoplast of the plant defence-related protease C14 [27] . In addition , the avirulence protein Avr3b from P . sojae is an ADP-ribose/NADH pyrophosphorylase and is predicted to modulate plant immunity through this activity [28] . A role in virulence has also been proposed for the P . infestans effector Avrblb1 ( IPIO1 ) through the disruption of plasma membrane/cell wall adhesions via its interaction with the lectin receptor kinase LecRK-I . 9 [29]–[30] . Studies of candidate RxLR effectors from H . arabidopsidis reveal that they localise to a number of subcellular locations , potentially indicating a diversity of host targets and modes of effector activity [31] . Of 49 candidate RxLRs , 33% were localised strictly in the plant nucleus and a further 33% were nucleo-cytoplasmic . The majority of the remaining RxLRs ( 26% ) were associated with the plant membrane trafficking network , most of which ( 18% ) were localised to the endoplasmic reticulum [31] . The large number of H . arabidopsidis candidate RxLR effectors localised to the plant nucleus is perhaps unsurprising; transcriptional changes are at the heart of PTI [6] , also many regulatory components of plant immunity that are active in the nucleus are trafficked there from a variety of subcellular locations [32]–[33] . In this study we demonstrate a novel virulence function for the P . infestans RxLR PITG_03192 ( Pi03192 ) . Yeast-2-hybrid and bimolecular fluorescence complementation ( BiFC ) analyses revealed two putative membrane-associated NAC ( NAM/ATAF/CUC ) transcription factors ( TFs ) as the potato host targets of Pi03192 . Interactions with these NAC TFs occur at the endoplasmic reticulum ( ER ) membrane . Virus induced gene silencing ( VIGS ) of the genes encoding these two putative TFs leads to an increase in susceptibility to P . infestans infection , suggesting these NAC TFs play a role in plant defence . Silencing of Pi03192 in P . infestans results in a reduction of virulence on potato and N . benthamiana . Critically , virulence of the silenced line on N . benthamiana can be enhanced by VIGS of either NTP1 or NTP2 . Both NAC targeted by P . infestans ( NTP ) 1 and NTP2 transcripts accumulate following treatment with culture filtrate ( CF ) from in vitro-grown P . infestans , which potentially contains a mixture of Phytophthora PAMPs and elicitors , but not following treatment with flg22 . In contrast , transcripts of both genes decrease during the early stages of P . infestans colonisation , suggesting that the pathogen manipulates transcription levels of these genes . On treatment with CF , but not flg22 , the putative ER membrane-associated NAC TFs , NTP1 and NTP2 translocate to the nucleus , where they are rapidly turned over by the 26S proteasome . We show that Pi03192 promotes virulence by preventing the nuclear accumulation of the two NAC TFs .
Candidate RXLR effector PITG_03192 ( Pi03192 ) , also named RD28 [19] , was one of the first predicted from P . infestans expressed sequence tags [14] . In keeping with the expression profiles of most RXLR effector genes , Pi03192 transcripts accumulate specifically during the first 3 days of P . infestans infection of potato [14] , [19] and tomato [19] . Nicotiana benthamiana is a model for functional studies in the Solanaceae , and is also a host for P . infestans . As such it has been extensively used to investigate pathogen and host gene functions in P . infestans-plant interactions [19] , [24] , [26]–[27] , [34] . To investigate whether Pi03192 acts to promote P . infestans colonisation the effector gene , minus signal peptide-encoding sequences , was expressed transiently in N . benthamiana . Figure 1 shows that , compared to expression of an empty vector control , Agrobacterium-mediated expression of Pi03192 inside N . benthamiana cells significantly enhances P . infestans colonisation by 9 days post-inoculation ( dpi ) . This suggests that Pi03192 , as a predicted translocated RxLR effector , acts within plant cells to promote ETS , prompting us to investigate its mode-of-action and potential host targets . To identify candidate host targets of P . infestans RxLR effectors a yeast-2-hybrid ( Y2H ) library composed of cDNA from potato plants infected with P . infestans [24] was screened with the candidate RxLR effector Pi03192 . After screening 8×106 yeast transformants , 16 clones expressing interacting proteins were identified . Sequences from 13 of 16 clones were found to encode the C-terminal region of a predicted membrane-associated NAC transcription factor ( TF ) , and the remaining 3 clones were found to encode the C terminus of a distinct predicted membrane-associated NAC TF ( Figure 2A and B ) . These NAC TF fragments interacted with Pi03192 but did not interact with another verified RxLR effector , PiAVR2 ( Figure 2A ) , which was used as a control . BLAST searches and alignments of ESTs from related solanaceous species tomato , potato and petunia were used to design primers to amplify the full-length potato NAC TF genes . Both genes were found to encode proteins containing a predicted N-terminal NAC DNA binding ( NAM ) domain and a predicted transmembrane ( TM ) domain at the C terminus of the protein ( Figure 2B ) . These genes were thus tentatively designated Solanum tuberosum NAC Targeted by Phytophthora ( StNTP ) 1 ( 13/16 clones ) and StNTP2 ( 3/16 clones ) . To further explore the relationship of StNTP1 and StNTP2 with NAC TFs from other plant species a phylogenetic tree was constructed by maximum likelihood based on the 337 most similar NAM domains extracted from plant sequences contained in the NCBI RefSeq sequence database ( Figure S1 and S2 ) . Both StNTP1 and StNTP2 are placed in clades with little or no representation of cereal NAC TF sequences . In each case the potato NAC sequences form a clade with NAC TFs from other Solanaceae , which are paired with a similarly expanded grouping of Arabidopsis thaliana NACs . In Arabidopsis thaliana 13 genes encoding proteins with an N-terminal NAM domains and C-terminal TM domains ( termed NTL1-13 ) have previously been reported , and classified into four phylogenetic subgroups [35] . The phylogenetic tree in Figure S1 differs from that proposed by Kim et al . [35] , and incorporates additional A . thaliana candidate NTLs not included in that analysis . This improved tree places StNTP1 in a clade with AtNTL6 , and StNTP2 in a clade that contains the three Arabidopsis NACs AtNTL1 , AtNTL3 and AtNTL7 ( Figure S1 and S2 ) Both clades contain sequences from a number of plant species , with or without predicted transmembrane domains at the C-terminal region . An alignment of the NAM domains of StNTP1 and StNTP2 with the 13 reported AtNTLs shows that the potato NAM domains contain the critical conserved residues required for DNA binding and for NAC TF dimerization ( Figure S3 ) . The subcellular localisations of StNTP1 , StNTP2 and Pi03192 were examined by Agrobacterium-mediated transient expression of each with N-terminal GFP fusions in N . benthamiana and imaging using confocal microscopy . Images of cells expressing each of the constructs revealed a network of fluorescence suggestive of localisation to the endoplasmic reticulum ( ER ) membrane in planta ( Figure 3A and B ) . To confirm the ER localisation of the RxLR effector , GFP-Pi03192 was co-expressed with an RFP-ER-tagged construct and imaged by confocal microscopy . Both were observed to co-localise to the ER network as shown by the merge of the green and red channels ( Figure 3A ) . The ER membrane localisation is consistent with the observation that both StNTPs possess predicted C-terminal TM domains . However , GFP-Pi03192 is also ER localised despite the absence of a predicted TM domain . Western blots hybridised with an antibody specific to GFP demonstrated that the GFP-StNTP1 , GFP-StNTP2 and GFP-Pi03192 fusion proteins were stable , showing bands of the predicted sizes , respectively 111 kDa , 106 kDa and 39 kDa ( Figure 3C ) . As NAC TFs are expected to localise to the nucleus following release from the ER , N-terminal fusions of GFP to StNTP1 and StNTP2 lacking the predicted TM domains were made . Surprisingly , confocal images showed no fluorescence of GFP-StNTP1ΔTM and GFP-StNTP2ΔTM ( Figure S4A ) . However , literature regarding membrane-associated transcription factors ( MTFs ) suggests that many are tightly regulated and rapidly turned over by the 26S proteasome [36]–[38] . When leaves expressing the GFP-StNTPΔTM constructs were treated with the proteasome inhibitor MG132 fluorescence was clearly observed , as anticipated , exclusively in the nucleus ( Figure S4A ) . Immunoblots to indicate fusion protein stability could only detect GFP-StNTPΔTMs in the presence of the proteasome inhibitor ( Figure S4B ) . This suggests that activity of these StNTPs is regulated in planta , at least in part , by rapid protein turnover . To investigate further the potential interaction between Pi03192 and either StNTP1 or StNTP2 in planta , bimolecular fluorescence complementation ( BiFC ) was utilised . The non-interacting P . infestans RxLR effector PiAVR2 was used as a negative control , as it has been shown using BiFC to interact in the host cytoplasm with a different plant protein , StBSL1 [26] . These constructs were co-expressed in N . benthamiana plants and examined using confocal microscopy . YFP fluorescence was observed when YN-Pi03192 was co-expressing with either YC-StNTP1 or YC-StNTP2 . However , no noticeable fluorescence was observed when co-expressing either YC-StNTP1 or YC-StNTP2 with YN-PiAVR2 ( Figure 4A ) . The fluorescence of YN-Pi03192 co-expressed with either YC-StNTP1 or YC-StNTP2 was observed to be at the ER ( Figure 4A , Figure S5 ) , showing that they are in close proximity in this location , consistent with co-localisation of these proteins ( Figure 3 ) . Fluorescence was subsequently quantified in leaf disks co-expressing each construct , and showed significantly increased fluorescence ( P≤0 . 001 , One way ANOVA ) when co-expressing YC-StNTP1 or YC-StNTP2 with YN-Pi03192 , compared to YC-StNTP1 or YC-StNTP2 with YN-PiAVR2 ( Figure 4B ) . Immunoblots of each construct used for SplitYFP show that all fusion proteins were stable in planta and of the predicted size ( Figure 4C ) . Therefore , protein instability does not account for the reduced SplitYFP fluorescence observed when using PiAVR2 , as opposed to Pi03192 . As Pi03192 was found to enhance P . infestans colonisation of the model host N . benthamiana ( Figure 1 ) , N . benthamiana was employed further to explore the roles of NTP1 and NTP2 in defence responses to this pathogen . Initially , PCR primers designed to amplify full-length StNTP1 and StNTP2 from potato ( Supplementary Table S1 ) were used to amplify homologous genes from N . benthamiana . This resulted in two genes with 86% and 87% identity at the nucleotide level , respectively , to StNTP1 and StNTP2 . These sequences were annotated in the N . benthamiana genome [39] as NbS00058586g0005 . 1 and NbS00026810g0011 . 1 . NbS00058586g0005 . 1 was found to be the reciprocal best BLAST match to StNTP1 in the N . benthamiana genome , and was thus renamed NbNTP1; similarly , NbS00026810g0011 . 1 was the reciprocal best BLAST match to StNTP2 , and renamed NbNTP2 . The NAM domains of NbNTP1 and NbNTP2 protein sequences are placed in the same clades as StNTP1 and StNTP2 , respectively , in our phylogenetic reconstruction ( Figure S1–S3 ) . To assess the effects of NbNTP1 and NbNTP2 on basal resistance to P . infestans , two independent Virus Induced Gene Silencing ( VIGS ) constructs were designed to specifically silence each gene independently in N . benthamiana . Different portions of the NbNTP genes were cloned in antisense into the pTRV2 Agro binary vector [40] to create silencing constructs pTRV-NTP1 I and II , and pTRV-NTP2 I and II ( Figure S6 ) . Agrobacterium strains containing these constructs were infiltrated into 2-week-old N . benthamiana seedlings , and plants were left for 2-to-3 weeks to allow silencing to become systemic . Gene silencing levels were checked using quantitative real time ( qRT ) -PCR . Both pTRV-NTP1 VIGS constructs were observed to knock down NbNTP1 expression by 80–90% and pTRV-NTP2 constructs knocked down NbNTP2 expression by 70–90% compared to gene expression in the TRV-GFP control plants ( Figure S6 ) Both pTRV-NTP1 VIGS constructs did not alter NbNTP2 transcript levels , and pTRV-NTP2 constructs did not alter NbNTP1 transcript levels . To investigate effects of silencing at the protein level , GFP-StNTP1 and GFP-StNTP2 fusions were transiently expressed in the NTP1 and NTP2 silenced plants and the levels of the GFP fusion proteins were examined by immunoblot analyses . Using a GFP antibody GFP-StNTP2 but not GFP-StNTP1 could be strongly detected in NTP1-silenced plants . In contrast , GFP-StNTP1 , but not GFP-StNTP2 protein was strongly detected in NTP2-silenced plants ( Figure S6 ) . This demonstrates that the constructs designed to silence NbNTP1 also silence StNTP1 , but not StNTP2 , whereas the constructs designed to silence NbNTP2 also silence StNTP2 , but not StNTP1 . Moreover , reduction in transcript levels was seen to lead to specific reduction in levels of the corresponding protein . Having shown that both sets of TRV-NTP1 and TRV-NTP2 VIGS constructs silence NTP1 and NTP2 , respectively , in N . benthamiana , we investigated whether P . infestans colonisation was enhanced by silencing them . Plants expressing these VIGS constructs , or the TRV-GFP control construct , were infected with a transgenic P . infestans strain expressing tandem dimer Tomato ( tdT ) [26] . TdT fluorescence was examined by confocal microscopy . We noted that colonisation was significantly delayed in TRV-GFP control plants , compared to plants lacking TRV ( Figure S7A ) . At early stages of infection ( 3 dpi ) , extensive intercellular mycelial growth indicative of spreading lesions was observed on TRV-NTP1 and TRV-NTP2 plants at 20–40% of inoculation sites . In contrast , at 3 dpi on TRV-GFP control plants infections were largely restricted to leaf-surface growth of germinating sporangia and , in some cases , production of invasive hyphae indicative of initial pathogen colonisation ( Figure 5A and B , Figure S7B ) . Thus , silencing either NTP1 or NTP2 promoted more rapid initial colonisation of N . benthamiana by P . infestans . The increase in susceptibility following TRV-NTP-mediated VIGS was quantified by counting the percentage of P . infestans drop inoculation sites that showed sporulating lesions at 7 dpi compared to the TRV-GFP control , which was used as a base line . Figure 5C shows an increase of between 10–20% in the number of P . infestans sporulating lesions on the TRV-NTP VIGS plants compared to the TRV-GFP control ( P<0 . 05; two tailed t-test ) . Figure 5D shows that , by 10 dpi , more P . infestans growth and sporulation was visible in NTP1- and NTP2-silenced plants than in the TRV-GFP-expressing control . The increase in susceptibility to P . infestans in TRV-NTP VIGS plants was measured at 10 dpi by quantifying sporangia harvested from leaves expressing each construct . Significantly more sporangia were recovered from TRV-NTP silenced plants than the TRV-GFP control plants ( Figure 5E ) . The increase in susceptibility of NTP1 and NTP2 silenced plants to P . infestans infection suggests that these genes are involved in conferring basal resistance to this pathogen . Therefore , it seems logical for P . infestans to produce effectors , such as Pi03192 , to interfere with the functions of these proteins . To examine the role of Pi03192 in P . infestans virulence a stably silenced transgenic line , hereafter referred to as 03192_IR , was made by transforming strain 88069 with the Pi03192 gene cloned as an inverted repeat . QRT-PCR was employed to determine the levels of silencing in germinating cysts . Figure 6A shows that 03192_IR is specifically silenced for Pi03192 expression and that the transcript accumulation of other RXLR genes with published roles in P . infestans-host interactions ( AVR3a , AvrBlb2 , and AvrBlb1 [ipiO1] ) [24] , [27] , [30] is unaffected . Compared to wildtype P infestans isolate 88069 , silenced line 03192_IR is less able to colonise both potato and N . benthamiana plants ( Figure 6B–D ) and forms both fewer and smaller lesions , indicating that Pi03192 function is important for full pathogenicity on these host plants . In N . benthamiana , colonisation by the 03192_IR line was restored to wild-type levels by transient agro-expression of Pi03192 inside host cells , compared to an empty vector control ( Figure 6E ) . When 03192_IR was inoculated onto either NTP1 or NTP2 silenced plants a significant increase in virulence was observed compared to growth on TRV::GFP controls . Figure 6F–G shows a significant increase in lesion size and increased symptom development , respectively , of 03192_IR on NTP1 and NTP2 silenced plants . This supports further the interactions of both NTP1 and NTP2 with Pi03192 and suggests Pi03192 exerts its virulence function by interfering with both NTP proteins , as silencing of either NTP1 or NTP2 using VIGS enhances pathogenicity of the compromised 03192_IR isolate . QRT-PCR was used to investigate expression of NbNTP1 and NbNTP2 across time-courses of P . infestans infection . Pi03192 expression was measured during P . infestans infection of N . benthamiana and found to be up-regulated at 16 and 24 hours post-inoculation ( hpi ) compared to the levels of expression in sporangia samples ( Figure 7A ) . The observed expression at 16 and 24 hpi time-points are consistent with previous observations of up-regulation of RXLR effector genes , generally , and Pi03192 specifically , during potato and tomato colonisation [14] , [19] and coincide with initial host cell penetration and formation of haustoria during the biotrophic phase [41] . QRT-PCR analysis of the same time-course samples showed that transcript accumulation of NbNTP1 and NbNTP2 co-ordinately decreased 2- to 6-fold in response to P . infestans infection ( Figure 7B ) . The expression levels of Pi03192 , StNTP1 and StNTP2 showed similar patterns during potato-P . infestans infection , demonstrating that the transcription of effector and NTP genes responds similarly in each pathosystem ( Figure S8 ) . We next investigated expression of NbNTP1 and NbNTP2 after infiltration of N . benthamiana leaves with culture filtrate ( CF ) from in vitro-grown P . infestans , which we selected because it potentially contains a range of Phytophthora PAMPs . To confirm this , qRT-PCR analysis was conducted on a range of PTI marker genes in N . benthamiana treated with either flg22 or P . infestans CF . The marker genes NbPTI5 and NbACRE31 [42] and NbWRKY7 and NbWRKY8 [43] showed similar patterns of up-regulation in response to either flg22 or CF treatments ( Figure S9 ) , consistent with our hypothesis that P . infestans CF likely contains Phytophthora PAMPs . The lower levels of up-regulation of these PTI marker genes with CF is likely because the P . infestans molecules acting to trigger this response are highly diluted compared to the defined flg22 peptide treatment . Similar to NTP1 and NTP2 , transcript accumulation of all 4 PTI marker genes decreased significantly during infection ( Figure S9 ) . In contrast to the observed decrease in NTP expression during infection , both NbNTP1 and NbNTP2 show coordinate 30- to 40-fold increases in transcript abundance measured at 3 hpi , with mRNA levels declining thereafter through the 48 hour time-course when the plant was treated with P . infestans CF ( Figure 7C and D ) . Treating the plant with the media used to grow P . infestans for CF preparation did not induce NbNTP expression ( Figure 7C and D ) . Interestingly , NbNTP1 and NbNTP2 transcripts failed to accumulated upon flg22 treatment ( Figure S10 ) . Taken together , these expression profiles distinguish flg22 and CF treatments , and suggest that NbNTP1 and NbNTP2 are up-regulated by a P . infestans-derived molecule in CF which activates a pathway that differs from the flg22/FLS2 PTI pathway . Both NbNTP genes show a similar pattern of expression in the different treatments , suggesting that they are co-ordinately regulated . Membrane-bound NAC transcription factors ( NAC-MTFs ) are often associated with rapid transcriptional changes in response to biotic and abiotic challenges , due to stress-triggered cleavage of the TM domain , allowing the active TF to re-localise to the nucleus [35] , [38] . As StNTP1 and StNTP2 have an ER membrane localisation and show elevated mRNA abundance in response to CF treatment , we hypothesised that application of CF would trigger their release from the ER membrane and lead to nuclear accumulation . To test the impact of Pi03192 effector on nuclear re-localisation of StNTP proteins , either a pFlub empty vector control ( Vec ) or pFlub-Pi03192 were co-expressed with either GFP-StNTP1 or GFP-StNTP2 . The pFlub vector contains a dual cassette expressing an untagged gene and an RFP-peroxisome tagged construct , which allows visualisation of cells ( i . e . containing peroxisomes tagged with RFP ) that also express an untagged gene of interest , in this case Pi03192 . N . benthamiana plants co-expressing either GFP-StNTP1 or GFP-StNTP2 with pFlub ( empty vector control ) were treated with either media or CF and examined using a confocal microscope . No change in localisation of fluorescence was observed , and both treatments showed GFP fluorescence associated only with the ER membrane surrounding the nucleus , albeit fluorescence was reduced following CF treatment ( Figure S11 ) . Accumulation of nuclear GFP-StNTP fluorescence was not evident following CF treatment ( Figure 8A ) . However , earlier experiments with the GFP-StNTPΔTM fusions demonstrated that nuclear NTP1 and NTP2 are turned over by the 26S proteasome ( Figure S4 ) . Therefore , samples were simultaneously treated with CF and proteasome inhibitor MG132 to block degradation . Confocal images following treatment with both CF and MG132 showed accumulation of both GFP-StNTP1 and GFP-StNTP2 inside the nucleus . In contrast , treatment with the medium used to grow P . infestans for CF preparation , alongside MG132 , did not result in nuclear accumulation of either StNTP protein ( Figure 8A ) . Moreover , whereas CF treatment with MG132 resulted in significant accumulation of GFP-NTP1 and GFP-NTP2 in nuclei , no such accumulation was observed following flg22 treatment with MG132 ( Figure S12 ) . This demonstrates that P . infestans CF treatment provides a specific trigger , which differs from that which activates the flg22/FLS2 PTI pathway , to release StNTP proteins from the ER membrane and allow their translocation inside the nucleus , after which they are turned over through the action of the 26S proteasome . To examine the effect of the RXLR effector Pi03192 on this process , leaves of N . benthamiana were infiltrated with Agrobacterium strains co-expressing GFP-StNTP1 or GFP-StNTP2 with pFlub-Pi03192 . Confocal images of these samples , treated with either CF alone or CF plus proteasome inhibitor MG132 , show that GFP-StNTP1 and GFP-StNTP2 fluorescence no longer accumulates inside the nucleus . Instead , even with MG132 treatment , GFP-StNTP1 and GFP-StNTP2 fluorescence was only detected in association with the ER , consistent with the interpretation that Pi03192 prevents release of GFP-StNTPs from the ER membrane ( Figure 8A ) . To confirm this result , a quantitative analysis was carried out on a series of confocal images of samples subjected to each of the treatments where the percentage of nuclei containing GFP-StNTP fluorescence was recorded . Figure 8B shows that the percentage of nuclei containing GFP-StNTP1 or GFP-StNTP2 fluorescence , co-expressed with empty pFlub vector and treated with CF plus MG132 , is significantly increased compared to all other treatments ( P≤0 . 001 ) , which do not differ significantly from each other . Immunoblots were performed to assess turnover of StNTP1 and StNTP2 when treated with CF , in the presence or absence of MG132 or Pi03192 . Figure 8C shows that , when co-expressed with the pFlub vector control , the detection of both GFP-StNTP1 and GFP-StNTP2 decreases upon CF treatment , consistent with the observed reduction in GFP fluorescence ( Figure S11 ) . This is likely due to release of the StNTP protein from the ER and turnover by the proteasome , whereas turnover is prevented in the presence of MG132 . Critically , when treated with CF , turnover of GFP-StNTP1 and GFP-StNTP2 is also prevented by co-expression with Pi03192 even in the absence of MG132 ( Figure 8C ) . Our results suggest it is likely that Pi03192 prevents release of StNTP1 and StNTP2 from the ER , rather than the effector preventing entry of these TFs into the nucleus . To further investigate this , we co-expressed pFlub vector ( control ) or pFlub-Pi03192 with GFP-StNTP1ΔTM or GFP-StNTP2ΔTM , and treated with MG132 . Pi03192 did not prevent nuclear accumulation of either GFP-StNTPΔTM fusion protein , indicating that the effector does not prevent their entry into the nucleus ( Figure S13 ) . We conclude that release of StNTP1 and StNTP2 from the ER , leading to them entering the nucleus , is prevented by the action of Pi03192 .
Little is known about the identities of host proteins that are targeted by effectors from filamentous plant pathogens such as oomycetes and fungi . Less is known about the roles those targets play in plant immunity , or other roles detrimental to disease progression , and less still about the modes of action of effector proteins upon such targets . Here we show that the P . infestans RXLR effector Pi03192 enhances infection when expressed transiently inside N . benthamiana cells . Pi03192 interacts with the C-terminal portions of two ER-associated potato NAC transcription factors , called StNTP1 and StNTP2 . Interaction in planta occurs at the ER . VIGS of the equivalent NbNTP1 and NbNTP2 genes in N . benthamiana leads to accelerated biotrophic colonisation , and increased sporulation in later stages of infection by P . infestans , compared to unsilenced plants . A stable transgenic P . infestans line ( 03192_IR ) silenced for Pi03192 expression shows decreased virulence on both potato and N . benthamiana , indicating that the effector has a non-redundant function and is essential for full virulence . Virulence of the 03192_IR line was restored to WT levels by transient in planta expression of Pi03192 , and was significantly enhanced on N . benthamiana plants that were VIGSed to knock down expression of either NTP1 or NTP2 , indicating that knock-down of an essential virulence component can be complemented by knock-down of its target ( s ) . Critically , we provide evidence that the mode-of-action of Pi03192 is to prevent PAMP-triggered re-localisation of StNTP1 and StNTP2 from the ER into the nucleus . Each of these observations is discussed below . Thirteen membrane-associated NAC TFs with putative C-terminal transmembrane ( TM ) domains [35] , [44] , termed NTM1-like ( NTL ) proteins , have been reported in Arabidopsis . Members of this family contribute rapid responses to biotic and abiotic stresses . NTL4 promotes production of reactive oxygen species in response to drought [45] . NTL8 is involved in GA-mediated signalling following salt stress [46] . NTL6 promotes SA-independent activation of defences , including up-regulation of PR genes , upon cold treatment [47] . NTL11 ( also called RPX ) has been shown to positively regulate the 26S proteasome [48] and NTL9 regulates leaf senescence in response to osmotic stress [49] . We find from searches of the RefSeq sequence database that there are more than 13 NTL genes ( proteins predicted to contain at least one NAM domain , and a transmembrane domain at the C-terminal region ) in A . thaliana ( Figure S1 and S2 ) . NTL9 was found to be a common interacting protein of RXLR effectors from Hyaloperonospora arabidopsidis ( Hpa ) and T3SS effector HopD1 from Pseudomonas syringae pv . tomato . Knockout of NTL9 in A . thaliana resulted in increased susceptibility to Hpa [23] . Comparison of NAM domains indicates that StNTP1 and three other potato NAC proteins with predicted TM domains are placed in the same clade as Arabidopsis NTL6 , while StNTP2 and a further three potato NAC proteins with predicted TM domains are found in a clade with Arabidopsis NTL1 , NTL3 and NTL7 ( Figures S1 , S2 ) . Both of these placings have exceedingly strong bootstrap support . It is not , however , possible to associate either StNTP1 or StNTP2 with a specific counterpart protein in Arabidopsis on the basis of this tree . NAM domains are expected to bind to specific regions of DNA , and so express the regulatory component of NAC TF function . Reconstruction of phylogeny on the basis of this protein domain may therefore reflect conservation of regulatory targeting , as the evolution of NAM domains will be constrained by the need to bind specific DNA sequence sites . Approximately half of the 337 sequences used to construct the tree were predicted to have TM domains at their C-terminus . Several clades containing sequences from a wide range of plant species are homogeneous for the presence of such a predicted domain . This suggests a general conservation across plant species of both a specific nuclear DNA target , and membrane association that is possibly a mechanistic requirement for rapid regulatory response by release from subcellular membranes . We note that the reconstructed phylogeny decomposes into a set of 40 distinct clades with 100% bootstrap support at their distal node , but that each such clade may contain multiple proteins from the same plant , either with or without a predicted TM domain . Most such clades contain either monocot or dicot plant proteins , but not both . Several clades contain a notable overrepresentation of proteins from Arabidopsis , with paired sequences from A . thaliana and A . lyrata , suggesting an expansion of this family in those species . Further work beyond the scope of this current study is required to determine experimentally whether proteins with similar NAM domains share common or overlapping promoter binding sites , or regulate expression of similar gene targets . Both NbNTP1 and NbNTP2 genes were co-ordinately up-regulated by treatment with CF from in vitro grown P . infestans , which we demonstrate activates early PTI-responsive marker genes in N . benthamiana . We therefore infer that CF potentially contains a mixture of Phytophthora PAMPs . Nevertheless , NbNTP1 and NbNTP2 transcripts did not accumulate following treatment with the defined PAMP , flg22 . Thus , up-regulation of NbNTP1 and NbNTP2 follows perception of a P . infestans-derived molecule in CF which activates a pathway independent of the FLS2/flg22 PTI pathway . During P . infestans host colonisation , transcript abundance of NbNTP1 and NbNTP2 ( Figure 7 ) , and of the PTI marker genes NbPTI5 and NbACRE31 [42] and NbWRKY7 and NbWRKY8 [43] ( Figure S9 ) , was sharply reduced , indicating that the active presence of the pathogen counteracts their induction . This indicates that the pathogen actively suppresses PTI . Moreover , the down-regulation of NbNTP1 and NbNTP2 indicates the potential contribution of these NAC TFs to preventing late blight disease . VIGS of either NbNTP1 or NbNTP2 resulted in enhanced susceptibility to P . infestans , confirming that their protein products reduce disease progression ( Figure 5 ) . The fact that VIGS of either NbNTP1 or NbNTP2 led to enhanced P . infestans colonisation implies that their roles in immunity are either functionally non-redundant , or that they act together , perhaps as heterodimers , and are thus inter-dependent . Hetero- and homo-dimerisation of NAC TFs has been demonstrated [50] , and we observe that the sites for dimerization are conserved in the NAM domains of both StNTP1 and StNTP2 ( Figure S3 ) although the contribution of the C-terminal domains is not clear . It is therefore at least plausible that these proteins act as a heterodimer , and this may explain why VIGS of either NbNTP1 or NbNTP2 alone enhances P . infestans colonisation . A transgenic P . infestans line was generated in which Pi03192 was silenced , but not other effectors which have demonstrated roles in virulence , such as AVR3a , AvrBlb1 , and AvrBlb2 [24] , [27] , [30] . Remarkably , the virulence of this silenced line , which was compromised on both potato and N . benthamiana , was significantly enhanced on the latter not only when Pi03192 was transiently expressed in plant cells , but also when either NbNTP1 or NbNTP2 were silenced by VIGS . The restoration of virulence following VIGS of either gene again suggests that the roles of NTP1 and NTP2 may be inter-dependent , or overlap , perhaps explaining why both are targeted by the Pi03192 effector . A NAC TF called TIP has been reported to interact with the coat protein of Turnip Crinkle Virus ( TCV ) , and this interaction was required to elicit the hypersensitive response triggered by HRT resistance protein [51] . GFP-TIP was found to localise to the nucleus in the absence of TCV . Notably , the presence of TCV coat protein resulted in exclusion of TIP from the nucleus , compromising its ability to regulate defence responses to TCV [52] . The authors postulated that exclusion of TIP from the nucleus by TCV coat protein was detected by HRP [52] . This indicates the potential for pathogens to interfere with the appropriate localisation of NAC TFs , and for the plant to ‘guard’ against such activity . To investigate the mode-of-action of Pi03192 , we studied re-localisation of StNTP1 and StNTP2 into the nucleus following treatment with P . infestans CF . As both TFs are turned over in the nucleus by the 26S proteasome , these experiments were conducted in the presence of the proteasome inhibitor MG132 . We showed that treatment with CF and MG132 resulted in a clear , measurable accumulation of GFP-StNTP1 and GFP-StNTP2 fluorescence in the nucleus . No such accumulation was observed following co-treatment with MG132 and the medium used for P . infestans growth to prepare CF , or by treatment with flg22 , demonstrating that CF contains a specific trigger from the pathogen that causes release of NTP1 and NTP2 from the ER in order to enter the nucleus . Critically , in the presence of Pi03192 , no increase in nuclear accumulation of GFP-StNTP1 or GFP-StNTP2 was observed following CF treatment , even when these plants were also treated with MG132 ( Figure 8 ) . GFP fluorescence was instead confined to the ER , indicating that the effector has either prevented release of the NTP TFs , or has prevented their entry into the nucleus . The observation that Pi03192 did not prevent accumulation of GFP-StNTP1ΔTM or GFP-StNTP2ΔTM following MG132 treatment , suggests that the effector does not inhibit entry into the nucleus . Moreover , the original Y2H screen revealed that Pi03192 interaction was maintained with just the C-termini of StNTP1 and StNTP2 , containing the TM domains . We conclude that Pi03192 prevents release of the NTP TFs from the ER . Immunoblots of the critical experiments showed that , in the presence of CF alone , the detection of both GFP-StNTP1 and GFP-StNTP2 was reduced , consistent with the 26S proteasome acting to degrade them when they are released from the ER to enter the nucleus . In agreement with this , treatment with both CF and MG132 restored stability of GFP-StNTP1 and GFP-StNTP2 . Crucially , in the presence of Pi03192 both GFP-StNTP1 and GFP-StNTP2 were strongly detected in the immunoblot following treatment with CF alone , showing that stability is restored by the effector in the absence of MG132 . Previous studies of Arabidopsis NTLs have shown that they are proteolytically cleaved from the ER so that they can enter the nucleus [35] , [37] , [44] . We were , however , unable to detect smaller products from GFP-StNTP1 and GFP-StNTP2 when treated with both CF and MG132 . It is possible that cleavage is sufficiently close to the TM domain , which represents only the last 18–29 amino acids of the C-termini of each , respectively , so as not to alter detectably the size of these proteins on a gel . Nevertheless , the failure to detect fluorescence from either of the GFP-NTP TFs in the nucleus when co-expressed with Pi03192 and treated with both CF and MG132 demonstrates that the effector has acted to prevent their entry into the nucleus where they can be turned over by the proteasome . The nucleus is a centre of activity for both PTI and ETI , and many critical regulators of either are trafficked there from various subcellular locations following pathogen perception [32]–[33] . However , few plant pathogen effectors have been shown to influence such re-localisation events and , indeed , few have been shown to target transcriptional regulators of plant immunity . Within the nucleus itself , host transcriptional reprogramming has been shown to be affected by transcriptional activator-like ( TAL ) effectors from Xanthomonas spp , whereas the type III effector XopD interacts directly with the host TF AtMYB30 to prevent its activity [reviewed in 32–33] . Here , we show that the P . infestans RXLR effector Pi03192 targets two NAC proteins , NTP1 and NTP2 , and prevents them from being released from the ER to enter the nucleus , where they contribute to prevent disease progression by this oomycete pathogen .
Y2H screening was performed using the ProQuest system ( Invitrogen ) . Briefly , DNA binding domain “bait” fusions were generated by recombination between pDonr201-03192 and pDEST32 , generating pDest32-03192 . This construct was transformed into yeast strain MaV203 , and nutritional selection used to recover transformants . A single transformant was grown up and used to prepare competent yeast cells , which then were transformed with a potato Y2H "prey" library , commercially prepared from P . infestans infected leaf material at 15 and 72 hpi . Candidate interacting preys were confirmed by retransformation with the 03192 bait construct or with a pDest32-AVR2 control to rule out the possibility that the observed reporter gene activation had resulted from interactions between the prey and the DNA binding domain of the bait construct or DNA binding activity of the prey itself . Nicotiana benthamiana seedlings were grown in individual pots in a glasshouse at 22°C ( 16 hours light , 8 hours dark ) , with 130–150 µE m−2 s−1 light intensity and 40% humidity . Agrobacterium tumefaciens strain AGL1 virG was used for all Agrobacterium transient expression experiments and was cultured for 24–48 h in Luria Broth at 28°C supplemented with the appropriate antibiotics , spun at 4000 rpm and resuspended in 10 mM MgCl2∶10 mM MES buffer with 200 µM Acetosyringone to OD600 = 0 . 3 for Western Blot or 0 . 01–0 . 1 for confocal imaging . Full length StNTP genes were cloned from potato cDNA with gene specific primers modified to contain the Gateway ( Invitrogen ) attB recombination sites . PCR products were purified and recombined into pDONR201 ( Invitrogen ) to generate entry clones via BP reactions using Gateway technology ( Invitrogen ) . Pi03192 and PiAvr2 were cloned into pDONR201 in the same way . Primer sequences are shown in Supplementary Table S1 . Protein fusions were made by recombining the entry clones with the following plant expression vectors using LR clonase ( Invitrogen ) . N terminal GFP fusions were made by recombining the entry clones with pB7WGF2 [53] . Split YFP constructs were made by recombining the entry clones for Pi03192 and PiAvr2 with pCL112 to generate YN protein fusions and StNTP1 and 2 were recombined with pCL113 to generate YC protein fusions [24] . Pi03192 was recombined with pFlub4 , a homemade variant of pMDC32 [54] where the HygR cassette was replaced with a Peroxisome mRFP tagged expression construct . Virus induced gene silencing ( VIGS ) constructs were made by cloning 250 bp PCR fragments of NTP1 and 2 from N . benthamiana cDNA and cloning into pBinary Tobacco Rattle Virus ( TRV ) vectors [40] between HpaI and EcoRI sites in the antisense orientation . A TRV construct expressing GFP described previously was used as a control [55] . Primer sequences are shown in Supplementary Table S1 . The two largest leaves of four leaf stage N . benthamiana plants were pressure infiltrated with LBA4404 A . tumefaciens strains containing a mixture of RNA1 and each NAC VIGS construct or the GFP control at OD600 = 0 . 5 each . Plants were used for assays or to check gene silencing levels by qRT-PCR 3 weeks later . A . tumefaciens containing each construct was pressure infiltrated into leaves of 4-week-old N . benthamiana . Cells expressing fluorescent protein fusions were observed using a Leica TCS-SP2 AOBS confocal microscope between one or two days post-infiltration . Images were obtained using an HCX APO 40×/0 . 9w water dipping lens . mRFP was imaged using an excitation wavelength of 568 nm from a ‘lime’ diode laser with emissions collected between 600 and 630 nm . GFP was imaged using 488 nm excitation from an argon laser , with emissions collected between 500 and 530 nm . Split-YFP was imaged using 514 nm excitation from an argon laser with emissions collected between 530–575 nm . Treatments were infiltrated 24 h after agro-infiltration of the constructs and leaves were observed under the confocal microscope 4 h later . Treatments are Media ( Pea Broth ) , CF ( P . infestans culture filtrate ) , MG132 ( 100 µM MG132 diluted in culture filtrate ) . For Pi03192 treatments GFP-NAC constructs were co-infiltrated with either pFlub-empty vector or pFlub-Pi03192 and subsequently treated with CF or MG132 as above . P . infestans strain 88069 expressing tandem dimer Tomato ( tdT ) fluorescent protein was imaged in planta using 10× dry lens and excited with 561 nm and emissions were collected between 570–600 nm using a lime diode laser . Quantification of fluorescence was performed using a SpectraMax M5 fluorimeter ( Molecular Devices ) . Leaf disks were cut from N . benthamiana leaves at 2 dpi Agro-infiltration and floated abaxial side up on H2O in 24-well plates readings were taken using a well scanning program taking reads from above . Softmax Pro software ( Molecular Devices ) was used to collect data . YFP fluorescence was excited at 514 nm and measured at 580 nm . GFP fluorescence was excited at 480 nm and measured at 520 nm . Leaf disks were harvested at 2 dpi after Agrobacterium infiltration with constructs expressing either GFP-NAC1 or 2 , GFP-Pi03192 , YN_PiAvr2 , YN_Pi03192 , YC_NAC1 , YC_NAC2 and were either untreated , or treated with one of the following: Buffer 3hpi ( 10 mM MgCl2∶10 mM MES ) , CF 3hpi ( P . infestans culture filtrate ) , MG132 3hpi ( 100 µM MG132 diluted in culture filtrate ) . For Pi03192 treatments GFP-NAC constructs were co-infiltrated with either pFlub-empty vector or pFlub-Pi03192 and subsequently treated with CF 3hpi or MG132 3hpi as above . 1 cm2 leaf disks were ground in LN2 and resuspended in 100 ul 2X SDS PAGE sample loading buffer and loaded onto a 12% Bis-Tris NuPAGE Novex gel run with 1X MOPS SDS running buffer for 1 . 5 h at 120 V ( Invitrogen ) . Gels were blotted onto a nitrocellulose membrane for 1 . 5 h at 30 V and stained with ponceau solution to show loading and transfer . Membranes were blocked in 5% milk in 1X PBS before addition of the primary antibodies at 1∶1000 dilutions: either a Monoclonal GFP antibody raised in mouse ( Sigma-Aldrich ) , a Monoclonal Anti HA antibody raised in rabbit ( Sigma-Aldrich ) or a Monoclonal Anti MYC antibody raised in rabbit ( Sigma-Aldrich ) . The membrane was washed with 1X PBST ( 0 . 2% tween 20 ) before addition of the secondary antibody at 1∶5000 dilution either Anti-Mouse IG HRP antibody ( Sigma-Aldrich ) or Anti-Rabbit IG HRP antibody ( Sigma-Aldrich ) . SuperSignal West Femto ( Thermo Scientific ) ECL detection was used according to the manufacturer's instructions . RNA was extracted using Tri Reagent [56] . RNA was treated to remove DNA contamination using a DNase Turbo Free kit ( Ambion ) according to the manufacturer's instructions . First strand cDNA was synthesised from 2 µg of RNA using Superscript II RNase HReverse Transcriptase ( Invitrogen ) according to manufacturer's instructions . Realtime qRT-PCR reactions were performed using Power SYBR Green ( Applied Biosystems ) and run on a Chromo4 thermal cycler ( MJ Research , UK ) using Opticon Monitor 3 software . Primer pairs were designed outside the region of cDNA targeted for silencing following the manufacturer's guidelines . Primer sequences in Supplementary Table S1 . Detection of real-time RT-PCR products , calculations and statistical analysis were performed as previously described [57] . Phytophthora infestans strains 88069 and an 88069-tdT transformant [26] were used for plant infection and were cultured on Rye Agar at 19°C for 2 weeks . Plates were flooded with 5 ml H2O and scraped with a glass rod to release sporangia . The resulting solution was collected in a falcon tube and sporangia numbers were counted using a haemocytometer and adjusted to 15000 sporangia/ml , 10 µl droplets were inoculated onto the abaxial side of detached N . benthamiana leaves stored on moist tissue in sealed boxes . For VIGSed plants the number of inoculated lesions which were sporulating at 7 dpi were counted and expressed as a percentage increase sporulating lesions compared to the GFP control plants . Sporangia counts were performed on 10 dpi leaves from VIGSed plants which had been immersed in 5 ml H2O and vortexed to release sporangia . A Haemocytometer was used to count the number of sporangia recovered from each leaf and was expressed as sporangia/ml . VIGSed plants inoculated with an 88069-tdT transformant were imaged using a confocal microscope at 3 dpi to examine P . infestans growth microscopically . Agrobacterium tumefaciens Transient Assays ( ATTA ) in combination with P . infestans infection were carried out as described [24] , [26] . Briefly , Agrobacterium cultures were resuspended in agroinfiltration medium at a final concentration of OD600 = 0 . 1 and used for transient expression in planta by agroinfiltration . After 1 day , each infiltration site was inoculated with 10 µl zoospores from P . infestans isolate 88069 at 15000 sporangia/ml . An inverted repeat construct of Pi03192 was cloned using primers indicated in Supplementary Table S1 in to the pSTORA vector and transformed into wildtype isolate 88069 as described previously [24] . The 03192_IR transformant was maintained on Rye agar plates supplemented with geneticin 20 µg/ml and gene expression levels from germinating cysts were determined by qRT-PCR as described previously [24] , using primer sequences in Supplementary Table S1 . P . infestans 03192_IR infection assays and ATTA experiments were performed as above . The HMMer3 [58] hmmsearch package was used to query: the NCBI RefSeq/nr sequence database; the PGSC Solanum phureja genome annotation v3 . 4; the Nicotiana benthamiana NBGI . 042210 release from the Dana Farber Cancer Institute; and Solanum lycopersicum annotations from the ITAG2 . 3 annotation and www . plantgdb . org ( all downloaded 22/3/2012 ) for NAM domain-containing proteins , using the HMM profile for the NAM domain family ( PF02365; http://pfam . sanger . ac . uk/family/PF02365 ) with the GA cutoff , producing a set of 2552 unique NAM domain protein sequences . These sequences were aligned using HMMer3's hmmalign package , and sequences with less than 40% sequence identity to any other sequence in the alignment discarded . The resulting set of 1700 sequences were clustered using MCL using an inflation value of 6 . 0 , and a single cluster of 406 sequences containing the NAM domains of both StNTP sequences was retained . The aligned sequences were back-translated by threading against the draft genomes from the genome sequencing projects from which they derived , or against their corresponding entry in the NCBI nr database as appropriate , using the Python script ( https://github . com/widdowquinn/scripts/blob/master/bioinformatics/get_NCBI_cds_from_protein . py ) , to give 374 aligned nucleotide sequences . The nucleotide alignment was processed with trimal [59] to remove all columns containing more than 10% gaps , producing an alignment with 337 sequences and 374 characters . jModelTest [60] was used to determine an appropriate substitution model for phylogenetic reconstruction which , by consensus of AIC and BIC , was GTR+I . A maximum likelihood tree was produced for this alignment using raxML [61] with 100 bootstrap trees , using the GTRCATI model . The resulting tree was rendered with FigTree ( http://tree . bio . ed . ac . uk/software/figtree/ ) . Transmembrane domains were predicted using the most complete available amino acid sequence available for each NAM domain-containing protein in the 2552 sequence set , using TMHMM v2 . 0 [62] . 161 NAM domain-containing proteins were predicted to contain a TM domain , of which 149 ( 92% ) were also in the cluster of 406 sequences containing the StNTP proteins identified by MCL . Pi03192 is PITG_03192 ( GenBank accession: XM_002906231 . 1 GI: 301117097 ) ; StNTP1 ( GenBank accession: KF437522 ) ; StNTP2 ( KF437523 ) ; NbNTP1 ( KF437524 ) and NbNTP2 ( KF437525 ) . | The plant immune system is activated following the perception of exposed , essential and invariant microbial molecules that are recognised as non-self . A major component of plant immunity is the transcriptional induction of genes involved in a wide array of defence responses . In turn , adapted pathogens deliver effector proteins that act either inside or outside plant cells to manipulate host processes , often through their direct action on plant protein targets . To date , few effectors have been shown to directly manipulate transcriptional regulators of plant defence . Moreover , little is known generally about the modes of action of effectors from filamentous ( fungal and oomycete ) plant pathogens . We describe an effector , called Pi03192 , from the late blight pathogen Phytophthora infestans , which interacts with a pair of host transcription factors at the endoplasmic reticulum ( ER ) inside plant cells . We show that these transcription factors are released from the ER to enter the nucleus , following pathogen perception , and are important in restricting disease . Pi03192 prevents the plant transcription factors from accumulating in the host nucleus , revealing a novel means of enhancing host susceptibility . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2013 | An RxLR Effector from Phytophthora infestans Prevents Re-localisation of Two Plant NAC Transcription Factors from the Endoplasmic Reticulum to the Nucleus |
Hematopoietic stem cells ( HSCs ) balance self-renewal and differentiation to maintain homeostasis . With aging , the frequency of polar HSCs decreases . Cell polarity in HSCs is controlled by the activity of the small RhoGTPase cell division control protein 42 ( Cdc42 ) . Here we demonstrate—using a comprehensive set of paired daughter cell analyses that include single-cell 3D confocal imaging , single-cell transplants , single-cell RNA-seq , and single-cell transposase-accessible chromatin sequencing ( ATAC-seq ) —that the outcome of HSC divisions is strongly linked to the polarity status before mitosis , which is in turn determined by the level of the activity Cdc42 in stem cells . Aged apolar HSCs undergo preferentially self-renewing symmetric divisions , resulting in daughter stem cells with reduced regenerative capacity and lymphoid potential , while young polar HSCs undergo preferentially asymmetric divisions . Mathematical modeling in combination with experimental data implies a mechanistic role of the asymmetric sorting of Cdc42 in determining the potential of daughter cells via epigenetic mechanisms . Therefore , molecules that control HSC polarity might serve as modulators of the mode of stem cell division regulating the potential of daughter cells .
Hematopoietic homeostasis depends on the ability of hematopoietic stem cells ( HSCs ) to balance symmetric and asymmetric divisions over an organism’s lifespan . Asymmetric divisions allow one daughter cell to differentiate while the other retains its stem cell potential . In contrast , symmetric divisions result in daughter cells that adopt equivalent fates [1–8] . The asymmetric or symmetric distribution of cellular components to daughter cells during mitosis is thought to determine their fate . This concept has been proposed as a way of controlling the size of the HSC pool [9] . However , the mechanisms that control the mode and outcome of HSC divisions with respect to the potential and ultimately function of daughter cells remain incompletely understood . While the number of phenotypic HSCs increases upon aging , their regenerative potential decreases , both in mice and in humans [10–13] . Aged HSCs differentiate preferentially into myeloid cells and less into cells of the lymphoid lineage . Aging of HSCs is caused by intrinsic changes in HSCs and also by extrinsic factors from the niche [14–20] . Epigenetic modifications and altered gene expression profiles ( e . g . , high expression of myeloid genes ) , together with loss of the polar distribution of tubulin and cell division control protein 42 ( Cdc42 ) in the cytoplasm and loss of the polarity of the acetylated form of the epigenetic marker H4K16 ( H4K16ac ) in the nucleus , are hallmarks of intrinsic HSC aging [10 , 21–23] . Cdc42 is a small RhoGTPase that cycles between an active guanosine triphosphate ( GTP ) -bound state and an inactive guanosine diphosphate ( GDP ) -bound state . A central , HSC-intrinsic molecular mechanism that causes aging of HSCs and that results in the above-mentioned phenotypic and functional impairments of aged HSCs is an increase in the activity of Cdc42 in aged stem cells [10 , 15 , 16 , 24 , 25] . Consequently , treatment of aged HSCs ex vivo with a Cdc42 activity specific inhibitor ( CASIN ) , which reduces the age-associated increase in Cdc42 activity , increased polarity in aged HSCs and rejuvenated their function in vivo [16] . The increase in Cdc42 activity in aged HSCs is caused by an intrinsic increase in the level of Wnt5a in HSCs [15] . Treatment of young HSCs with Wnt5a induced elevated activity of Cdc42 and a loss of cell polarity and premature aging of HSCs [15] . We hypothesized here that HSC polarity determines the mode ( symmetry/asymmetry ) of HSC divisions and therefore that there is a difference in the symmetry/asymmetry of division upon changes in the activity level of Cdc42 in HSCs . We thus investigated whether the targeted alteration of Cdc42 activity by either CASIN ( to decrease Cdc42-GTP levels in aged HSCs ) or Wnt5a treatment ( to increase Cdc42-GTP levels in young HSCs ) , which translates into polarity changes , influences the balance of asymmetric or symmetric HSC divisions ( mode of division ) and therefore drives functional differences among daughter cells ( output of division ) upon aging and rejuvenation ( S1A Fig ) . By comparing the mode of HSC division in vitro and the output of division in vivo , we demonstrate that young and aged CASIN-treated HSCs divide asymmetrically while aged and young Wnt5a-treated HSCs undergo symmetric divisions . We conclude that , by modulating Cdc42 activity and consequently cell polarity , it is possible to adjust the mode and the epigenetic outcome of HSC divisions .
We first microscopically tracked the kinetics of single HSC doubling in vitro ( Fig 1A ) . There was no difference in the kinetics of first and second divisions between young , aged , aged treated with CASIN , and young treated with Wnt5a HSCs ( Fig 1B ) nor in the relative distribution of mitotic cells to metaphase , anaphase , or telophase ( S1B Fig ) . Next , 3D confocal immunofluorescence ( IF ) was used to quantify Cdc42 and H4K16ac allocation to daughter cells during mitosis . Young HSCs were found to segregate Cdc42 and H4K16ac asymmetrically into daughter cells . Aged HSCs displayed a symmetric distribution of both Cdc42 and H4K16ac in daughter cells , and inhibition of Cdc42 activity in aged HSCs increased the frequency of asymmetric divisions among aged HSCs . Consistently , in the presence of Wnt5a , young HSCs presented a shift towards a symmetric allocation ( Fig 1C and 1D and Fig 2A and 2B; S1C–S1J Fig; S1–S8 Movies; S1 Table; S2 Fig ) . To note , polarity was not affected by cell culture conditions ( S1K Fig ) , and Cdc42 protein was observed asymmetrically or symmetrically already during metaphase ( S1L Fig ) . H4K16ac levels were stably detected throughout mitosis ( S1M Fig ) , and the difference in allocation was quantified at telophase . When setting the threshold to a 75% difference in Cdc42 amount between daughter cell pairs ( which translates into at ratio of 1 over 4 , see also cartoon scheme in Fig 1D ) , 80% of young HSCs divided asymmetrically ( Fig 2C ) . The frequency of asymmetric division was significantly reduced upon aging ( S1 Table and Fig 2D ) . Consistent with polarity/apolarity before division , aged CASIN-treated HSCs divided more frequently asymmetrically , while Wnt5a treatment of young HSCs increased the frequency of symmetric divisions ( Fig 2E and 2F ) . Because it is not yet possible to image in living cells Cdc42 and H4K16ac allocation to nascent daughter cells , we used mathematical association analyses to investigate whether there is a link between polarity in mother HSCs and the mode of HSC division . The probability that a polar stem cell will undergo an asymmetric division was termed da ( with probability for symmetric division defined as 1 − da ) . Similarly , ds represented the probability that an apolar stem cell divides symmetrically ( with 1 − ds for asymmetric ) ( Fig 3A ) . By applying a maximum likelihood estimation to the previously published observed frequencies of polar versus apolar HSCs in young , aged , young Wnt5a-treated , and aged CASIN-treated cell populations [15 , 16] , we found that the probabilities da ≈ 0 . 96 and ds ≈ 0 . 82 best explained the corresponding modes of division ( Fig 3A ) . Modelling thus indicates that there is a 96% chance that a polar cell will divide asymmetrically , and apolar cells will divide with an 82% chance symmetrically . Encouraged by this strong association between HSC polarity/apolarity and asymmetric/symmetric mode of division , we further designed a mechanistic mathematical model to investigate the causative link to the fates of the daughter cells ( see S1 Text and S3 Fig ) . To mimic the experimental observations , we assumed that Cdc42 is either distributed uniformly ( apolar ) or peaked ( polar ) along a circular , abstracted cell shape . We directly coupled the active proportion of intracellular Cdc42 ( Cdc42-GTP ) to the shape of this distribution , thereby functionally connecting the observed increase in Cdc42 activity with the loss of polarity during aging [15] . The mathematical approach suggested that a switch-like construct could determine daughter cell fate , in which different fates manifest based on the intracellular protein concentration immediately after cell division . We also implemented a transcriptional autoregulative feedback for total Cdc42 concentration , thereby establishing bistability ( Fig 3B ) . For certain parameter choices , such a system yielded 2 characteristic states: either Cdc42 was highly expressed due to self-activation ( e . g . , indicative of HSCs ) or Cdc42 was lowly expressed ( e . g . , indicative of the progenitor state ) ( Fig 3C and 3D ) . The choice of either the HSC or the progenitor state was determined by Cdc42 concentration right after cell division . Distinct concentrations between daughter cells due to an asymmetric mode of division resulted also in distinct states ( Fig 3C and 3D ) . Because Cdc42 activity , and thereby HSC polarity , is altered during aging , the model predicted that asymmetric outcomes of division are more likely for young HSCs , while a higher incidence of symmetric divisions in aged HSCs will preferentially promote self-renewal , predicting a progressive expansion of the stem cell pool over time ( Fig 3E ) . Indeed , an increase in the number of HSCs in aged mice and humans has been verified by several groups [8 , 15 , 16 , 22 , 26–28] . To experimentally verify the predictions of the mathematical models and to further determine the function of daughter cells , we performed single daughter cell transplants using Rag2−/−γc−/−KitW/Wv recipient mice [29] . After the first division in vitro , we injected each single daughter cell into an individual recipient mouse ( Fig 4A ) and assessed its contribution to peripheral blood ( PB ) every 4 to 8 weeks up to 24 weeks after transplantation . A daughter cell that was able to contribute to B cells , T cells , and myeloid cells 24 weeks after transplantation with an overall engraftment of more than 0 . 1% qualified as a stem cell [30 , 31] . When a contribution at 24 weeks was not detected in all cell lineages tested or the overall contribution of all donor-derived cells was below 0 . 1% , the initially transplanted cell was scored as a progenitor cell . Based on the scores of the outcomes of the daughter cell transplants , the initial HSC division was then defined as symmetric ( two daughter stem cells or two daughter progenitor cells ) or asymmetric ( one daughter stem cell and one daughter progenitor cell ) ( Fig 4B; S4A–S4D Fig; S5A Fig ) . Young HSCs divided primarily asymmetrically , while aged HSCs showed mainly symmetric division outcomes ( Fig 4C; S2 and S3 Tables; S5A Fig ) . Treating aged HSCs with CASIN significantly increased the frequency of asymmetric divisions of aged HSCs; in contrast , treating young HSCs with Wnt5a ( Cdc42 inducer ) increased the frequency of symmetric divisions ( Fig 4C; S2 and S3 Tables; S5A Fig ) . The highest overall contribution to PB was detected in mice transplanted with young daughter stem cells at both early and late time points after transplantation ( Fig 4D and 4E; S5A Fig ) . Young Wnt5a-treated daughter stem cells presented with a decrease in engraftment only at early time points ( 4 weeks ) , while there was no difference in engraftment between young and young Wnt5a-treated daughter stem cells at the end point ( 24 weeks ) ( Fig 4D and 4E ) . Aged daughter stem cells from CASIN-treated mother HSCs showed an about 2-fold increase in overall engraftment compared to daughter stem cells from aged untreated HSCs ( Fig 4E; S5A Fig ) . Young daughter stem cells presented with the highest level of contribution to B cells , while aged and young Wnt5a-treated cells displayed a significantly reduced B cell output ( Fig 5A and 5B; S5A Fig ) . Daughter stem cells from aged CASIN-treated HSCs differentiated significantly more than the other experimental samples into T cells ( Fig 5A and 5B; S5A Fig ) . As for daughter progenitor cells , no difference was detected among samples , besides the observation that young and aged CASIN-treated sample sets presented with a trend in higher reconstitution at 24 weeks after transplant ( S5B–S5D Fig ) . Our data further demonstrate that the initial population of mother HSCs the experiments were initiated with contained , on average , 77 . 5% of cells with stem cell function ( based on their ability to give rise to at least one daughter stem cell ) , and there was no difference between the groups of young and aged HSCs and treated HSCs ( S3 Table and S6 Fig ) . Frequent symmetric divisions of aged HSCs might result in a local accumulation of HSCs in bone marrow . We set up experiments to test this hypothesis and stained whole-mount preparations of the long bones of young and aged mice ( S7A Fig ) . Indeed , in aged animals , HSCs often were found in clusters of 2 to 3 stem cells , while HSCs in young animals were almost always found as solitary cells ( Fig 6A and 6B ) . As a consequence , the distance of HSCs to their nearest neighbor HSC was significantly decreased in aged bones , and around 30% of all HSCs were found less than 20 μm ( proximity ) from another HSC ( Fig 6C and 6D and S7B and S7C Fig ) . This translated into 34% of all aged HSCs being in clusters compared to less than 2% for young HSCs ( Fig 6E and S7D Fig ) . We next investigated whether HSC clustering was intrinsic to aged HSCs . To this end , we transplanted 1 , 000 young and aged HSCs from AcRFP mice ( that ubiquitously and constitutively express the red fluorescent protein [RFP] ) into young Rag2−/−γc−/−KitW/Wv recipient mice and imaged the bone marrow 6 weeks after transplant . We observed a higher frequency of HSC clusters ( 58% of all HSCs ) in the bone marrow of recipients of aged HSCs compared to recipients of young HSCs ( 11% of all HSCs ) ( Fig 6F and 6G and 6J ) . Aged HSCs were also closer to each other and more frequently located in proximity of another HSC when compared to the distribution of young HSCs ( Fig 6H and 6I ) . The frequency of clusters , compared to non-transplanted controls , increased for both aged and young donor HSCs ( Fig 6J and S7E Fig ) , suggesting a contribution of the transplantation to aging of HSCs . In summary , clusters of HSCs in bone marrow are intrinsic to HSCs , are primarily formed by aged HSCs , and are consistent with an elevated frequency of symmetric divisions of aged HSCs . To investigate the extent of the correlation between the divisional asymmetry/symmetry of HSCs and the transcriptome of daughter cells , we performed single-cell RNA sequencing ( scRNA-seq ) on paired daughter cells ( Fig 7A ) . A global supervised “between-group analysis” ( BGA ) showed distinct clustering of daughter cells derived from the different experimental arms ( Fig 7B; S4 Table; S8 Fig; S9A–S9C Fig ) . Daughter cells from aged and aged CASIN-treated HSCs showed a higher transcriptional heterogeneity compared to daughter cells from young and young treated-with-Wnt5a HSCs . When daughter pairs were separated based on chronological age , the aged CASIN-treated arm was closer to the young than to the aged control arm . However , the daughter cells from young Wnt5a-treated HSCs were also closer to the young control than to the older control arm ( Fig 7B ) . To evaluate asymmetry/symmetry based on the difference of the transcriptome between the daughter cells , we first performed Self-Organizing Maps ( SOM ) analyses followed by three different analytical approaches ( S10 Fig ) . We selected statistically significant genes ( p-value and q-value < 0 . 05 ) that fall into distinct clusters of highly correlated genes ( metagenes ) that , summarily , presented a portrait that captures the global transcriptional landscape of a given cell . Cells with similar portraits and significant sets of genes will implicitly have high levels of transcriptional similarity . In the first analysis , we performed a hypergeometric-model–based gene set enrichment analysis using 13 previously published HSC and polarity signatures [32–39] by simultaneously interrogating all the gene sets ( Fig 7C; S11 Fig; S5 Table ) . When we then compared cells belonging to a pair , the majority of the pairs showed a high degree of similarity ( concordance ) across all arms ( Fig 7D ) . In the second approach , we quantified the degree of similarity between SOMs of daughter pairs by comparing the genes that showed statistically significant log fold-change compared to the mean global expression profile ( Fig 7E ) . Each pair was classified as concordant or discordant based on the magnitude of genes showing the same direction of change . Again , we observed a high level of concordance for all daughter pairs ( Fig 7F ) . Finally , taking into consideration the results of the transplants ( Fig 4C ) and the mathematical modeling ( Fig 3E ) , we sought to confirm higher asymmetric outcomes ( discordant transcriptome profiles ) in young and aged CASIN-treated HSCs and increased symmetric division outcomes ( concordant transcriptome profiles ) in aged and young Wnt5a-treated HSCs . To this end , we analyzed differentially expressed genes among daughter cells from young and aged CASIN-treated HSCs to highlight discordant divisions among the different pairs ( Fig 7G; S6 Table ) . Results showed a percentage of asymmetric divisions slightly higher for all arms compared to the previous gene set enrichment and fold change analysis approaches; however , again results were not in agreement with the transplants and the mathematical prediction , and no significant difference was observed among the four arms ( Fig 7H ) . In the end , we analysed enrichment for biological processes in the 134 up-regulated genes in young and aged CASIN-treated daughter pairs . We observed a significant enrichment in epigenetic regulation of gene expression and chromatin organization ( Fig 7I ) . In the down-regulated gene sets , structural constituents of ribosomes were significantly enriched ( Fig 7I; S9D Fig ) . No significant enrichment was found for the genes up-regulated in aged and young Wnt5a-treated cells ( S7 Table ) . Therefore , the transcriptome in daughter cells to a large extent does not correlate with the function of a daughter cell as determined in the transplantation experiments . It is well established that genes specific for certain cell types or that encode crucial lineage determining regulators might be activated in a stepwise fashion at the levels of chromatin accessibility first , followed by transcriptional activation [40] . Therefore , to investigate whether chromatin accessibility might be distinct in daughter , cells we performed single-cell transposase-accessible chromatin sequencing ( scATAC-seq ) on daughter cells from young and aged HSCs ( Fig 8A; S12 Fig; S13 Fig ) . ATAC-seq was selected because it reflects alterations in chromatin accessibility that are possibly also linked to changes in the level of H4K16ac ( see Fig 2A and [41–44] ) . The ATAC-seq profile of single daughter cells showed the expected distribution of insert sizes and an enrichment at transcription start sites ( TSSs ) of nucleosome-free fragments and the expected profile with respect to CCCTC-binding factor ( CTCF ) binding sites ( S13 Fig ) . ATAC-seq peaks were annotated predominantly to TSSs , introns , and intergenic regions up to 1 kb apart from TSSs ( S14A Fig ) . The overall scATAC-seq profile of daughter pairs of young HSCs showed a marked increase in the value of peak ratio compared to daughter pairs of aged HSCs . This resulted in a difference in ATAC-seq peak amounts between daughters from young HSCs , while daughter pairs of aged HSCs presented more often with having equal amounts of ATAC-seq peaks and a ratio quite uniform around 1 ( Fig 8B ) . The difference in the ATAC-seq peak amount between paired daughters allowed for scoring of divisions as symmetric or asymmetric . Briefly , for those pairs for which we had a sufficient number of peaks for the reshuffling simulation test of the null hypothesis that cells in a pair show the same peak counts in the same genomic positions , asymmetry assignment was based on the significance of no overlap and the ratio of the peak counts ( as a measure of global chromatin accessibility ) greater than 2 ( S8 Table and S15 Fig ) . Based on these statistical parameters , young HSCs presented with 57% of asymmetric divisions , while aged HSCs showed 40% of asymmetric divisions ( Fig 8C; S8 Table; S15 Fig ) . The symmetry/asymmetry difference between young and aged cells determined by 3D-IF for Cdc42 and H4K16ac , in transplantation assays and as predicted by the mathematical model , is thus similar to the difference in the ATAC-seq peak profile of paired daughters of young and aged HSCs . We next annotated scATAC-seq peaks to genes . Similar to the analyses performed for the scRNA-seq data , we used gene set enrichment analysis to align daughter pairs to known stem cell signatures ( Fig 8D and S16 Fig ) and presented them again in radar plots . The data revealed also for these analyses a higher frequency of asymmetric divisions for young HSCs ( 64% ) compared to asymmetric division for aged HSCs ( 40% ) ( Fig 8E and S16 Fig ) . Individual daughter cells from young and aged pairs were subsequently grouped according to the mode of division as established based on the ATAC-seq peak ratio ( S8 Table and Fig 8B and 8C ) : both daughter cells of all symmetric divisions were assorted as stem cells because the majority of daughter cells stemming from symmetric divisions were , according to the transplantation assay , stem cells . As for asymmetric divisions , the daughter cells with the highest ATAC-seq peak number were assigned to the daughter progenitor cells group because daughter cells of symmetric divisions , which result in daughter stem cells , presented with a quite uniform and low peak count ( Fig 8F ) . Next , transcription factor ( TF ) footprint analyses on ATAC-seq peaks identified significantly enriched TF binding sites common to daughter stem cells of both young and aged HSCs ( S14B and S14C Fig and S17 Fig and S9 Table ) . Among them were TFs like NFY , SpiB , Sp5 , Sp1 , Runx , ERG , PU . 1 , and Maz ( S14B–S14D Fig ) . Of these , the frequencies of target sequences with a Sp5 , Sp1 , Maz , and ERG motif were even further increased in daughter stem cells of aged HSCs ( S14D Fig ) . In daughter progenitor cells from both young and aged HSCs , the enriched TFs set was very distinct from the one enriched in stem cells ( S14E and S14F Fig ) , and factors like c-Myc , Usf2 , and Stat3 further showed a decline in enrichment in aged daughter progenitors , while PU . 1:IRF8 , CEBP , CEBP:AP1 , and Nanog showed the opposite trend ( S14E Fig ) . However , only the frequency of target sequences for the E2F6 motif was significantly more enriched in aged daughter progenitors ( S14G Fig ) . Finally , gene ontology ( GO ) analysis revealed 603 and 490 significantly enriched pathways in young and aged daughter stem cells and 474 and 360 pathways in young and aged daughter progenitor cells ( S18 Fig and S10 Table ) . More than 75% of young daughter progenitor cells shared enrichment for only 1 . 47% of all significantly enriched GOs . These daughter-progenitor–specific enriched GOs for young HSCs were platelet homeostasis , mitosis , interferonγ and interferon signalling , IL-2 and IGF1 signalling , and pyruvate and lipid metabolism ( Fig 8G ) [45–49] . In daughter progenitors of aged HSCs , 100% of the cells were enriched for only 2 . 77% of the GOs , which included platelet homeostasis , TNF signalling , propionyl-CoA catabolism , and cholesterol biosynthesis ( Fig 8H ) [50–52] . To further dissect differences between young and aged daughter progenitor cells , we performed gene set enrichment analyses for a previously reported multipotent hematopoietic progenitor cell signature [53] . Despite only 2 aged and 3 young daughter progenitor cells showing significant enrichment , both young and aged daughter progenitor cells showed a similar set of genes overlapping across cells in a given age group but also across the two age groups , indicating a similar pattern of chromatin accessibility in and around these genes in both young and aged daughter progenitor cells ( S14H–S14I Fig ) . Most of the daughter stem cells from both young and aged HSCs were enriched for GO pathways linked to glycolysis and gluconeogenesis and also for signaling linked to RhoGTPases , implying an important role of regulation of the glycolytic metabolism and of small RhoGTPase signaling ( like Cdc42 ) for the regulation of stem-ness of daughter stem cells ( Fig 8I and 8J ) [54–56] , as also predicted by our mathematical model ( Fig 3 ) . Similarly informative might be the finding that major differences in GO enrichment among daughter stem cells from young and aged HSCs were linked to signalling pathways involved in the interaction of HSCs with the stem cell niche: while daughter stem cells from young HSCs are enriched for GOs associated with TGFβ , VEGFR2 , and EGFR signalling , daughter stem cells from aged HSCs are enriched for Wnt signalling ( Fig 8I–8J ) [15 , 56–58] . In summary , open chromatin configuration and thus the epigenetic make-up of a daughter cell strongly correlates with the potential of the daughter cell .
The asymmetric or symmetric sorting of epigenetic information upon HSC division has been proposed as a way of controlling the potential of nascent daughter cells [9] . Here , we demonstrate , using a comprehensive set of paired daughter analyses that include single-cell 3D confocal imaging , single-cell transplants , scRNA-seq , and scATAC-seq , that the mode of HSC division is strongly linked to the polarity status before mitosis , which is in turn determined by the level of the activity Cdc42 in stem cells . These results consequently imply that the level of activity of Cdc42 in mother HSCs [4 , 15] regulates the mode ( symmetric/asymmetric ) of HSC divisions , which results in a distinct allocation of epigenetic regulators like H4K16ac as well as of Cdc42 itself to daughter cells . Our data , combined with our mathematical model on the mode of divisions and the role of Cdc42 in daughter cells , further support that polar HSCs ( that are the majority in young animals ) undergo asymmetric divisions while apolar HSCs ( that are the majority in aged animals ) undergo primarily symmetric divisions . Aging and rejuvenation of HSCs , as a consequence of changes in the activity of Cdc42 in HSCs [15 , 16] , also imply a change in the mode of HSC divisions and thus link the mode of division of HSCs to changes in the function of daughter stem cells . In addition , our data ( both experimental and mathematical modeling ) strongly support that symmetric divisions , which are the primary mode of division of aged HSCs , might be directly linked to the increase in the number of HSCs in bone marrow upon aging . A striking new feature , clustering of HSCs within the bone marrow of aged animals ( Fig 6 ) , is most likely a consequence of the elevated frequency of symmetric divisions of aged HSCs . An interesting novel concept derived from HSC clustering is that , in such clusters , HSCs might serve as niche cells for other HSCs . Recently published data support a dominant role of hematopoietic progenitor cells but not HSCs in maintaining long-term hematopoiesis in an unperturbed setting in young animals [59 , 60] . Our findings are consistent with this novel concept because young HSCs will divide primarily asymmetrically and provide the progenitor cells that support hematopoiesis . Our data might predict that , upon aging , this dominant role of progenitor cells in driving hematopoiesis will shift towards a more relevant contribution of stem cells to hematopoiesis due to the reduced number of progenitors generated from aged HSCs . The final fate of daughter cells after HSC divisions might be defined by signals intrinsic to stem cells that are allocated during division to daughter cells but might also be driven by extrinsic signals from the environment acting on daughter cells after mitosis is completed . The stem cell intrinsic signals need to be , by definition , epigenetic in nature ( not linked to the DNA sequence ) , as daughter cells present with a genome that is identical to their mother cell . Our data and our mathematical model support that the decision on the potential of daughter cells is driven , to a large extent , by the mode of the division of the mother HSC , which is decided upon in the mother HSCs . We show distinct allocation of a polarity protein ( Cdc42 ) and an epigenetic mark ( H4K16ac ) to daughter cells , which correlates with the potential of the daughter cells themselves . Our data also demonstrate that the overall transcriptome in daughter cells does not correlate with their potential , while the amount and the location of open chromatin regions is tightly linked to it . Furthermore , mathematical modelling and our ATAC-seq-linked GO signatures in daughter stem cells predict an important role for RhoGTPase signaling as well as metabolic signaling for the maintenance of the stem cell potential in daughter stem cells upon stem cell divisions . While a role of glycolysis in stem cell maintenance has been already described by multiple laboratories [54 , 55] , an important role for RhoGTPase signaling for stem cell maintenance has only been recently suggested [61 , 62] . The daughter progenitor cell signature is distinct from the stem cell signature , while sharing for both young and aged daughter-cell gene loci linked to platelet homeostasis . It has been described that HSCs can rapidly differentiate into platelets [63] . In addition , progenitor cells display enrichment for signatures linked to interferon signaling , which has been , in the past , primarily associated with stem cell quiescence/activation [45 , 64 , 65] . Recently , it has been reported that H4K16ac can directly control chromatin accessibility [44] , so it is likely that a symmetric or asymmetric distribution of H4K16ac to daughter cells ( due to polarity/apolarity of the mark in mother stem cells ) might contribute to distinct levels of chromatin accessibility and thus affect the fate of daughter cells . These findings are consistent with recent data supporting that epigenetic memory in HSCs is persistent and guides cellular function [66] . The data presented here also imply that an open chromatin configuration precedes changes in gene expression in daughter cells . Our ATAC-seq analyses imply that young HSCs might also establish a specific chromatin configuration that allow stem-cell–extrinsic signalling ( TGFβ , VEGF , IGF1 , and EGF signalling ) to act on nascent daughter stem cells , while aged daughter stem cells present with open chromatin regions linked to Wnt signalling . A role for all of these factors for the biology of young and aged HSCs has already been described , further supporting these conclusions [15 , 56–58] . In summary , our data imply an important role for Cdc42 activity/polarity in HSCs for driving the symmetric/asymmetric mode of division as well as a role of epigenetic mechanisms for determining the potential of daughter cells . The frequency of polar HSCs decreases upon aging , which results in more symmetric divisions but daughter stem cells with impaired potential .
This study involves animal research ( mice , C57BL/6 strain ) . All mice were housed in the animal barrier facility under pathogen-free conditions either at the University of Ulm or at CCHMC . All mouse experiments were performed in compliance with the German Law for Welfare of Laboratory Animals and were approved by the Institutional Review Board of the University of Ulm or by the IACUC of CCHMC ( approval number: TVA 0 . 1172 and IACUC2013-0154 ) . This study does not involve human participants and/or tissues , and it does not involve collection of plant , animal , or other materials from a natural setting . C57BL/6 mice ( 10- to 12-week-old ) were obtained from Janvier . Aged C57BL/6 mice ( 20- to 26-month-old ) were obtained from the internal divisional stock ( derived from mice obtained from both The Jackson Laboratory and Janvier ) as well as from NIA/Charles River . Congenic young and aged C57BL/6 . SJL-Ptprca/Boy ( BoyJ ) mice were obtained from Charles River Laboratories or from the internal divisional stock ( derived from mice obtained from Charles River Laboratories ) . Pan-RFP mice carrying constitutively active ROSA26-tdRFP alleles ( indicated in the manuscript as Ac-RFP mice ) were obtained from Professor Hans Joerg Fehling ( Institute of Immunology , Ulm University ) and were previously generated by intercrossing C57BL/6-Gt ( ROSA ) 26Sortm1Hjf/Ieg mice [67] with animals from a germline Cre-deleter strain [68] . Offspring in which the ROSA26-driven fluorescent tdRFP reporter had been activated irreversibly as the result of loxP/Cre-mediated recombination in the germline were backcrossed for >10 generations onto C57BL/6 , thereby eliminating the Cre recombinase transgene . Ac-RFP mice were used as homozygotes . Rag2−/−γc−/−KitW/Wv mice were obtained from the internal divisional stock ( derived from mice obtained from Hans-Reimer Rodewald [29] ) . All mice were housed in the animal barrier facility under pathogen-free conditions either at the University of Ulm or at CCHMC . All mouse experiments were performed in compliance with the German Law for Welfare of Laboratory Animals and were approved by the Institutional Review Board of the University of Ulm or by the IACUC of CCHMC . SMART-Seq version 4 Ultra Low Input RNA kit from Clontech ( catalog number 634892 ) was used . The kit generates Illumina compatible RNA-Seq libraries . Sorted single cells were cultured with and without treatment in the presence of cytokines until first cell division ( 40–44 hours ) . The daughter cells were manually separated , washed with phosphate-buffered saline ( PBS ) , and collected for RNA sequencing . The cDNA synthesis and amplification was done as recommended by Clontech . The amplified cDNA generated from single cells were used to make libraries using Illumina’s Nextera XT DNA Library Preparation kit ( catalog number FC-131-1096 ) as per Illumina’s instructions . The generated libraries were quantified using an agilent bio-analyzer , pooled , and subjected to next-generation sequencing in a Hi-Seq 2500 for pair-end 75 bp sequencing condition . Bone marrow mononuclear cells were flushed out from long bones ( tibiae and femurs ) of young and aged mice and were isolated by low-density centrifugation ( Histopaque 1083 , Sigma ) . Low-density bone marrow cells were stained with a cocktail of biotinylated lineage antibodies . Biotinylated antibodies used for lineage staining were all rat anti-mouse antibodies: anti-CD11b ( clone M1/70 ) , anti-B220 ( clone RA3-6B2 ) , anti-CD5 ( clone 53–7 . 3 ) , anti-Gr-1 ( clone RB6-8C5 ) , anti-Ter119 , and anti-CD8a ( clone 53–6 . 7 ) ( all from eBioscience ) . After lineage depletion by magnetic separation ( Dynalbeads , Invitrogen ) , cells were stained with anti-Sca-1 ( clone D7; eBioscience ) , anti-c-kit ( clone 2B8; eBioscience ) , anti-CD34 ( clone RAM34; eBioscience ) , anti-CD127 ( clone A7R34; eBioscience ) , anti-Flk-2 ( clone A2F10; eBioscience ) , and Streptavidin ( eBioscience ) . Single long-term HSCs ( gated as LSK CD34−Flk2− ) [15] were sorted using a BD FACS Aria III ( BD Bioscience ) into Terasaki plates . HSCs were cultured in IMDM plus 10% FBS plus P/S plus 100 ng/mL TPO , G-CSF , and SCF ( PeproTech ) at 37°C , 5% CO2 , 3% O2 . CASIN was used at a dose of 5 μM . Wnt5a ( R&D ) was used at a dose of 100 ng/mL . Each well of the Terasaki plate was microscopically tracked , and within 1 hour from HSC division , daughter cells were separated by mechanical pipetting , deposited in distinct Terasaki wells , and controlled again by microscopy . Eventually , the daughter pair was injected into recipient mice ( 1 daughter in 1 mouse , the second daughter into a littermate ) . For daughter cell pair transplantation , young ( 2- to 4-month-old ) and aged ( 24-month-old ) C57BL/6 . SJL-Ptprca/Boy ( BoyJ ) ( Ly5 . 1+ ) mice were used as donors . Single daughter cell HSCs were separated by pipetting under microscopy control and were directly injected into the retro-orbital vein of recipient mice . PB chimerism was determined by FACS analysis at week 4 , 12 , 16 , and 24 post transplant . The transplantation experiment was performed 8 times with a cohort of 10 to 16 recipient mice per group each transplant . In total , 110 recipients were injected , and 94 mice showed detectable chimerism ( contribution to PB >0 . 1% ) at a minimum of 1 time point of the analyses . PB immunostaining was performed according to standard procedures , and samples were analyzed on a LSRII flow cytometer ( BD Biosciences ) . Monoclonal antibodies to Ly5 . 2 ( clone 104 , eBioscience ) and Ly5 . 1 ( clone A20 , eBioscience ) were used to distinguish recipient from donor cells . For PB lineage analysis , the antibodies used were all from eBioscience: anti-CD3ε ( clone 145-2C11 ) , anti-B220 ( clone RA3-6B2 ) , anti-Mac-1 ( clone M1/70 ) , and anti-Gr-1 ( clone RC57BL/6-8C5 ) . The engraftment is plotted as percentage of donor-derived cells among total white blood cells . B cells , T cells , and myeloid cell lineage data are plotted as the percentage of B220+ , CD3+ , and Myeloid ( Gr-1+ , Mac-1+ , and Gr-1+Mac-1+ ) cells among donor-derived cells . Please note that data are plotted in a logarithmic scale to accommodate sample heterogeneity . Freshly sorted HSCs were seeded on fibronectin-coated glass coverslips . At 32 to 34 hours after culturing in IMDM plus 10% FBS plus P/S plus 100 ng/mL TPO , G-CSF , and SCF ( PeproTech ) ± CASIN 5 μM ± Wnt5a ( R&D ) 100 ng/mL at 37°C , 5% CO2 , 3% O2 , cells were fixed with BD Cytofix Fixation Buffer ( BD Biosciences ) [15] . After fixation , cells were gently washed with PBS , permeabilized with 0 . 2% Triton X-100 ( Sigma ) in PBS for 20 minutes , and blocked with 10% Donkey Serum ( Sigma ) for 30 minutes [15] . Primary and secondary antibody incubations were performed for 1 hour at room temperature . The primary antibodies were anti-alpha tubulin antibody ( Abcam , rat monoclonal ab6160 ) , anti-Cdc42 , and anti-H4K16ac obtained from Millipore and Abcam ( we tested 2 different antibodies for each target , and results were consistent; all 4 antibodies were rabbit polyclonal; anti-Cdc42 from Millipore was previously validated [4] ) . The secondary antibodies for IF were anti-rat DyLight488-conjugated antibody , anti-rat DyLight647-conjugated antibody , and anti-rabbit DyLight549-conjugated antibody ( all obtained from Jackson ImmunoResearch ) . Coverslips were mounted with ProLong Gold Antifade Reagent with or without DAPI ( Invitrogen , Molecular Probes ) . Samples were imaged with an AxioObserver Z1 microscope ( Zeiss ) equipped with a 63× PH objective . Images were analyzed with Zen software . Alternatively , samples were analyzed with an LSM710 confocal microscope ( Zeiss ) equipped with a 63X objective . Primary raw data were imported into the Volocity Software package ( version 6 . 2 , Perkin Elmer ) for further processing and conversion into 3D images . After intravenous injection of APC-anti-CD31 ( clone MEC13 . 3 , BioLegend ) and Alexa Fluor 647-anti-CD144 ( clone BV13 , BioLegend ) antibodies , long bones were harvested after postmortem heart perfusion with 4% paraformaldehyde ( PFA ) in PBS and were post-fixed in 4% PFA/PBS solution for 24 hours at 4°C . Subsequently , bones were embedded without bisecting in optimum cutting temperature ( OCT ) compound ( Tissue-Tek ) and were snap frozen in liquid nitrogen and stored at −80°C . Bones were shaved along the longitudinal axis on a cryostat until the BM cavity was exposed ( S5A Fig ) . The bones were purified from melting OCT . Specimens were fixed again in 4% PFA/PBS at RT for 30 minutes . Tissues were blocked and permeabilized with buffer containing 20% donkey serum and 0 . 5% Triton X-100 , were incubated with a fluorescent-labeled antibody PE-anti-CD150 or AlexaFluor488-anti-CD150 ( clone TC15-12F12 . 2 , BioLegend ) as well as Biotin-labeled primary antibodies anti-CD41 ( clone MWReg30 ) , anti-CD48 ( clone HM48-1 ) , anti-CD11b ( clone M1/70 ) , anti-B220 ( clone RA3-6B2 ) , anti-CD5 ( clone 53–7 . 3 ) , anti-Gr-1 ( clone RB6-8C5 ) , anti-Ter119 , and anti-CD8a ( Clone 53–6 . 7 ) ( all from eBioscience ) 1 to 3 days at 4°C and stained with Streptavidin-eFluor450 ( eBioscience ) or Streptavidin- FITC ( eBioscience ) for 2 hours at RT . The fluorescently labeled bone tissues were placed cut-face down onto a 4-well-μ-Slide and were covered in antifade to prevent tissue desiccation . The preparations were immediately examined under Zeiss LSM 710 or Leica TCS SP8 confocal microscopes and analyzed with the image analysis software Volocity ( version 6 . 2; Perkin Elmer ) . Raw Fastq files were adapter trimmed using Trimm Galore ( Babraham Institute ) , a wrapper tool around Cutadapt and FastQC . Reads with Phred score of 20 or higher were kept for further analysis . Reads were aligned to the mouse reference genome version 10 ( GRCm38/mm10 ) using tophat [69] . Transcript abundance estimation , fragments per kilobase of exon per million fragments mapped ( FPKM ) calculations , and normalization was done using Cufflinks [70–73] , a package of tools for transcriptome assembly and differential expression analysis of RNA-Seq data . Correspondence-based BGA was done using R and bioconductor package made4 [74 , 75] . SOM and metagene analysis was performed using oposSOM [76] . Calculation of concordant and discordant daughter pairs was done using quadrant count ratio ( QCR ) by counting genes that fall in the 4 quadrants . A chi-squared test was used to test the significance of difference of concordant ( quadrants I and III ) and discordant genes ( quadrants II and IV ) . Gene set enrichment analysis was carried out using stem-ness and cell-polarity–related gene sets . The analysis approach simultaneously interrogated 13 previously published HSC and polarity signatures [32–39] , and each daughter cell pair was depicted in a radar plot such that vertices in the plot correspond to each of the considered gene sets: Mm_HSC_Runx1_Wu [32] , Mm_HSC_Tcf7_Wu[32] , Mm_LT_HSC_Venezia [36] , Mm_Proliferation_Venezia [36] , Mm_Quiescence_Venezia [36] , Polarity_factors_Ting [37] , Novel_HSC_regul_polar_Ting [37] , Grover et al , 2016 [38] and Kowalczyk et al , 2015 [39] ( S6 Table ) . Cells were first individually analyzed , and significance of gene set enrichment was tested using hypergeometric test . Concordance between daughter pair was then visualized using radar plots and tested using goodness-of-fit test . Knowledge-driven comparison between young ( young LT-HSC/CASIN-treated aged LT-HSC cells ) versus aged ( aged LT-HSC/WNT5A-treated young LT-HSC ) arms was made by populating genes that show same direction of significant change ( p-value cutoff 0 . 05 ) in expression in a given arm but not in the other . Genes fulfilling this criterion were used to generate the Venn diagram shown in Fig 4G . Daughter pair concordance and discordance in each treatment group was determined based on the level of significance of correlation that each pair shows . GO analysis and analysis of enrichment for biological processes was done according to standard GO Consortium dataset by using PANTHER version 11 [77] . The protocol was adjusted from previously published procedures [78] . Briefly , single HSCs from young and aged C57BL/6 mice were sorted into Terasaki plates with IMDM supplemented with 10% FBS , P/S , and 100 ng/mL of SCF , G-CSF , and TPO at 3% O2 , 5% CO2 . Cells were tracked microscopically , and after the first division , the daughter cells were separated by pipetting and singularly subjected to fragmentation of open chromatin regions using Tn5 transposase ( Illumina ) , followed by a pre-amplification step , library preparation , and subsequent paired-end sequencing . For the pre-amplification , NEBNext Ultra II Q5 Master Mix was used with Primer 1: 5’GTCTCGTGGGCTCGGAGATGTGTATAAGAGACAG3’ and Primer 2: 5’TCGTCGGCAGCGTCAGATGTGTATAAGAGACAG3’ . For dual-indexing , 10 μL of the pre-amplified ATAC reaction was used as input for Nextera index kit ( Illumina ) . The generated libraries were quantified using an agilent bio-analyzer and a qPCR kit ( New England Biolabs ) , pooled , and subjected to next-generation sequencing in a Hi-Seq 3000 for paired-end 75 bp sequencing condition . We used the ENCODE guidelines and current standards of ATAC-seq data analysis . This was achieved by using a combination of ENCODE prototype scripts and in-house algorithms . Briefly , sequences were aligned to the UCSC mouse reference genome mm10 using Bowtie 2 [79] . Peak calling was performed using MACS [80] . Additional analysis , visualization , and DNA footprint analyses were done using various in-house R scripts and published Bioconductor packages ( ATACSeqQC ) [81] . Overlap analysis between peak files was done using bedtools [82] . Random chromosome-controlled reshuffling of peaks was performed using bedtools followed by Monte Carlo Simulation based on in-house script . TF motif analysis and annotation were done using HOMER [83] . Gene set enrichment analysis was carried out using stemness and cell-polarity–related gene sets . The analysis approach simultaneously interrogated 13 previously published HSC and polarity signatures [32–39] , and each daughter cell pair was depicted in a radar plot in which vertices in the plot correspond to each of the considered gene sets: Mm_HSC_Runx1_Wu [32] , Mm_HSC_Tcf7_Wu [32] , Mm_LT_HSC_Venezia[36] , Mm_Proliferation_Venezia [36] , Mm_Quiescence_Venezia [36] , Polarity_factors_Ting [37] , Novel_HSC_regul_polar_Ting [37] , Grover et al , 2016 [38] , and Kowalczyk et al , 2015 [39] ( S6 Table ) . Cells were first individually analyzed , and significance of gene set enrichment was tested using hypergeometric test . Concordance between daughter pair was then visualized using radar plots and tested using goodness-of-fit test . Data were tested to meet normal distribution . The variance was similar between groups that were statistically compared . All data are plotted as mean plus 1 SEM unless otherwise stated . The SEM is used to indicate the precision of an estimated mean . Such a data representation does not affect the statistical analyses as variance information is utilized in the test statistics . A paired Student t test was used to determine the significance of the difference between means of 2 groups . One-way ANOVA or two-way ANOVA were used to compare means among 3 or more independent groups . Bonferroni post-test to compare all pairs of dataset was determined when overall p-value was < 0 . 05 . All statistical analyses were determined with Prism 4 . 0 c version . In order to choose sample size , we used GraphPad StatMate Software version 2 . 0b , estimating an SD between 2 and 8 ( depending on the experiment and the possibility of increasing sample size ) . For transplantation experiments , mice showing signs of sickness and with clear alterations of blood parameter and/or showing signs of major disease involving also nonhematopoietic tissues were excluded from analysis . As for in vitro experiments , samples were excluded from analysis in the case of clear technical problems ( error in immune-blotting or staining procedures or technical problems with reagents ) . All criteria for exclusions of samples from in vivo or in vitro experiments were preestablished . Each figure legend contains the number ( n ) of biological repeats ( samples obtained from experiments repeated on different days and starting from different mice ) of the statistical analysis . Mice for experiments were randomly chosen from our in-house colonies or from suppliers . All mice were C57BL/s6 females unless otherwise stated . The investigator was not blinded to the mouse group allocation nor when assessing the outcome ( Rag2−/−γc−/−KitW/Wv mice , aged mice , or young mice transplanted with aged BM stem cells require particular care and follow-up ) . | Stem cells are unique cells that can differentiate to produce more stem cells or other types of cells and can divide both symmetrically ( to produce daughter cells with the same fate ) and asymmetrically ( to produce one daughter cell that retains stem cell potential and one that differentiates ) . The mechanisms that control the outcome of stem cell divisions have been the focus of many studies; however , they remain mainly unknown . Here , we have analyzed these mechanisms in murine hematopoietic stem cells ( HSCs ) by directly comparing the epigenetic signature , the transcriptome , and the function of the two daughter cells stemming from the first division of either a young or an aged HSC . We observe that , while young HSCs divide mainly asymmetrically , aged HSCs divide primarily symmetrically . We find that the mode of division is tightly linked to stem cell polarity and is regulated by the activity level of the small RhoGTPase cell division control protein 42 ( Cdc42 ) . In addition , we show that the potential of daughter cells is further linked to the amount of the epigenetic mark H4K16ac and also to the amount of open chromatin allocated to a daughter cell , but it is not linked to its transcriptome . In summary , our study suggests that HSC polarity linked to Cdc42 activity drives the mode of division , while epigenetic mechanisms determine the functional outcome of the stem cell division . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"cell",
"physiology",
"medicine",
"and",
"health",
"sciences",
"cell",
"division",
"analysis",
"cell",
"cycle",
"and",
"cell",
"division",
"cell",
"processes",
"cloning",
"surgical",
"and",
"invasive",
"medical",
"procedures",
"cell",
"polarity",
"stem",
"cells",
"bioassays",
"and",
"physiological",
"analysis",
"molecular",
"biology",
"techniques",
"epigenetics",
"chromatin",
"research",
"and",
"analysis",
"methods",
"transplantation",
"blood",
"and",
"lymphatic",
"system",
"procedures",
"cell",
"analysis",
"animal",
"cells",
"chromosome",
"biology",
"gene",
"expression",
"molecular",
"biology",
"hematopoietic",
"stem",
"cells",
"cell",
"biology",
"stem",
"cell",
"transplantation",
"cell",
"transplantation",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types"
] | 2018 | Aging alters the epigenetic asymmetry of HSC division |
Dengue fever is a virus infection that is spread by the Aedes aegypti mosquito and can cause severe disease especially in children . Dengue fever is a major problem in tropical and sub-tropical regions of the world . We invited dengue experts from around the world to attend meetings to discuss dengue surveillance . We reviewed literature , heard detailed reports on surveillance programs , and shared expert opinions . Presentations by 22 countries were heard during the 2 . 5 day meetings . We describe the best methods of surveillance in general , the stakeholders in dengue surveillance , and the steps from mosquito bite to reporting of a dengue case to explore how best to carry out dengue surveillance . We also provide details and a comparison of the dengue surveillance programs by the presenting countries . The experts provided recommendations for achieving the best possible data from dengue surveillance accepting the realities of the real world ( e . g . , limited funding and staff ) . Their recommendations included: ( 1 ) Every dengue endemic country should make reporting of dengue cases to the government mandatory; ( 2 ) electronic reporting systems should be developed and used; ( 3 ) at minimum dengue surveillance data should include incidence , hospitalization rates , deaths by age group; ( 4 ) additional studies should be completed to check the sensitivity of the system; ( 5 ) laboratories should share expertise and data; ( 6 ) tests that identify dengue virus should be used in patients with fever for four days or less and antibody tests should be used after day 4 to diagnose dengue; and ( 7 ) early detection and prediction of dengue outbreaks should be goals for national surveillance systems .
Dengue virus , which is most commonly transmitted by the Aedes aegypti mosquito , is the most important mosquito-borne viral disease affecting humans [1] . Caused by one of four serotypes , dengue fever ( DF ) produces a spectrum of clinical illness that ranges from an influenza-like illness to a fatal shock syndrome ( DSS ) . Most patients that progress to shock first develop a more severe form of infection called dengue hemorrhagic fever ( DHF ) . We estimate that 3 . 6 billion people in 124 countries are at-risk for infection and 500 million people infected each year [2] . Over two million cases of DHF occur annually , and approximately 21 , 000 deaths are likely attributable to dengue [2] . Dengue prevention is limited vector control and treatment is limited to supportive care to avoid shock . To address the need for dengue prevention , several dengue vaccines are in development . One candidate entered expanded phase 2 clinical trials in 2009 [3] . Decision making prior to vaccine introduction and monitoring for effectiveness and safety after introduction require adequate country specific disease surveillance data [4] . To assess the status of dengue surveillance and to develop recommendation to improve surveillance data quality , two Dengue Prevention Boards convened to discuss dengue surveillance in representative countries . This report describes the results of that work .
As part of its program to facilitate the development and introduction of a dengue vaccine in endemic countries , Pediatric Dengue Vaccine Initiative ( PDVI ) [5] has sponsored two Boards consisting of dengue experts primarily from endemic countries , the Asia-Pacific Dengue Prevention Board ( APDPB ) and the Americas Dengue Prevention Board ( AmDPB ) [6] . These experts are in-country advocates for improved dengue prevention and control activities , most working in anticipation of dengue vaccines . The Boards meet regularly to assess various aspects of dengue prevention and control . Accurate burden of disease data will be needed for informed decision making regarding vaccine introduction [7]; however , often the only data available are from national surveillance . For this and other reasons , the Boards along with PDVI selected surveillance for their first topic to address . The format for their work was two working meetings of Board members and invited consultants and representatives from the Ministries of Health or other agencies involved in dengue surveillance . The objectives of the meetings were to assess the state of dengue surveillance in selected countries and reach a consensus on best practices . The Asia-Pacific Board met on June 19–21 , 2007; the Americas Board , on January 17–19 , 2008 . In addition to Board members , meeting attendees included national and international experts in surveillance and dengue , representatives of ministries of health , WHO and regional offices ( e . g . SEARO ) , PAHO and the Caribbean Epidemiology Center ( CAREC ) . Oral presentations , facilitated discussions , and a survey of presenters were used to determine the key issues and best practices . In total , there were presentations on the surveillance programs from twenty two countries given by representatives of Ministry of Health or other agency participating in dengue surveillance in-country ( e . g . Institute Pasteur ) ( for APDPB: Australia , Cambodia , French Polynesia , India , Malaysia , the Philippines , Sri Lanka , Singapore , Thailand , Japan , Vietnam; for AmDPB: Argentina , Brazil , Costa Rico , Colombia , Cuba , Honduras , Mexico , Puerto Rico , Nicaragua , United States ( South-west border states ) , and Venezuela ) ( Figure 1 ) . Because ensuring adequate surveillance requires participation from several disciplines , experts presented on topics of surveillance , epidemiology , entomology , and virology . Each country provided a detailed description of their national dengue surveillance system and results ( Table 1 & 2 ) . Attendees then synthesized the comments and opinions of the Board members . Full reports of each meeting are be available on the website of the Prevention Boards [6] .
The core functions of a comprehensive surveillance system are detection , reporting , investigation , confirmation , analysis , interpretation , and response . Cooperation is essential between the healthcare system and the public health authority because for rapid response to emerging public health threats the public health authority is dependent on healthcare system to generate timely and accurate case reports . At the time of the meeting , WHO had published guidelines on the diagnosis of dengue including case definitions; but these guidelines were published more than 10 years ago [8]—in 2009 WHO published new guidelines with major changes in dengue case classifications [9] . Regional offices have also drafted guidelines [10]–[12] . The guidelines agree on major issues with minor variations ( for example , some include leukopenia or hepatomegaly in the case definitions , but not all include a “suspected case” category ) . One major difficulty with all previous guidelines is case classification [13] . Because case fatality rates are much higher among patients with DHF , correct classification is important for triage , treatment , and prognosis . Obtaining a platelet count , hematocrit , and radiographic imaging is often not possible , too time consuming , or too expensive in many healthcare facilities in endemic countries—but the results of these tests are required diagnostic criteria for DHF . There was wide recognition of the need for a simplified classification system that is still helpful for case management [13] , [14] . Although meeting attendees reported using similar dengue case definition systems , surveillance methods varied between countries . Laboratory methods also vary as well as the testing algorithms and the interpretation of positives . For example , in Brazil and Colombia , healthcare providers complete case reports on both ambulatory and hospitalized patients , however , in Thailand and Vietnam the majority of reported cases are hospitalized . In only 12/22 ( 55% ) of countries represented at the meeting confirmed all officially reported cases with laboratory testing . Nearly every country includes suspected dengue cases regardless of age , but in Cambodia surveillance is conducted only among children less than 15 years of age . In Singapore and Brazil , monitoring vector indices is an integral part of the dengue surveillance system , while in Puerto Rico it is not . The attendees reported that these differences were not currently a problem for country level analyses but make inter-country , regional , and global analyses and comparisons difficult . Moreover , some difference ( e . g . lack of dengue surveillance among adults in Cambodia ) could be an impediment to strategic planning and implementation of a dengue vaccine since the disease also affects adults as well as children . Moreover , the vaccination of children is likely to also have an impact on adult disease burden [15] , [16] , further improving the cost-effectiveness . Since surveillance data are needed for health ministries to target control responses when outbreaks are detected , data must be collected in a timely fashion . In order to better understand the overall process , attendees reviewed the steps from infection to reporting ( Figure 2 ) . The incubation period is , on average , one week following the bite of an infectious mosquito . Several more days pass before symptoms become severe enough to cause the patient to seek medical attention , and still more time is required for the symptoms of DHF to develop . Outpatient clinic-based surveillance will detect cases earlier than inpatient facilities , potentially allowing more time for public health action . The medium for reporting ranged from paper case report forms , to hand-held computers , to internet-based systems . A case study from Nicaragua showed that hand-held computers , although initially requiring significant investment in infrastructure and training , do reduce reporting time . In Kolkata , India , special mapping of cases has been used to target control activities . In Singapore and Brazil , ministries are also using intranet-based data entry software allowing staff to directly enter data on cases and Ae . aegypti breeding sites in the field . The data are then immediately available to plan interventions and follow-up . All countries are dependent on paper forms for case reporting before any additional investigation or action . Time is required for that report to reach the surveillance office , to be entered and analyzed , and finally be reported . However , many countries are developing improved methods for data collection for targeted interventions . Another key issue is the needs of stakeholders with interests in dengue diagnosis and surveillance . These stakeholders include the general public , senior policy-makers , academics , and legislators . A diverse group , their interests range from the parents of sick children who want immediate and accurate test results—knowing the diagnosis allows them to cope better—to healthcare workers , staff in laboratories , public health and vector control authorities . All want a point-of-care test to speed accurate diagnosis and treatment and allow rapid public health intervention . Others ( e . g . general public , including travellers and Ministries of Health ) are more likely interested in more accurate tests to allow improved burden of disease estimates which could effect budget allocations for control . In most countries diagnostic testing and surveillance relies on healthcare practitioners and laboratory staff to report cases but they receive little benefit . Confirmatory diagnostic tests such as virus isolation or reverse transcriptase-polymerase chain reaction testing ( RT-PCR ) require expertise and equipment usually found only in reference laboratories . However , several attendees explained that the time required for a sample to reach and to be processed at centralized facilities often results in delays that render the results useless to the treating physician . Further delays occur if the information provide on a sample is incomplete or if batch-testing of samples is conducted . After testing , the report generated requires verification , approval and delivery ( e . g . mailing ) . As a consequence , health care providers in most countries must treat patients empirically [17] . The attendees concluded that simplified case reporting [18] , rapid turnaround of results , and training healthcare providers in reporting [19] can be important ways to encourage continued reporting of cases . Mandatory reporting , they explained , rarely guaranteed reporting .
Every dengue endemic country should systematically gather data in an established dengue surveillance system [12] , and each system should have a quality assurance mechanism . Legislation should make dengue a notifiable disease in every affected country [12] to improve the capture of cases by surveillance . However , even mandatory reporting is not sufficient; additional efforts are needed to improve and maintain a high level of quality reporting . All suspected cases must be reported to a central dengue unit in the health ministry as rapidly as possible and providers should be reminded that timely reporting can lead to effective response [26] . Laboratory confirmation of suspected cases should always be sought , except during outbreaks . Once an outbreak is confirmed no added information is gained by testing all samples; a subset of the samples is usually sufficient to track the outbreak [17] . That said , health providers should be informed that not all samples submitted during outbreaks will necessarily be tested . In outbreaks , data collection and analysis should be completed as rapidly as possible . Reporting should be encouraged from all levels of healthcare facilities in both the public and the private sectors [12] . In particular , mechanisms to involve the private sector should be developed; one possible way to encourage reporting is a rapid turn around of dengue diagnostic test results which can be provided free of charge[12] . While the turn around may not be quick enough to affect patient care , rapid return of results to submitting providers has intrinsic value for improving their diagnostic acumen . Reporting should be expanded to also include cases presenting to outpatient facilities , but staff in such settings may need further training to ensure the quality of data . To confirm and understand the burden of disease , periodic additional studies ( e . g . using capture-recapture methods ) should be conducted and incorporated into the system when possible . This will also determine the representativeness of the surveillance data . Laboratory confirmation improves the specificity of surveillance [27] , but laboratory methods and protocols should be standardized . This can be achieved through national and international networking of dengue laboratories to share expertise , protocols and data . A critical element for the successful laboratory diagnosis of an acute dengue infection is the timely collection of high quality samples . Monitoring the time from case identification to receipt of blood samples in the testing facility may assist in maintaining high quality specimens . RT-PCR and virus isolation are the two recommended methods for virus identification . Monitoring serotypes and sequencing isolates can provide useful markers for outbreak prediction [28] . Detection of the non-structural protein antigen NS1 may also be useful , but it must undergoing further evaluation [17] . For serologic testing , the hemagglutination-inhibition assay remains the gold standard of serological assays and should be maintained in those laboratories capable of performing it; however , enzyme-linked immunosorbent assays for IgM and IgG are considered the minimum requirement for confirmation of cases [29] . At least weekly reporting of aggregate results was considered by the attendees as the minimum standard during peak transmission . To conserve resources , reporting could be reduced to biweekly during periods of low transmission . During an outbreak , more frequent reporting , perhaps on a daily basis , would be useful . However , it is important to note that reporting would be affected by the operating hours of the reporting facilities ( e . g . facilities closed on weekends or holidays could artificially reduce reported cases and create surveillance artifacts ) . Reports should reach the surveillance units within 48 hours of form completion . Especially since dengue occurs frequently in young adults in Southeast Asia , it is recommended that the usual categories for reporting in health information systems should be used , namely less than 1 year , 1–4 years , 5–14 years and older than 15 years . However , reporting the median age of cases across all ages is also a useful statistic to track , and may be more useful for comparison if countries are using different age categories . Moreover , if the median age is reported by countries not including cases of all ages because the peak incidence is in children , the overall age distribution could be modeled with the available data . Electronic reporting systems should be developed and used broadly and such applications would facilitate formal reporting among countries . Meeting attendees emphasized the need to determine the incidence of severe cases through measurement of incidence rates of dengue fever , dengue hemorrhagic fever , and dengue shock syndrome , with hospitalization rates and mortality rates broken down by age group consistently applying the WHO regional case definitions . Weekly incidence of dengue , with data stratified by age , gender , and location should also be rapidly reported to allow effective use of vector control resources and to monitor intervention programs . In addition analyses should be conducted to detect and forecast dengue outbreaks through determination of the national threshold for outbreak alert and response [33] , to monitor the seasonality , age distribution , and transmission patterns and to evaluate and guide the introduction of potential dengue vaccines . Vector surveillance requires baseline data for comparisons . When relevant data are available , analyses should be conducted to identify locations and patterns of the vector population ( species , density , and vector-control indices ) and should also be used to monitor interventions ( with disease reduction as a measure of impact , and house index , container index , and Breteau index as indicators of outcome ) . In conclusion , the two Dengue Prevention Boards met to discuss the practice and logistics of dengue surveillance . The attendees applied their practical experience and discussed the strengths and weakness for the countries represented at the meeting . They then suggested best practices in dengue surveillance in endemic countries . For PDVI , improved surveillance serves many purposes including generating more accurate estimates of disease burden , further demonstrating the need for a dengue vaccine , supporting clinical evaluations of candidate dengue vaccines and providing more robust surveillance for monitoring the impact of the eventual introduction of dengue vaccines in national immunization programs . | The Pediatric Dengue Vaccine Initiative organized Dengue Prevention Boards in the Asia-Pacific and the Americas regions consisting of dengue experts from endemic countries . Both Boards convened meetings to review issues in surveillance . Through presentations , facilitated discussions , and surveys , the Boards identified best practices in dengue surveillance including: ( 1 ) Dengue should be a notifiable disease in endemic countries; ( 2 ) World Health Organization regional case definitions should be consistently applied; ( 3 ) electronic reporting systems should be developed and used broadly to speed delivery of data to stakeholders; ( 4 ) minimum reporting should include incidence rates of dengue fever , dengue hemorrhagic fever , dengue shock syndrome , and dengue deaths , and hospitalization and mortality rates should be reported by age group; ( 5 ) periodic additional studies ( e . g . , capture/recapture ) should be conducted to assess under-detection , under-reporting , and the quality of surveillance; ( 6 ) laboratory methods and protocols should be standardized; ( 7 ) national authorities should encourage laboratories to develop networks to share expertise and data; and ( 8 ) RT-PCR and virus isolation ( and possibly detection of the NS1 protein ) are the recommended methods for confirmation of an acute dengue infection , but are recommended only for the four days after onset of fever—after day 4 , IgM-capture enzyme-linked immunosorbent assay is recommended . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"public",
"health",
"and",
"epidemiology"
] | 2010 | Best Practices in Dengue Surveillance: A Report from the Asia-Pacific and Americas Dengue Prevention Boards |
Bdelloid rotifers are a class of microscopic invertebrates that have existed for millions of years apparently without sex or meiosis . They inhabit a variety of temporary and permanent freshwater habitats globally , and many species are remarkably tolerant of desiccation . Bdelloids offer an opportunity to better understand the evolution of sex and recombination , but previous work has emphasised desiccation as the cause of several unusual genomic features in this group . Here , we present high-quality whole-genome sequences of 3 bdelloid species: Rotaria macrura and R . magnacalcarata , which are both desiccation intolerant , and Adineta ricciae , which is desiccation tolerant . In combination with the published assembly of A . vaga , which is also desiccation tolerant , we apply a comparative genomics approach to evaluate the potential effects of desiccation tolerance and asexuality on genome evolution in bdelloids . We find that ancestral tetraploidy is conserved among all 4 bdelloid species , but homologous divergence in obligately aquatic Rotaria genomes is unexpectedly low . This finding is contrary to current models regarding the role of desiccation in shaping bdelloid genomes . In addition , we find that homologous regions in A . ricciae are largely collinear and do not form palindromic repeats as observed in the published A . vaga assembly . Consequently , several features interpreted as genomic evidence for long-term ameiotic evolution are not general to all bdelloid species , even within the same genus . Finally , we substantiate previous findings of high levels of horizontally transferred nonmetazoan genes in both desiccating and nondesiccating bdelloid species and show that this unusual feature is not shared by other animal phyla , even those with desiccation-tolerant representatives . These comparisons call into question the proposed role of desiccation in mediating horizontal genetic transfer .
The bdelloid rotifers are a class of microscopic invertebrates found in freshwater habitats worldwide . Two life history characteristics make these soft-bodied filter feeders unusual among animals . First , bdelloids famously lack males [1] or cytological evidence of meiosis [2 , 3] and are only known to reproduce via mitotic parthenogenesis . They are therefore one of the best-substantiated examples of a eukaryotic taxon that has evolved apparently without sex or meiosis for tens of millions of years [1 , 2 , 4 , 5] . Famously labelled ‘an evolutionary scandal’ [6] , bdelloids have diversified into over 500 species [7 , 8] defying the usual fate of asexual lineages [9–11] . Their persistence has implications for theories of the evolution of sex and recombination , a fundamental puzzle in biology [12–15] . A second key feature is that most bdelloid species are remarkably tolerant of desiccation and can survive the loss of almost all cellular water at any stage in their life cycle , including as adults [16 , 17] . As water evaporates , animals contract their bodies into flat , ellipsoid ‘tuns’ and enter a dormant state called anhydrobiosis , during which all metabolic activities associated with life are suspended [5 , 16 , 18] . Individuals can remain in this condition for long periods , usually days or weeks but occasionally several years [19 , 20] . The return of water restores metabolism and reproduction , with little evidence of negative fitness consequences for survivors [21] . Species that live in limnoterrestrial habitats such as puddles , leaf litter , and moss are subject to rapid and repeated cycles of drying . The ability to survive desiccation has been proposed to play a key role in bdelloid evolution [5 , 22] . Early marker-based analyses of bdelloid genomes recovered highly divergent gene copies that were interpreted as nonrecombining descendants of ancient former alleles [4] . Along with the low copy number of vertically inherited transposable elements ( TEs ) [23] , this result was considered positive genetic evidence of long-term asexual evolution . However , subsequent investigations of larger genomic regions revealed evidence of tetraploidy , probably arising from an ancient hybridisation or genome duplication event affecting diploid ancestors prior to the diversification of bdelloid families [5 , 24 , 25] . Genes generally have up to 4 copies , arranged as 2 pairs , with greater divergence between pairs ( ‘ohnologs’ , also known as homeologs in other polyploid systems ) than within pairs ( ‘homologs’ ) [5 , 24 , 25] . Another extraordinary feature was that a remarkably high proportion of bdelloid genes show similarity to nonmetazoan orthologs , mostly from bacteria but also fungi and plants , suggesting a rate of horizontal gene transfer ( HGT ) into bdelloid genomes at least an order of magnitude greater than that observed in other eukaryotes [26] . Many genes originating by HGT from nonmetazoans are expressed and functional [26 , 27] . The first whole-genome sequence for a bdelloid substantiated many of these findings [28] . The tetraploid genome of Adineta vaga comprises both homologous regions with low divergence ( median 1 . 4% for protein-coding genes ) and conserved gene order ( i . e . , high collinearity ) as well as ohnologous regions with much higher divergence ( median 24 . 9% ) and degeneration of gene order ( i . e . , low collinearity ) . The genome encodes remarkably few TEs ( approximately 3% of the genome ) but a high proportion of foreign genes ( approximately 8% ) , many of which occur as quartets and were therefore presumably acquired prior to tetraploidisation . Unusual structural features were also reported , including a large number of breaks in collinearity between homologous regions and linkage of homologs on the same assembly scaffold , often arranged as genomic palindromes . This assembly cannot , therefore , be decomposed into haploid sets , a finding that was interpreted as further evidence of long-term ameiotic evolution in A . vaga [28] . To what extent are the genomic characteristics of bdelloid rotifers explained by their unusual biology and ecology ? An important discovery was that bdelloids can survive doses of ionising radiation that would be lethal to nearly any other animal , owing to their ability to repair the resulting DNA double-strand breaks ( DSBs ) and recover from extensive genome fragmentation [29 , 30] . Experiments in A . vaga showed that comparable genome fragmentation also occurs during desiccation , and this led to the view that bdelloid genomes may be shaped by the need for recovering animals to repair DSBs arising from repeated desiccation [31] . One hypothesis is that homologous gene copies are used as reciprocal templates for repairing DSBs , a process that would act to homogenise homologous regions periodically via gene conversion and select against individuals with excessive divergence because template mismatches would disrupt DNA repair [24 , 28–31] . In this scenario , the molecular consequences of desiccation are directly linked to the patterns of intragenomic divergence observed in bdelloid genomes , via breakage and repair of DNA . However , the link between desiccation and DSBs is not unequivocal , and evidence from other anhydrobiotic taxa is mixed . For example , DNA integrity is largely maintained in desiccating tardigrades [32–34] , but not in the chironomid midge Polypedilum vanderplanki [35] . In bdelloids , a key prediction is that species that undergo desiccation more frequently should experience higher rates of DSB repair , resulting in more opportunities for gene conversion and thus a lower level of homologous divergence . A related hypothesis is that foreign DNA present in the environment may become incorporated into bdelloid genomes via nonhomologous recombination during DSB repair following desiccation , resulting in a higher rate of HGT than is experienced by other eukaryotes [26 , 28 , 30 , 31] . Evidence for high levels of nonhomologous transfer inspired further suggestions that DSB repair might even facilitate homologous horizontal transfer and genetic exchange between individual animals [26 , 28 , 30 , 31] . These ideas remain controversial , however , and recent claims of evidence for DNA transfer between individuals of A . vaga [36] have subsequently been identified as artefacts of experimental cross-contamination [37] . A separate recent study reported a striking pattern of allele sharing among 3 individuals of another bdelloid species , Macrotrachela quadricornifera , which was interpreted as evidence of sexual reproduction via an unusual form of meiosis ( similar to that of plants in the genus Oenothera ) [38 , 39] . However , no evidence for such a mechanism was apparent in a larger study of the genus Adineta [40] . In the absence of clear evidence for either occasional sex or horizontal genetic transfer between individual bdelloids ( but without discounting either possibility ) , the nature of bdelloid recombination remains an open question . Showing both extensive anhydrobiotic capabilities and putatively ancient asexuality , bdelloid rotifers sit at a unique junction in animal evolution . To better understand the relative contributions of these features to bdelloid genome evolution , we have taken advantage of natural variation in the capacity of species to survive desiccation by sampling and comparing whole genomes from multiple taxa . In particular , many species in the genus Rotaria live in permanent water bodies and do not survive desiccation in the laboratory [17 , 41] . Here , we present high-coverage , high-quality genome sequences for 3 species from 2 genera: the desiccation-tolerant species A . ricciae and the obligately aquatic , nondesiccating species R . macrura and R . magnacalcarata ( Fig 1 ) . These are compared with the published genome of A . vaga . Using a range of assembly approaches , we confirm the conservation of ancestral tetraploidy in all species but demonstrate substantial variation in genome size among species . We then test predictions regarding the effects of desiccation tolerance on intragenomic homologous divergence and investigate genome architecture within species to ask whether the unusual genomic structures observed in A . vaga are a general feature across bdelloids . Finally , we contrast a range of genome characteristics , including homologous divergence , HGT content , and repeat abundance across a wider range of animal taxa , allowing us to place some of the unique features of bdelloid genomes in a wider metazoan context .
Reference genome sequences for A . ricciae , R . macrura , and R . magnacalcarata were assembled using a combination of long- and short-read sequencing technologies ( S1 Table ) . Kmer spectra of raw and filtered sequencing reads indicated high ( >100x ) but variable coverage across sites in each genome ( S1 Fig ) . In addition , a large proportion of low-coverage kmers indicated substantial polymorphism in the R . magnacalcarata raw data , most likely corresponding to population variation in the multi-individual DNA sample collected for this species ( see Materials and methods ) . Contaminating reads from non–target organisms were excluded by scrutinising initial draft assemblies for contigs showing abnormal guanine-cytosine ( GC ) content , coverage , or taxonomic annotations , removing 3 . 1% of reads from the R . macrura dataset , 2 . 1% from R . magnacalcarata , and 6 . 3% from A . ricciae ( including approximately 9 Mb sequences annotated as Pseudomonas spp . ) ( S2 Fig ) . Given the complex patterns of intragenomic divergence and gene copy number observed in other bdelloid species [24 , 25 , 28] , we adopted 2 assembly strategies . First , reference assemblies were generated with a focus on high assembly contiguity . Reference sequences were constructed using the Platanus assembler [42] and improved using Redundans [43] ( S3 and S4 Figs ) . Second , ‘maximum haplotype’ assemblies were generated with a focus on maximum separation and resolution of homologous regions , even if this reduced assembly contiguity . This was intended to minimise the confounding effects of assembly ‘collapse’ , a phenomenon whereby homologous regions with no or low divergence are assembled as a single contig with 2-fold coverage relative to separately assembled regions . Further assembly metrics are provided in S1 Data . All assemblies have been submitted to DDBJ/ENA/GenBank under the project accession ID PRJEB23547 . Genome metrics for A . ricciae , R . macrura , and R . magnacalcarata reference assemblies are shown in Table 1 , alongside those for the published assembly of A . vaga ( accession GCA_000513175 . 1 , hereafter referred to as the ‘2013’ assembly ) [28] . The reference assembly of A . ricciae spanned 174 . 5 Mb , comprising 4 , 125 scaffolds with an N50 ( length-weighted median ) scaffold length of 276 . 8 kb ( Fig 2A ) . The reference assemblies for R . macrura and R . magnacalcarata spanned 234 . 7 and 180 . 5 Mb over 29 , 255 and 20 , 900 scaffolds , respectively , with N50 scaffold lengths of 73 . 2 and 53 . 3 kb . The proportion of undetermined bases ( i . e . , gaps denoted as Ns ) was low in all cases , accounting for 2 . 1% , 0 . 3% , and 0 . 8% of the A . ricciae , R . macrura , and R . magnacalcarata reference assemblies , respectively . The GC content was 35 . 6% for A . ricciae , 32 . 6% for R . macrura , and 31 . 9% for R . magnacalcarata; therefore , GC content in A . ricciae is higher relative to A . vaga ( 30 . 8% ) and both Rotaria species . Gene completeness was assessed by comparing sets of core eukaryotic genes to each reference assembly using Core Eukaryotic Gene Mapping Approach ( CEGMA ) and Benchmarking Universal Single-Copy Orthologs ( BUSCO ) [44 , 45] . Recovery of full-length CEGMA genes ( n = 248 ) was 98% , 94% , and 98% for A . ricciae , R . macrura , and R . magnacalcarata , respectively , and gene duplication ( average copy number per CEGMA gene ) was 2 . 9 , 1 . 6 , and 1 . 7 , respectively . The equivalent recovery of a larger set of BUSCO core metazoan genes ( n = 978 ) was 90% for all assemblies , with duplication scores of 2 . 0 , 1 . 2 , and 1 . 2 , respectively . The equivalent completeness and duplication scores for the A . vaga 2013 assembly were 96% and 3 . 0 for CEGMA , and 91% and 2 . 0 for BUSCO . The number of genes predicted from each reference assembly varied considerably among species . Gene prediction was performed using BRAKER [47] if RNA sequencing ( RNASeq ) data were available or MAKER/Augustus [48 , 49] if not , giving initial estimates of 55 , 801 , 26 , 284 , and 36 , 377 protein-coding genes for A . ricciae , R . macrura , and R . magnacalcarata , respectively ( Table 1 ) . Genes with BLAST matches to TEs ( E-value ≤1 × 10−5 ) and short genes with no matches to UniProt90 ( i . e . , likely spurious gene models ) were removed , resulting in ‘high-quality’ sets of 49 , 857 , 24 , 594 , and 29 , 359 protein-coding genes for downstream analyses ( Table 1; S2 Data ) . Reannotation of the A . vaga 2013 assembly using MAKER/Augustus resulted in 67 , 364 predicted genes , reducing to 57 , 431 after quality control ( S2 Data ) . Therefore , the reference genomes of R . macrura and R . magnacalcarata appear to encode approximately half the number of genes observed in A . ricciae and A . vaga . Correspondingly , the mean intergenic distance was higher in R . macrura ( mean 3 . 9 kb ) and R . magnacalcarata ( 2 . 2 kb ) than in A . vaga ( 1 . 4 kb ) or A . ricciae ( 1 . 2 kb ) ( S5 Fig ) . We checked for misannotations by comparing each set of predicted proteins to the corresponding assembly using TBLASTN ( E-value ≤1 × 10−20 ) . This did not reveal any highly similar matches ( discounting hits that overlapped with existing gene models ) , indicating that ‘missing’ Rotaria genes were not the result of poor gene prediction . However , the assemblies of all 4 species showed a large number of matches at lower similarity ( 30%–35% median identity at the amino acid level ) ( Fig 2C ) . These protein hits to putative noncoding regions may indicate pseudogenes , resulting either from degradation of coding regions following ancestral tetraploidisation or more recent duplications that have subsequently decayed and no longer encode functional proteins . The structure of predicted genes also varied among species . The average intron length was 104 and 108 bp for A . ricciae and A . vaga , respectively , but up to 3 times longer in R . macrura and R . magnacalcarata ( 362 and 208 bp , respectively ) ( Fig 2D , Table 1; S6 Fig ) . Distributions of intron lengths showed 2 distinct classes , with the majority of introns falling in the range 30 to 100 bp but a substantial minority showing a higher variance around a much larger mean ( inset of Fig 2D ) . The proportion of single-exon genes ( SEGs ) was substantially lower for R . macrura relative to other species , likely reflecting the lack of RNASeq guidance during annotation ( Table 1 ) . The repeat content of bdelloid assemblies was measured following 2 approaches: ( 1 ) comparisons to known metazoan repeats , sampled from Repbase , and ( 2 ) comparisons to Repbase plus an additional library modelled ab initio from each assembly , using RepeatModeler ( see Materials and methods ) . For ( 1 ) , the relative abundances of TEs were low for all species , with the total proportion of interspersed repeats accounting for 1 . 2% of the assembly span for both A . vaga and R . magnacalcarata , 0 . 9% for R . macrura , and 0 . 8% for A . ricciae ( S3 Data ) . Including simple and low-complexity repeats resulted only in modest increases , to 2 . 0% , 3 . 4% , 2 . 2% , and 3 . 0% for A . ricciae , A . vaga , R . macrura , and R . magnacalcarata , respectively . For ( 2 ) , however , the inclusion of ab initio repeats resulted in considerably increased repeat content for all species but to a greater extent in Rotaria ( 16 . 8% for A . ricciae , 18 . 4% for A . vaga , 22 . 0% for R . macrura , and 27 . 6% for R . magnacalcarata ) . A large proportion of ab initio repeats were marked as ‘unclassified’ , and their nature is yet to be determined ( Fig 2B; S3 Data ) . The composition of bdelloid genomes with respect to genome size evolution is considered further below . Our assembly results show an apparent 2-fold difference in the number of genes encoded by Adineta species relative to Rotaria species , suggesting substantial differences in either ploidy or divergence patterns between bdelloid genera . To investigate the evolutionary relationships among genes within each species , we estimated nucleotide divergence and collinearity among separately assembled gene copies using MCScanX [50] . This analysis identifies collinear blocks of genes , defined as pairs of genomic regions that show conserved gene order ( see Materials and methods ) . We plotted the average synonymous divergence ( KS ) between genes within each collinear block against a ‘collinearity index’ , defined as the number of collinear genes divided by the total number of genes within a given block ( following [28] ) . Both the A . ricciae reference and the ( reannotated ) A . vaga 2013 assembly showed a clear delineation of genes into both homologs ( low KS and high collinearity ) and ohnologs ( high KS and low collinearity ) ( Fig 3A; S4 Data ) , as has been observed previously for A . vaga [28] . The number of A . ricciae genes that form homologous collinear blocks is 36 , 593 ( 73 . 4% total genes ) ; for A . vaga , it is 37 , 061 ( 64 . 5% ) ( S2 Table ) . Comparisons between Adineta species show approximately half as many homologous collinear blocks in A . ricciae relative to A . vaga ( 475 versus 905 ) , but these contain twice as many genes ( median 24 versus 11 ) . Therefore , the extent to which we have successfully captured homologous gene copies in A . ricciae appears to be at least equivalent to that for A . vaga . Strikingly , however , only ohnologous relationships are inferred in Rotaria genomes: collinear blocks composed of homologous genes are not observed ( Fig 3A ) . Comparisons of ohnologous blocks across all species also suggest that the extent of ohnologous collinearity is higher in Adineta species than in Rotaria species ( more ohnologous blocks comprising a greater proportion of genes ) , notwithstanding confounding factors such as differences in the level of assembly fragmentation ( S2 Table ) . Assuming that the ancestor of extant Rotaria lineages was also tetraploid [25] , the apparent ‘loss’ of homologous copies in Rotaria species may be caused by either ( 1 ) the genuine loss of homologous gene copies from Rotaria genomes—resulting in a shift from tetraploidy to highly diverged diploidy—or ( 2 ) extremely low levels of divergence between Rotaria homologs , such that the majority of homologous sites are identical and cannot be separately assembled ( i . e . , are collapsed ) . To differentiate between these hypotheses , we characterised patterns of nucleotide polymorphism and read coverage across each genome , as has been used to investigate the genomes of other polyploid or asexual species [51–55] . Widespread assembly collapse of homologous regions should result in single-nucleotide polymorphisms ( SNPs ) with a frequency around 50% and a total coverage ( read depth of reference plus alternative bases ) that is approximately equal to the genome-wide average , analogous to collapsed heterozygous sites in a segregating diploid genome ( e . g . , see Fig 2A of [42] ) . These patterns are not predicted under the hypothesis of gene loss in an uncollapsed assembly ( i . e . , all haplotypes separately assembled ) , where SNPs may arise in repetitive regions ( TEs , tRNAs , low-complexity regions , etc . ) but are unlikely to show a frequency of 50% or consistent read depth . Reads were mapped to the reference assembly of each species , using single-clone ( A . ricciae and A . vaga ) or single-individual ( R . macrura and R . magnacalcarata , whole genome amplified [WGA] ) libraries . High-quality biallelic SNPs showing a minor allele frequency ( MAF ) distributed around 50% were detected in all assemblies , indicating at least partial collapse in all cases ( Fig 4A , S5 and S6 Data ) . The relative platykurtosis observed in Rotaria species may be an artefact of WGA ( inflation of low-frequency SNPs ) or lower coverage in general ( S7 Fig ) . In A . vaga , the majority of sites ( approximately 76% ) show coverage around 90x , representing separately assembled regions , with a minor peak at 180x , representing collapsed regions ( Fig 4B; S8 Fig ) . The majority of SNPs occur in regions of 180x coverage , as would be expected under the scenario of localised assembly collapse [28] . For both R . macrura and R . magnacalcarata , however , read depth at SNP sites ( reference plus alternative alleles ) varied in concert with the genome-wide coverage distribution ( Fig 4B ) . These patterns indicate that the majority of SNPs occur in collapsed regions , supporting the hypothesis of widespread assembly collapse in Rotaria species . A different pattern is observed in A . ricciae , however . Here , a small proportion of sites ( 11% ) are distributed around a peak at 75x coverage , which presumably represents the 1-fold coverage value , but the majority of sites ( 81% ) show 150x coverage and are thus presumably 2-fold covered ( i . e . , present in double copy ) ( Fig 4B; S9 Fig ) . Furthermore , SNP depth is unimodal and is centred on the 150x coverage peak , indicating that the majority of variant sites occur in regions of putative 2-fold coverage . Given the successful capture of the majority of homologous gene copies in A . ricciae ( approximately 73% , S2 Table ) , we infer that these conflicting signals are likely derived from another source of coverage heterogeneity that is unrelated to homologous collapse . This is unlikely to be due to an additional whole-genome duplication in A . ricciae , given that both Adineta species have 12 chromosomes [56 , 57] , but may be caused by other phenomena that affect DNA stoichiometry at the level of either the genome ( e . g . , segmental or partial genome duplications ) or the sample itself ( e . g . , endopolyploidy [58] or cryptic population structure ) ( S1 Text , S10 Fig ) . Further investigations of the A . ricciae genome are required to test these hypotheses . A . ricciae displayed a further difference from other bdelloid genomes: a clear elevation in KS , both for homologs ( compared to A . vaga; mean KSAr = 0 . 135 versus KSAv = 0 . 05; t = 47 , P < 0 . 01 ) and for ohnologs ( e . g . , mean KSAr = 1 . 267 versus KSAv = 0 . 613; t = 124 , P < 0 . 01 ) ( Fig 3B; S3 Table ) . No such elevation was observed in the rate of nonsynonymous substitution in A . ricciae , compared to the other species . However , the A . ricciae genome also shows the highest GC content of the 4 species ( approximately 5% higher than A . vaga , and 3% to 4% higher than either Rotaria species ) . Therefore , one explanation for the increase in KS may be selection for increased GC content in A . ricciae , with continued purifying selection at nonsynonymous sites [59] . In the A . ricciae reference and A . vaga 2013 assemblies , the majority of homologs were separately assembled , allowing for the identification of homologous gene copies and the estimation of their sequence divergence using a BLAST-based approach [51 , 55] . The median divergence between separately assembled homologous gene copies was 4 . 55% ( mode = 3 . 75% ) in A . ricciae and 1 . 42% ( mode = 1 . 25% ) in A . vaga ( in agreement with [28] ) ( Fig 5A ) . However , these estimates may be inflated because they fail to consider homologous regions with low divergences that are collapsed . Alternative estimates of homologous divergence based on SNPs detected in the collapsed A . vaga assembly were correspondingly lower ( 0 . 955% and 0 . 788% , based on alignment of libraries ERR321927 and SRR801084 , respectively ) ( S5 Data ) . Exploration of alternative assembly strategies , which aimed to minimise as much as possible the phenomenon of assembly collapse , did not result in the separate assembly of Rotaria homologous regions ( S1 Data ) , indicating substantially lowered homologous divergence relative to Adineta . To estimate the divergence between collapsed Rotaria homologous gene copies , we instead counted the number of SNPs occurring in coding regions in each assembly . Based on single-individual , WGA mate-pair libraries aligned to the R . macrura and R . magnacalcarata reference assemblies , a total of 13 , 115 and 36 , 594 SNPs were detected across 40 . 0 and 40 . 3 Mb of coding sequences ( CDSs ) , respectively ( S5 and S6 Data ) . Assuming that all detected SNPs are the result of homologous collapse , an upper limit for the divergence between homologs is estimated at 0 . 033% and 0 . 075% for R . macrura and R . magnacalcarata , respectively ( Fig 5B ) . These results indicate that homologous divergence in nondesiccating Rotaria species is at least an order of magnitude lower than that observed in anhydrobiotic Adineta species . This contradicts hypotheses that emphasise the role of desiccation in shaping patterns of divergence in bdelloid genomes . If the rate of desiccation-induced DSB repair is positively correlated with the rate of gene conversion , a lower level of homologous divergence is expected in species with higher rates of desiccation . In fact , we observe the opposite: divergence between homologs in nondesiccating Rotaria species is considerably lower than in A . ricciae ( median 4 . 6% ) , A . vaga ( 1 . 4% ) ( here and [28] ) , and Philodina roseola ( 3% to 5% ) [24] , all of which are capable of anhydrobiosis . Across sexual eukaryotes , estimates of allelic divergence range from about 0 . 01% to 8% [60] . Therefore , none of these bdelloid species is beyond the range of observed values for sexual taxa , although some fall near the extremes of this distribution ( e . g . , Rotaria are towards the lower end , in contrast to A . ricciae and P . roseola ) . How can we reconcile theory with these observations ? The simplest explanation is that homologous divergence is unlinked to desiccation and that the observed differences are instead reflective of underlying phylogeny . Alternatively , it may be that homogenisation between homologous gene copies in Rotaria is not caused by desiccation-induced DSB repair but by gene conversion arising during a different process , such as mitotic crossing over [61–63] . In the yeast Saccharomyces cerevisiae , for example , various forms of mitotic recombination can produce tracts of gene conversion many kilobases long , often initiated from DNA nicks that are subsequently processed into DSBs [64–66] . Such processes may be especially pronounced and irreversible in asexuals: rapid loss of heterozygosity is observed in recent asexual lineages of the water flea Daphnia pulex , in which high rates of initial heterozygosity in hybrid asexual lineages are rapidly eroded via gene conversion and hemizygous deletion , which may ultimately limit their longevity [67] . However , this begs the question: why should the same homogenising mechanisms not operate in desiccation-tolerant species ? One possibility is that desiccation-tolerant species have low or negligible background rates of gene conversion while hydrated , thanks to selection for highly effective DNA repair and error-checking mechanisms imposed by environments that desiccate on a regular basis . Such a repair system might faithfully prevent loss of diversity in the context of mitosis , even while occasional gene conversion remains an unavoidable consequence of the more demanding repairs required after desiccation . Alternatively , perhaps similar homogenising forces do indeed operate in desiccation-tolerant rotifers but are counteracted by mutations generated during repair of desiccation-induced DSBs , whose net effect is to sustain high rates of homologous divergence [68] . Positive evidence for a link between desiccation and DSBs in bdelloids is currently limited to experiments in a single species , A . vaga [31] , and evidence from other anhydrobiotic taxa is mixed [32 , 33 , 35] . Further work on DNA integrity and genome evolution in bdelloids is needed to address these divergent predictions . A final explanation that cannot be entirely excluded is that low homologous divergence in Rotaria genomes results from cryptic sexual reproduction , constrained by small population sizes ( i . e . , inbreeding ) , although no males have so far been detected in Rotaria or any other bdelloid [69] . The A . vaga 2013 assembly ( accession GCA_000513175 . 1 ) showed a number of unusual structural genomic features , including breaks in homologous collinearity , physical linkage of homologous genes ( i . e . , encoded on the same scaffold ) , and genomic palindromes of the form g1A , g2A , g3A…g3B , g2B , g1B , where A and B denote homologous copies of genes g1 , g2 , and g3 . Such features would result in chromosomes that cannot be decomposed into haploid sets and thus imply a genome architecture that is incompatible with conventional meiotic pairing and segregation , as might be predicted under the hypothesis of long-term asexuality [28] . To test for such structures in other bdelloid genomes , we first analysed the reannotated A . vaga 2013 assembly . A total of 298 breaks in collinearity ( 32 . 9% of 905 homologous blocks ) were detected ( an example is shown for scaffold AVAG00001 in Fig 6A ) . In addition , 25 homologous blocks were encoded on the same genomic scaffold , 2 as tandem arrays and 23 as palindromes ( Fig 6B ) . Thus , our detection methods were able to recover the same signals of ameiotic evolution reported by Flot et al . ( 2013 ) [28] , for the same A . vaga assembly . The method of Flot et al . ( 2013 ) was to construct contigs from Roche 454 Titanium and GS-FLX data using the MIRA assembler [70] , followed by correction and scaffolding using high-coverage Illumina paired-end and mate-pair data ( section C1 of [28] supplement ) . We attempted to reassemble the A . vaga paired-end Illumina data independently ( incorporating both mate-pair and 454 data for scaffolding ) using a variety of established short-read assemblers . However , this consistently resulted in highly fragmented assemblies ( e . g . , N50 of approximately 1 to 2 kb ) , except when allowing for the collapse of homologous regions ( S1 Data ) . The lack of contiguity in A . vaga maximum haplotype assemblies precluded us from using alternative assembly approaches to investigate the features detected in the 2013 assembly . A lack of separately assembled homologous gene copies in either Rotaria species similarly precluded structural analysis . The closely related species A . ricciae , however , showed both high assembly contiguity and a majority of separately assembled homologous gene copies . For this species , we detected only 8 collinearity breaks ( 1 . 7% of 466 homologous blocks ) between homologs ( an example is shown in Fig 6C ) , in contrast with hundreds inferred from the A . vaga 2013 assembly . Assuming that the A . ricciae and A . vaga 2013 assemblies are structurally accurate and that the homologous gene copies captured in both assemblies reflect the ancestral tetraploidisation common to all bdelloids , collinearity appears to be markedly more conserved in A . ricciae relative to A . vaga . However , many of the detected breaks in both species span regions separated by scaffold gaps ( Ns introduced during the joining of contigs ) , suggesting that at least some detected collinearity breaks may be the result of scaffolding errors despite requisite care during assembly ( Fig 6A and 6C ) . For example , an unscaffolded A . ricciae assembly showed only a single break in collinearity , although the increased fragmentation of this assembly ( N50 = 18 . 7 kb ) may limit our ability to detect such breaks . We did not detect any cases of homologous genes arranged as palindromes in A . ricciae: only 2 cases of linked homologous blocks were detected , both tandem repeats ( Fig 6D ) . Overall , these results suggest that certain unusual genomic features , previously interpreted as positive signatures of long-term ameiotic evolution in A . vaga [28] , are largely absent from the closely related A . ricciae ( and remain untested in other bdelloids ) . These patterns may reflect true differences between Adineta species , although no marked dissimilarity in karyotype is evident [56 , 57] . Alternatively , they may be either false-positive or false-negative artefacts of applying alternative assembly methodologies to complex genomes with different patterns of intragenomic divergence . Evidence from other taxa is limited . For example , similar features have been reported in the recently assembled genomes of the parthenogenetic springtail Folsomia candida [71] and in certain apomictic species of Meloidogyne root-knot nematodes [72] . However , these involved only small proportions of each respective genome , and the latter study used the same assembly approach that was applied to A . vaga [72] . Furthermore , recent investigations of genome evolution in asexual , nondesiccating Diploscapter nematodes have revealed a high degree of collinearity among homologous genes , despite high levels of divergence ( approximately 4% , thus similar to A . ricciae ) [54 , 73] . These data suggest that transitions to asexuality do not necessarily lead to the erosion of collinearity . Similar structural variants are also detected in many sexual organisms—for example , humans [74 , 75] , cichlid fishes [76] , and cows [77]—and may involve translocations or duplications that are many kilobases in length . Therefore , further work is required to improve and validate assembly contiguity of bdelloid genomes and to ascertain the evolutionary significance of these features . Although variation in genome size among bdelloids has been inferred previously based on cytofluorometry of oocytes [78] , there are some inconsistencies between reported values that indicate possible errors ( see S2 Text for details ) . Based on our assembly results , estimations of global genome properties such as total span and gene number can be estimated bioinformatically , using both kmer- and assembly-based approaches . The maximum haplotype assembly for A . ricciae was approximately 201 Mb in length encoding 63 , 000 genes , while reannotation of the partially collapsed 217-Mb A . vaga 2013 assembly showed approximately 67 , 000 genes . While maximum haplotype assemblies for A . vaga were highly fragmented ( and therefore poorly annotated ) , a collapsed A . vaga assembly , reduced to 109 Mb , encoded 31 , 600 genes . Notwithstanding potential complexities of coverage heterogeneity in the A . ricciae data ( discussed above ) , these values suggest that the full complement of genes in A . ricciae may be in the region of 60 , 000 to 65 , 000 genes across a total span of about 200 Mb ( S11 Fig ) . The genome of A . vaga is likely to be of approximately equivalent size [28] . The largely collapsed reference assemblies of R . macrura and R . magnacalcarata showed about 25 , 000 and 35 , 000 genes , respectively , and thus are in broad agreement with observations from Adineta . This also implies that the total genome size for Rotaria is in the region of 400 to 500 Mb , assuming the majority of sites are in double copy ( S11 Fig ) , and indicates that the genomes of Rotaria species may be considerably larger relative to Adineta species . What mechanisms might explain these observed differences ? Based on comparisons to known metazoan TEs from Repbase , the abundance of TEs and low-complexity repeats was low in all species , suggesting that expansions of known TEs or simple repeats in the Rotaria lineage is unlikely to be a major driver . However , the inclusion of the largely unclassified ab initio repeats did result in a marked increase in total repetitive sequences for all species ( 17% , 18% , 22% , and 28% for A . ricciae , A . vaga 2013 , R . macrura , and R . magnacalcarata , respectively ) . The relative increase is greatest in the Rotaria species , suggesting that a substantial fraction of the R . macrura and R . magnacalcarata reference assemblies are covered by repeats whose exact nature remains to be elucidated . In addition , average intron sizes in Rotaria genes are longer ( by at least 100% ) , driven primarily by an increase in the number of long introns . Intriguingly , a similar association between desiccation tolerance and genome ‘compaction’ has been observed in tardigrades: Hypsibius dujardini has a genome size of 104 Mb and only survives desiccation under certain conditions , whereas Ramazzottius varieornatus has a much smaller genome size of 56 Mb and is capable of rapid anhydrobiosis [34 , 79] . Future sampling of more phylogenetically independent comparisons of desiccating and nondesiccating species is needed to test these ideas . To better understand how bdelloid genomes compare to those of other metazoans , we characterised an additional 13 species from across the Protostomia ( S4 Table ) based on genome size , gene density , patterns of orthologous gene clustering , HGT content , and repetitive sequence content ( Fig 7 , S12 Fig , S5 and S6 Tables ) . Our comparison included a broad taxonomic range of species from different ecological niches , including molluscs [80–83] , annelids [81] , the platyhelminth Schistosoma haematobium [84] , the desiccation-tolerant tardigrade R . varieornatus [34] , the orthonectid intracellular parasite Intoshia linei [85] , and chromosomal-level reference genomes for Caenorhabditis elegans [86] and Drosophila melanogaster [87] . Phylogenetic relationships among species were not estimated directly but inferred from the literature [85 , 88 , 89] . We assessed the extent of horizontal transfer into protostome genomes using both sequence comparison and phylogenetic approaches . The extent to which HGT contributes to the genomes of multicellular eukaryotes is controversial . For example , a recent claim of 17% nonmetazoan genes encoded in the genome of the tardigrade H . dujardini was later shown to be derived mostly from contaminating non–target organisms [91–94] . Nonetheless , a high proportion of genes from a variety of nonmetazoan sources has consistently been inferred in bdelloid genomes from a range of independent data , including fosmid sequences [26] , transcriptomes [27 , 41] , and whole-genome data [28] . To measure the level of horizontal transfer , we developed an HGT assessment pipeline that uses both sequence comparison and phylogenetic approaches to build a body of evidence for the foreignness of each predicted gene . Our goal was not to unequivocally assert the evolutionary history of individual genes but rather to apply these tests consistently across the set of animal genomes as a fair comparison for estimating HGT . Our initial screen identified 6 , 221 ( 12 . 5% ) , 8 , 312 ( 14 . 5% ) , 3 , 104 ( 12 . 6% ) , and 3 , 443 ( 11 . 7% ) genes from A . ricciae , A . vaga , R . macrura , and R . magnacalcarata , respectively , as HGT candidates ( HGTC ) ( Fig 7D , S7 Data ) . These values are substantially higher than the proportion of HGTC observed in any other protostome species included in this analysis , using the same pipeline and thresholds ( the highest proportion of HGTC for a non-bdelloid was 3 . 6% , for the annelid worm Capitella teleta ) . This is also noticeably higher than estimates based solely on the Alien Index . For each HGTC , we then assessed ( 1 ) the presence of predicted introns , ( 2 ) scaffold linkage to another gene of unambiguous metazoan origin , ( 3 ) presence on a scaffold that encodes a high HGTC proportion that might indicate contamination , ( 4 ) membership within a ‘clan’ of nonmetazoan orthologs , and ( 5 ) monophyly of the HGTC with all present nonmetazoan orthologs to the exclusion of all metazoan orthologs ( see Materials and methods ) . Testing for clan membership with nonmetazoan orthologs reduced the proportion of HGTC to 9 . 1% , 7 . 6% , 6 . 2% , and 6 . 5% for the 4 bdelloids , compared with <1% for all other species ( S7 Data ) . The final test was not applicable for the majority of HGTC because metazoan orthologs were often not detected; thus , the number of genes that additionally showed evidence for monophyly with nonmetazoan orthologs was 189 , 190 , 111 , and 82 for A . ricciae , A . vaga , R . macrura , and R . magnacalcarata , respectively , and was reduced to a handful or 0 in all other species . Sequential BLAST analysis of HGTC ‘clan’ genes showed that the majority ( approximately 80% ) were found in all 4 bdelloids , suggesting that most detected HGTC genes are of ancient origin ( S13 Fig ) . Correspondingly , many HGTC genes were also found as either pairs or quartets ( for A . ricciae and A . vaga ) or as singletons or pairs ( for R . macrura and R . magnacalcarata ) ( S13 Fig ) . These comparisons support previous findings of a high proportion of nonmetazoan genes in bdelloid genomes [26 , 28 , 41 , 95] . Compared to other metazoans , and at all levels of scrutiny , the 4 bdelloid genomes analysed here showed a substantially greater proportion of genes from nonmetazoan sources than do any other species in our comparison . Our results confirm a substantial proportion of foreign genes in the nondesiccating Rotaria genomes , in agreement with recent findings based on transcriptomes [41] . Our assessment also showed very low levels of HGT ( approximately 1% ) into the genome of the anhydrobiotic tardigrade R . varieornatus , in agreement with recent estimates [34 , 79] . In addition , recent genome investigations of the anhydrobiotic chironomid insect P . vanderplanki , which experiences a large number of DNA breakages during desiccation [35] , also did not reveal an elevated rate of HGT [96] . Taken together , these findings bring into question the association between anhydrobiosis and elevated rates of HGT that previously has been suggested for bdelloids [26 , 28 , 30 , 31 , 95] . One explanation may be that differences in HGT content reflect species-specific differences in the mechanism of anhydrobiosis , in combination with particular ecological properties of each species . Further comparative work is thus required to elucidate any relationship between anhydrobiosis and horizontal transfer . An alternative possibility is that HGT content in bdelloids does not reflect a deviation in the rate of import of foreign genes but an increased rate of retention arising from their putative longstanding asexuality . Based on transcriptome data from Rotaria species , Eyres et al . ( 2015 ) estimated the rate of gain to be low in absolute terms , on the order of approximately 10 HGT gains per lineage per million years [41] . Perhaps this is a typical background rate of import for organisms with similar ecological and physiological properties to bdelloid rotifers , but acquired genes are able to persist for longer in an ameiotic background given the lack of mechanisms such as segregation and unequal crossing over that would otherwise remove them . If so , foreign genes incorporated by asexuals , even if initially deleterious , might persist over the extended timescales necessary for domestication . The high proportion of nonmetazoan genes accumulated in bdelloid genomes may therefore owe more to a long-term lack of meiotic sex than to anhydrobiosis . We also quantified the abundance of TEs and low-complexity repeats in each animal genome . We chose to focus on the quantification of known repeats and thus did not perform ab initio repeat modelling for non-bdelloid species . There was considerable variation in TE abundance among species ( Fig 7E ) , with the total proportion of genome covered by interspersed repeats varying from 0 . 3% in the tardigrade R . varieornatus to 27 . 5% in the oyster Crassostrea gigas ( S3 Data ) . The relative abundance of different classes of repeats , including long interspersed elements ( LINEs ) , short interspersed elements ( SINEs ) , long terminal repeats ( LTRs ) , and DNA elements , also differed greatly among taxa , as did the amount of simple and low-complexity repeats . The proportion of total repeats ( TEs plus low-complexity repeats ) ranged from 0 . 6% in R . varieornatus to 42% in the annelid worm Helobdella robusta . All 4 bdelloid species display a low abundance of TEs , in agreement with previous findings [23–25 , 28 , 97] . However , 2 other species also show low levels of TEs: I . linei , an intracellular parasite of marine invertebrates with a highly reduced genome ( 42 Mb ) [85] , and R . varieornatus , also with a relatively small genome ( 56 Mb ) [34] . In fact , R . varieornatus encodes the fewest TEs of the species analysed here ( 0 . 6% as a proportion of assembly span ) , followed by the 4 bdelloids ( 2%–3% ) . These estimates of TE abundances in I . linei and R . varieornatus are substantially lower than the total repeat content of these genomes ( 28% and 20% , respectively ) , which includes a high proportion of ab initio repeats ( inferred directly from the assembled nucleotides ) marked as ‘unclassified’ ( accounting for approximately 18% and approximately 19% total repeats , respectively [79 , 85] ) , matching our finding of higher ab initio repeat content in bdelloids . Additional work is required to elucidate the nature of these unclassified repeats in bdelloids and in other taxa . What evolutionary forces may explain the low abundance of TEs in these species ? Asexuality and anhydrobiosis have both previously been posited as factors contributing to the low number of TEs in bdelloid rotifers . For example , under long-term asexual evolution , TEs may proliferate freely within a genome and thus drive that lineage to extinction ( an extension of Muller’s ratchet ) or become lost , domesticated , or otherwise silenced [30 , 98–100] . Frequent cycles of desiccation and rehydration may also favour the evolution of reduced repeat content , via selection against deleterious chromosomal rearrangements brought about by ectopic recombination of TEs during the repair of DSBs [29 , 30] . Our comparisons did not detect any substantial variation in the abundance of known TEs between desiccating ( 1 . 2% and 0 . 8% for A . ricciae and A . vaga , respectively ) and nondesiccating ( 0 . 9% and 1 . 2% for R . macrura and R . magnacalcarata , respectively ) species , despite a considerable increase in the inferred genome size of Rotaria species . Moreover , the proposed mechanism involving desiccation relies on DSB repair during rehydration , a process which is presumably limited in the aquatic species R . macrura and R . magnacalcarata and may also not apply in the case of R . varieornatus , whose DNA is protected during anhydrobiosis [34] . However , the vast majority of bdelloid rotifers are resistant to desiccation , suggesting that anhydrobiosis was probably the ancestral state [17] . Therefore , it may be that TEs and other repeats were already largely eradicated in the most recent common ancestor to nondesiccating Rotaria species , prior to their adaptation to a fully aquatic lifestyle and loss of anhydrobiosis . Finally , we also tested for the presence of a suite of 41 sex-related genes [101] in bdelloids using both TBLASTN ( comparing to the genome ) and HMMER ( comparing to the proteome ) . Tested genes included 11 associated with meiosis , 19 involved in recombinational repair , 6 involved in DNA damage detection , 4 involved in DSB repair via nonhomologous end-joining , and 1 involved in bouquet formation ( S8 Data ) . A positive match using TBLASTN and/or HMMER was recorded in at least 1 bdelloid species for all tested genes ( 40 of 41; 98% ) with the exception of RED1 , which is involved in crossover regulation and was not detected in any bdelloid at any significance threshold ( S8 Data ) . However , RED1 was not detected in D . melanogaster and only as a poor match in C . elegans and thus may represent an ancestral loss that predates the bdelloids . Overall , these findings suggest that bdelloids do encode the majority of genes involved in meiosis and sex-related functions . However , the presence of these genes does not necessarily indicate the presence of sex or meiosis because they are likely to be retained for other functions related to homologous recombination and DSB repair [102] .
The bdelloid rotifers have drawn attention because 2 features of their life history are remarkable among metazoans: their apparent ancient asexuality and their ability to withstand desiccation at any life stage . In this work , we have generated whole-genome sequence data for 3 additional bdelloid species with the overall aim of assessing hypotheses regarding the contributions of asexuality and anhydrobiosis to their genome evolution . We find that both desiccating and nondesiccating species are ancestrally tetraploid , in agreement with previous work , but that homologous divergence in nondesiccating Rotaria species is substantially lower than that observed in anhydrobiotic Adineta species and may be low even compared to estimates of allelic heterozygosity from sexual eukaryotes . This finding runs counter to predictions based on current hypotheses regarding the genomic effects of desiccation and thus requires a reevaluation of the causes and consequences of intragenomic interactions between bdelloid homologs . Comparisons of genome architecture revealed that a number of unusual genome features posited as evidence of long-term ameiotic evolution in A . vaga were largely absent from the closely related species A . ricciae , for which a comparable assembly is now available . In addition , we find that bdelloids encode the majority of genes that are required for meiosis and syngamy in sexual taxa but emphasise that the precise function of these genes in bdelloids is currently unknown . We reconfirm previous reports that bdelloids encode a high proportion of nonmetazoan genes . Here too , a role for desiccation tolerance had been hypothesised . We find that high HGT content is a potentially unique feature of bdelloid genomes among animals , but comparisons to other desiccation-tolerant taxa raise questions about the role of anhydrobiosis . Our extensive assembly results also allow for a refinement of the global parameters of bdelloid genomes and suggest substantial genome size differences between genera . The phylogenetic nonindependence of our comparative analysis currently precludes any certainty in linking these observed trends to desiccation tolerance . Further elucidation will be possible when data for anhydrobiotic species within Rotaria become available in the future . Overall , we conclude that many features of the bdelloid genomes analysed here are not markedly inconsistent with those found in sexual taxa , except for the remarkably high prevalence of HGT . Finally , we hope that our approach may offer useful guidance for future studies involving the de novo assembly of non–model organisms with complicated genome characteristics from complex raw data . Our goal was to explore the assembly parameter space for each dataset , taking into consideration a number of potential confounding factors including polyploidy , intragenomic divergence , and sample polymorphisms . Our assembly results showed good contiguity and gene-completeness metrics , indicating a high level of overall quality . Nonetheless , we reiterate the caution that a full understanding of genome architecture and evolution in bdelloid rotifers will be possible only with highly contiguous , chromosome-level assemblies , towards which future efforts will be directed .
Clonal cultures of A . ricciae [103] rotifers were grown as previously described [27 , 104–106] . Briefly , rotifers were grown in T75 tissue culture flasks ( Nunc ) with 15 to 25 ml ddH2O and fed twice a week with 10 μl of either bacteria ( Escherichia coli TOP10 [ThermoFisher] in water ) or a solution of yeast extract and peptone ( 2 . 5% w/v each ) . Approximately 50 , 000 rotifers were starved overnight before collection and harvested by centrifugation at 10 , 000 g for 5 minutes before treatment according to the relevant DNA or RNA extraction protocol . A starter culture for R . macrura was generated from approximately 100 wild-caught animals isolated from a small pond near Lake Orta , Italy . Populations were grown in sterile distilled water and fed with autoclaved and filter-sterilised organic lettuce extract . Prior to DNA extraction , animals were washed twice in sterile distilled water and starved overnight ( approximately 16 hours ) before being washed again with HyPure molecular-grade water . Genomic DNA from approximately 420 animals ( 260 derived from a single founding animal; the remainder derived from a subpopulation of approximately 10 wild-caught founders ) was extracted using the DNeasy Blood & Tissue kit ( Qiagen ) following the standard protocol . DNA was extracted in batches and pooled to generate sufficient material . Paired-end data for R . magnacalcarata have been described previously [41] . Both R . macrura and R . magnacalcarata PE libraries are derived from multiple individual samples . For mate-pair library construction for both R . macrura and R . magnacalcarata , DNA was extracted from a single individual and subjected to WGA using the Repli-G Single Cell kit ( Qiagen ) , following the manufacturer’s protocol . DNA concentration and quality were ascertained using a Qubit ( Invitrogen ) and a NanoDrop spectrophotometer ( Thermo Scientific ) . Desiccation tolerance of R . macrura and R . magnacalcarata were tested using protocols as previously described [17 , 41] ( see S3 Text for further details ) . For A . ricciae , a short-insert library with an insert size of 250 bp was prepared using Illumina Nextera reagents and sequenced ( 100 bases paired-end ) on an Illumina HiSeq 2000 at the Eastern Sequence and Informatics Hub ( Cambridge , UK ) . Two long-insert ( mate-pair ) libraries both with inserts of 3 kb were also sequenced ( 51 bases paired-end ) at GATC Biotech ( London , UK ) . In addition , a PacBio ( Pacific Biosciences ) long-read library with an insert of 10 kb was sequenced using 3 SMRT Cells on a PacBio RS II ( The Genome Analysis Centre , Norwich , UK ) . An RNASeq library with an insert size of 250 bp was sequenced ( 150 bases paired-end ) on an Illumina NextSeq500 at the Department of Biochemistry , University of Cambridge ( Cambridge , UK ) . A short-insert library ( 500-bp insert ) for R . macrura was prepared using Illumina TruSeq reagents at the Centre for Genomic Research ( CGR ) at the University of Liverpool ( Liverpool , UK ) . Mate-pair libraries with 2-kb inserts were also prepared at CGR using Nextera reagents , and all libraries were sequenced ( 150 bases paired-end ) over 3 lanes of an Illumina HiSeq4000 at CGR . Short-insert data for R . magnacalcarata have been described previously [41] . All raw data have been submitted to the Sequence Read Archive ( SRA ) , an International Nucleotide Sequence Database Collaboration ( INSDC ) , under the accession IDs ERR2135445–55 ( S1 Table ) . For A . ricciae , R . macrura , and R . magnacalcarata data , adapter sequences and low-quality bases were removed from Illumina data using Skewer v0 . 2 . 2 [107] , and data quality was manually assessed using FastQC v0 . 11 . 5 [108] . Genome coverage was estimated by generating kmer distributions using BBMap ‘kmercountexact’ v36 . 02 [109] , and library insert sizes—along with initial genome size estimates—were calculated using SGA ‘preqc’ [110] . Error correction of reads was performed using BBMap ‘tadpole’ ( k = 31 ) , discarding any pairs of reads containing unique kmers . Contaminant reads derived from non–target organisms were filtered using BlobTools v0 . 9 . 19 [111] . Briefly , trimmed and error-corrected paired-end data were digitally normalised to approximately 100x using BBMap ‘bbnorm’ [109] , and a preliminary draft assembly was generated using Velvet v1 . 2 . 10 [112] , setting a kmer length of 75 . Taxonomic annotations for all contigs were determined by comparing contigs against the NCBI nucleotide database ( nt ) and a custom database containing recently published whole-genome sequences of metazoans within the Spiralia ( Lophotrochzoa ) group ( S4 Table ) using BLAST ‘megablast’ ( E-value ≤1 × 10−25 ) [113] , and the UniRef90 database using Diamond ‘blastx’ [114] . Finally , read coverage for each contig was estimated by mapping non-normalised reads to each draft assembly using BWA ‘mem’ v0 . 7 . 12 [115] . Taxon-annotated GC coverage plots ( ‘blobplots’ ) [111 , 116] were generated using BlobTools ( default parameters ) and inspected manually . Putative contaminant sequences were identified as contigs showing atypical GC content , read coverage , and/or taxonomic classification . Given the a priori expectation that a substantial number of bdelloid genes may derive from nonmetazoan sources , we did not exclude any contigs based on taxonomy alone . Paired reads were excluded from further analysis only if both mapped to an identified contaminant contig or if one of the pair mapped to a contaminant while the other was unmapped . Additional rounds of filtering were performed if previously unassembled contaminant sequences became evident upon reassembly . Filtered reads were assembled into contigs using the Platanus assembler v1 . 2 . 4 [42] with default parameters . Mate-pair libraries were filtered to remove contaminating FR-orientated reads ( i . e . , reads originating from short fragments ) by excluding reads that mapped within ≤500 bases from the terminus of a contig . Contigs were scaffolded using SSPACE v3 . 0 [117] , and undetermined bases were filled using GapFiller v1 . 10 [118] . The A . ricciae assembly was further scaffolded with the PacBio library using SSPACE-LongRead v1 . 1 [119] . RNASeq reads for A . ricciae were assembled de novo using Trinity v2 . 2 . 0 [120] ( default parameters ) and used for additional scaffolding with L_RNA_Scaffolder [121] and SCUBAT v2 [122] . An available transcriptome for R . magnacalcarata [41] was similarly utilised . A final round of assembly ‘polishing’ was performed using Redundans v0 . 12b [43] , and scaffolds less than 200 bases in length were discarded . Assembly completeness was evaluated using the CEGMA v2 . 5 [44] and BUSCO v3 . 0 . 0 [45] gene sets , choosing the Eukaryota ( n = 303 ) and Metazoa ( n = 978 ) databases in the latter case and increasing the search limit to 8 . Alternative assemblies were also generated using Velvet [112] , SPAdes [123] , and dipSPAdes [124] for comparison . The reference assembly pipeline above was designed to maximise assembly contiguity but may lead to assembly collapse , the extent of which is undetermined a priori . Therefore , maximum haplotype assemblies were also generated for each species for comparison , defined as assemblies with minimal reduction due to assembly collapse . Maximum haplotype assemblies were generated using either dipSPAdes ( default settings ) or Platanus with the ‘bubble crush’ reduction parameter set to 0 . Details of assembly parameters trialled are given in S1 Data . Collapsed and maximum haplotype ( re ) assemblies for A . vaga were also generated following the same procedures , using Illumina short-insert libraries ( accession IDs SRR801084 and ERR321927 ) for contig building as well as mate-pair ( accession ID ERR321928 ) and 454 data for scaffolding ( see [28] for details ) . Repetitive regions were masked prior to gene prediction . Repeats were modelled ab initio using RepeatModeler v1 . 0 . 5 [125] . Repeats arising from duplicated genes or recent gene family expansions ( e . g . , alpha-tubulin in R . magnacalcarata [126] ) were removed from the custom repeat library by comparing each repeat library to the SwissProt database ( BLASTX , E-value ≤1 × 10−5 ) and retaining only those sequences with descriptions for known repeat elements . The filtered RepeatModeler library was merged with known Rotifera repeats from Repbase v22 . 02 [127] ( accessed using the command ‘queryRepeatDatabase . pl -species ‘rotifera’ ) and compared to each assembly using RepeatMasker v4 . 0 . 7 [128] . Low-complexity regions and simple repeats were additionally soft-masked . Gene prediction was then performed using BRAKER v1 . 9 [47] where RNASeq data was available ( A . ricciae and R . magnacalcarata ) . Briefly , RNASeq reads were aligned to the masked assembly using STAR , specifying the ‘twoPassMode Basic’ parameter to improve splice junction annotation . The resultant alignment BAM file was then input to the BRAKER pipeline with default settings . For R . macrura , an initial set of gene models was constructed using MAKER v3 . 00 [48] , using evidence from SNAP [129] and GeneMark-ES v4 . 3 [130] . MAKER-derived gene models were then passed to Augustus v3 . 2 . 1 [49] for final refinement . Transfer and ribosomal RNA genes were predicted using tRNAscan-SE v1 . 3 . 1 [131] and RNAmmer [132] , respectively ( S9 Data ) . The A . vaga 2013 assembly ( GCA_000513175 . 1 ) was also reannotated for consistency with these results , using both approaches outlined above ( in conjunction with RNASeq library accession ERR260376 ) . To test if CDSs had been inadvertently missed during gene prediction , we compared proteins to the source nucleotide sequences from which they had been predicted using TBLASTN ( E-value ≤1 × 10−20 ) . Matches to existing gene models were discounted by removing alignments that showed any overlap with gene regions ( BEDtools ‘intersect’ [133] with the ‘-v’ option ) , leaving only hits to regions of the genome that had not already been annotated as a gene . Syntenic regions within and between genomes were identified using MCScanX [50] , calling collinear ‘blocks’ regions with at least 5 homologous genes and fewer than 10 ‘gaps’ ( i . e . , missing genes ) . Rates of synonymous ( KS ) and nonsynonymous ( KA ) substitution between pairs of collinear genes were estimated by aligning proteins with Clustal Omega [134] and back-translating to nucleotides before calculating KA and KS values using BioPerl [135] . The collinearity of each block was calculated by dividing the number of collinear genes in a block by the total number of genes in the same region [28] . We also counted the number of collinearity breakpoints between adjacent homologous blocks across each genome , defining a breakpoint as an occurrence in which homologous blocks cannot be aligned without rearrangement . Collinearity plots were generated using the Circos software [136] in conjunction with the circosviz . pl program from the mmgenome toolkit [137] . Collinearity analysis scripts are available at https://github . com/reubwn/collinearity . Orthologous relationships among proteins from the same set of protostomes as above were inferred using OrthoFinder v1 . 1 . 4 [138] with default settings . All genomic , GFF , and protein sequence datasets were downloaded from NCBI GenBank no later than May 2017 . For SNP finding , data were mapped using Bowtie2 v2 . 2 . 6 [139] with the ‘--very-sensitive’ preset to minimise mismapped reads , and SNPs and indels were called using Platypus v0 . 8 . 1 [140] , setting a minimum mapping quality of 30 , a minimum base quality of 20 , filter duplicates to 1 , and a minimum read depth to approximately 25% of the average coverage of each individual library . VCF manipulation and SNP statistics were calculated using VCFlib v1 . 0 . 0-rc1 [141] . For A . vaga , SNPs were called based on the Illumina dataset ERR321927 mapped to the published genome sequence [28] . For R . macrura and R . magnacalcarata , SNPs were called based on WGA mate-pair libraries mapped as single-end because paired-end data for these samples were composed of multiple nonclonal lineages . We assessed the extent of horizontal transfer into bdelloid genomes using a combination of sequence comparison and phylogenetics-based approaches and applied the same tests to a set of 13 publicly available proteomes from species across the Protostomia ( S4 Table ) . Protein sequences were first compared to the UniRef90 database [142] ( downloaded November 29 , 2016 ) using Diamond ‘blastp’ [114] ( E-value ≤1 × 10−5; maximum target sequences = 500 ) . To avoid potential bias from bdelloid sequences already submitted to GenBank , all hits to the phylum Rotifera ( NCBI taxonomy ID 10190 ) were omitted from further analysis . For each query , 2 HGT metrics were then calculated: ( 1 ) HGT Index ( hU [27] ) , defined as BOUT − BIN , where BIN is the best ( highest ) Diamond bitscore from comparisons to ‘ingroup’ taxa and BOUT is the corresponding score for hits to ‘outgroup’ taxa; and ( 2 ) Consensus Hit Support ( CHS ) , defined as the proportion of all hits that support a given query’s ingroup/outgroup classification , itself inferred from the highest sum of bitscores to ingroup or outgroup across all hits [94] . The CHS score therefore takes into account the taxonomic distribution of all hits for each query and militates against misclassifications based on hU scores alone . We defined the ingroup as ‘Metazoa’ and the outgroup as ‘non-Metazoa’ and marked all proteins with an hU ≥30 and CHSOUT ≥90% as putative HGTC . We then looked at the distribution of all HGTC across the genome and discarded any candidate found on a scaffold encoding ≥95% of genes of putative foreign origin ( i . e . , ‘HGT-heavy’ scaffolds that may be derived from contaminant sequences that were not removed during assembly ) . For each HGTC , physical linkage ( i . e . , presence on the same scaffold ) to a gene with good evidence for metazoan origin ( hU ≤0 , CHSIN ≥90% ) and the number of predicted introns were also recorded . Finally , phylogenetic support for HGT was then assessed: for each HGTC , the sequences of 15 metazoan and 15 nonmetazoan UniRef90 hits ( when present ) were extracted and aligned using MAFFT v7 . 309 [143] with default parameters , and a maximum likelihood phylogeny was constructed using IQ-TREE v1 . 5 . 3 [144] , specifying automatic model selection and 1 , 000 ultrafast bootstrap replicates . The functionality of GNU Parallel [145] was used to compute multiple trees simultaneously , and clusters with fewer than 4 taxa were not analysed . Branching patterns of resultant trees were then assessed using a custom script written in R v3 . 3 . 1 [146] , utilising functions from the ‘ape’ v4 . 1 package [147] . HGT analysis scripts are available at https://github . com/reubwn/hgt . The abundance of known TEs was assessed for the same set of protostomes using RepeatMasker , except using a Repbase ( v22 . 02 ) repeat library specific to the Metazoa ( i . e . , ‘queryRepeatDatabase . pl -species ‘metazoa” ) . Custom species-specific repeat libraries ( e . g . , using RepeatModeler ) were not generated for this analysis; only known repeats from Repbase were compared . The total span of LINEs/SINEs , LTR elements , DNA elements , and simple repeats relative to the assembly span for each species was then computed from the RepeatMasker results . We also calculated a genome density metric , defined as the number of protein-coding genes per Mb of haploid genome , i . e . , accounting for variation in ploidy among species . The presence of meiosis- and other sex-related genes was assessed following the approach of Tekle et al . [101] . A total of 41 orthologous groups were downloaded from the OrthoMCL database ( v5 ) ( http://orthomcl . org/orthomcl/; accessed September 2017 ) ( S8 Data ) . Searches were conducted using both TBLASTN ( E-value ≤1 × 10−5 ) against the reference assemblies or HMMER3 ( http://hmmer . org/ ) against the predicted protein sets , after alignment with Clustal Omega [134] . Presence was recorded if any query within each orthologous group showed a TBLASTN alignment with ≥50% identity over ≥50% query length and/or if HMMER reported an alignment above the default significance threshold . Multiple hits to the same location ( caused by paralogy or hits to similar domains ) were recorded if top hits overlapped among queries . The genomes and proteomes of D . melanogaster and C . elegans were also searched for comparison . | Bdelloid rotifers are microscopic animals that live in freshwater habitats throughout the world . Two life history characteristics distinguish these common invertebrates as extraordinary . First , they have existed for millions of years apparently without sex: males have not been reported , and females produce genetically identical daughters via parthenogenesis . Second , most bdelloid species are highly resilient to desiccation and can survive without water for extended periods of time ( a process known as anhydrobiosis ) . These 2 attributes have been predicted to leave signatures in bdelloid genomes . Here , we present new draft genomes for 3 bdelloid species and employ comparative genomics to explore the potential impacts of bdelloids’ unusual lifestyle on genome structure and content . We find that many proposed genomic consequences of asexuality and desiccation tolerance do not hold true for all species . The genomes of bdelloids may be more similar to those of other animals than previously thought , though a remarkable exception is the high proportion of genes acquired horizontally from nonmetazoan taxa . Our findings necessitate a reevaluation of the effects of asexuality and desiccation on genome evolution in bdelloid rotifers . | [
"Abstract",
"Introduction",
"Results",
"and",
"discussion",
"Concluding",
"remarks",
"Materials",
"and",
"methods"
] | [
"horizontal",
"gene",
"transfer",
"genome",
"evolution",
"invertebrate",
"genomics",
"sequence",
"assembly",
"tools",
"gene",
"transfer",
"genome",
"analysis",
"genomic",
"libraries",
"genomics",
"molecular",
"evolution",
"comparative",
"genomics",
"animal",
"genomics",
"eukaryota",
"rotifers",
"genetics",
"biology",
"and",
"life",
"sciences",
"computational",
"biology",
"evolutionary",
"biology",
"evolutionary",
"processes",
"organisms"
] | 2018 | Comparative genomics of bdelloid rotifers: Insights from desiccating and nondesiccating species |
In the study of circadian rhythms , it has been a puzzle how a limited number of circadian clock genes can control diverse aspects of physiology . Here we investigate circadian gene expression genome-wide using larval zebrafish as a model system . We made use of a spatial gene expression atlas to investigate the expression of circadian genes in various tissues and cell types . Comparison of genome-wide circadian gene expression data between zebrafish and mouse revealed a nearly anti-phase relationship and allowed us to detect novel evolutionarily conserved circadian genes in vertebrates . We identified three groups of zebrafish genes with distinct responses to light entrainment: fast light-induced genes , slow light-induced genes , and dark-induced genes . Our computational analysis of the circadian gene regulatory network revealed several transcription factors ( TFs ) involved in diverse aspects of circadian physiology through transcriptional cascade . Of these , microphthalmia-associated transcription factor a ( mitfa ) , a dark-induced TF , mediates a circadian rhythm of melanin synthesis , which may be involved in zebrafish's adaptation to daily light cycling . Our study describes a systematic method to discover previously unidentified TFs involved in circadian physiology in complex organisms .
The circadian rhythm evolved as an adaptation to the Earth's daily light cycle . In complex animals , each tissue or cell type contains a functional clock that is controlled by the central pacemaker . In mammals , the suprachiasmatic nucleus ( SCN ) in the hypothalamus acts as the central circadian pacemaker that coordinates most aspects of behavior and physiology [1] . In lower vertebrates , circadian pacemakers are also present in the eyes and the pineal gland [2] . It is not surprising that circadian rhythm controls physiological processes universal to all tissues such as metabolism [3] and cell cycle [4] . However , tissue-specific functions are also controlled by the circadian clock . Meta-analysis of all existing mammalian circadian gene expression studies indicates that there are thousands of genes showing circadian expression in a given tissue , although only a small number of circadian genes are expressed in more than one tissue [5] . This raises the question as to how genome-wide tissue-specific circadian gene expression is regulated ? A simple answer is that all circadian gene expression is entirely regulated by core circadian genes . However , there are only a few transcription factors ( TFs ) within the core circadian circuit , including Arntl/Clock ( Bmal1/Clock ) and Bhlhe40/41 ( Dec1/2 ) ( binding the E-BOX ) , Rora/b/c and Nr1d1/2 ( Rev-erb α/β ) ( binding RRE ) , and Dbp/Hlf/Tef and Nfil3 ( binding the D-BOX ) , each of which either activates or represses target genes in mammals . Recent ChIP-seq experiments have revealed 2 , 049 genome-wide binding sites for Bmal1 in mouse liver targeting only 16% of the circadian oscillating genes that are expressed in liver and are represented in the circadian gene database [5] , [6] . From this we can conclude that , circadian oscillations in gene expression in different tissues are unlikely to all be under the direct regulation of known core circadian TFs , but rather they are likely to be regulated by other TFs that are themselves regulated by core circadian genes [7] , [8] . In particular , tissue-specific circadian functions might be regulated by tissue-specific TFs relaying signals from core circadian genes . Typically , circadian gene expression studies are conducted in a few selected tissues at a time . In mouse , the most studied circadian model in vertebrates , genome-wide circadian gene expression has been studied in only 14 tissues [5] . An additional difficulty is that many tissues or cell types , with potential importance to circadian rhythm are difficult to access without specialized techniques . For example , in Drosophila , Kula-Eversole et al . have studied circadian gene expression in PDF-expressing ventral lateral neurons: s-LNvs and l-LNvs and observed marked differences with those in fly head [9] . In this study , we attempt to circumvent these difficulties by a genome wide analysis of circadian gene expression in whole larval zebrafish ( Danio rerio ) . The zebrafish larva is an established model for the study of animal development [10] . Decades of gene expression studies in zebrafish , most of which used in situ hybridization ( ISH ) provide a rich resource of spatial information on gene expression ( http://zfin . org ) . Therefore , circadian oscillating genes identified in this study can be mapped to specific tissues and cell types using a spatial expression atlas . We have identified a number of tissue-specific TFs that were not previously known to be involved in circadian rhythm . Through the analysis of gene regulatory networks , we identified a novel circadian TF , microphthalmia-associated transcription factor a ( mitfa ) , controlling circadian melanin synthesis in pigment cells . mitfa shows strong circadian rhythm in expression and is dark-induced . This observation lead us to hypothesize that melanin synthesis itself is similarly circadian controlled . We confirmed this experimentally and speculate that a dark-adaptation pathway is involved . Circadian- regulated melanin synthesis is likely to be crucial to species , including fish and reptiles , that change skin color to evade predators in response to the daily light-cycle . Our study highlights the possibility that different organisms may employ the circadian rhythm differently to control their specific behavior allowing adaptation to their ecological niche .
We monitored the circadian behavior of larval zebrafish under an infrared behavioral monitoring platform ( Text S1 , Figure S1A ) . Wild-type ( WT ) larval zebrafish were raised in 14 h∶10 h light/dark ( LD ) cycle from birth . Starting at 5 days post-fertilization ( dpf ) , they were monitored under either LD or constant dark ( DD ) conditions . In both conditions , the fish demonstrated robust circadian changes in their locomotor activities ( Figure S1B ) implying that circadian activities have been fully established at this stage of zebrafish development . The amplitude of oscillation was significantly lower in DD larvae compared to that in LD larvae indicating strong dependence of circadian activities on light entrainment . We subsequently collected larval zebrafish starting at 5 dpf every 4 hours for a period of 48 hours under both LD and DD conditions . The whole-genome transcriptome profiles of the animals were assayed on Agilent zebrafish microarrays . Circadian oscillating genes were identified using Fisher's g test , and their circadian phases were determined by fitting to cosine functions with shifting phases ( Materials and Methods ) . We identified 2 , 856 circadian oscillating genes in both LD and DD conditions with an overall False Discovery Rate ( FDR ) ≤5% ( Table S1 ) . These account for over 17% of expressed genes in larval zebrafish ( Figure 1A ) . We thus refer to these zebrafish circadian oscillating genes in both LD and DD as “zebrafish circadian oscillating genes” ( ZCOGs ) . The mean circadian phases under LD and DD are used to represent the circadian phases of ZCOGs . This dataset displays a prominent bimodal distribution with peaks at CT2 ( Circadian Time 2; CT0 is lights-on and CT14 is lights-off ) and CT16 corresponding to 2 hours after lights-on and lights-off respectively ( Figure 1B ) . We annotated the biological functions of ZCOGs using the Gene Ontology ( GO ) and KEGG databases . ZCOGs are enriched in GO terms related to light and abiotic stimuli and transporter functions ( Table S2 ) . As expected , circadian rhythm is the most significantly enriched KEGG pathway among ZCOGs . We then used a sliding window approach to identify the specific time window when the biological functions of ZCOGs are enriched ( Figure 1D ) . Components of the lysosome are enriched around CT16 , proteasome components are enriched around CT20 , and those involved in ribosome biogenesis are enriched around CT0 . We compiled a spatial gene expression atlas from annotations in the ZFIN database , which contains expression information for 96 , 189 records encompassing 6 , 429 genes in 1 , 173 tissues or cell types in larval zebrafish . A total of 1 , 310 ZCOGs were mapped to the tissues or cell types of expression and we computed the enrichment of circadian oscillating genes in these tissues . We found that ZCOGs are most enriched in retina photoreceptive layer , intestine bulb , and retina . Zebrafish homologs of almost all known mammalian core circadian genes are represented among ZCOGs ( Table 1 ) , indicating that the basic circadian circuit is highly conserved between zebrafish and mammals . To search for novel circadian genes conserved between mammals and zebrafish , we obtained an extended list of mouse circadian genes oscillating in at least six tissues from our previously constructed circadian gene databases [5] . The circadian phases of these genes have very low variability across different mouse tissues . We identified 74 ZCOGs that are homologs of 48 mouse circadian genes ( Table S3 ) . Many circadian genes in zebrafish have undergone gene duplications including arntl1 ( bmal1 ) , clock , and per1 . Duplicated circadian genes tend to have similar circadian phases . In addition to commonly known core circadian clock genes , heat/cold shock proteins , purine nucleoside phosphorylases , and ubiquitins are also among the conserved circadian oscillating genes between mouse and zebrafish . Interestingly , we observed a nearly anti-phase ( 9 hour shift ) relationship between zebrafish and mouse among conserved circadian genes ( Figure 1C ) . Such phase shift could arise from different entrainment conditions between zebrafish and mouse ( 14∶10 LD for zebrafish and 12∶12 LD for mouse ) and other species differences such as activity patterns . In larval zebrafish , the amplitude of circadian activity is significantly reduced under DD compared to that under LD conditions . Using the same false discovery rate ( FDR ) parameters , the number of circadian oscillating genes under DD is also significantly lower than that under LD . To examine the effects of light entrainment on circadian gene expression , we first compared the phases and amplitudes of circadian oscillations of ZCOGs between DD and LD conditions . Among 2 , 856 ZCOGs , the circadian phases are highly consistent between DD and LD ( circular correlation coefficient r = 0 . 93 ) with only 169 ( 6% ) ZCOGs showing phase shifts of more than four hours . In contrast , 233 ZCOGs oscillate at reduced amplitudes with only 21 ZCOGs oscillating at increased amplitudes under DD compared to LD . Thus light entrainment tends to enhance the amplitudes of circadian oscillation of gene expression . Next , we searched for genes having reduced amplitudes as a consequence of either decreased peak levels or elevated trough levels in DD compared to LD . To include the genes that might have lost rhythmicity under DD , we examined 3 , 677 LD oscillating genes ( FDR≤10% ) regardless of their rhythmicity under DD . Among them , we identified 464 genes showing decreased peaks and 113 genes showing elevated troughs in DD compared to LD ( Materials and Methods ) . We defined 366 peak decreased genes with LD circadian phases lying in the light period as “light-induced” because their expression increases in the light period in LD but remains at low level in DD ( Figure 2A; Table S4 ) . Similarly , 50 trough-increased genes with LD circadian phases lying in the dark period were defined as “dark-induced genes” . Furthermore , we observed that the LD circadian phases of light-induced genes were predominantly distributed around CT5 . The expression of these genes increases rapidly after lights-on and reaches a peak after 5 hours in LD . Thus we defined 197 light-induced genes with LD circadian phases between CT2 and CT8 as “fast light-induced genes” . This group includes known light-sensitive genes ( e . g . per2 , cry5 , cry-dash , and tefa ) . It also contains genes showing significantly enriched expression in epiphysis and the retinal photoreceptive layer ( e . g . arrestin 3 ( arr3a/b ) and interphotoreceptor retinoid-binding protein ( irbp ) ) . The remaining light-induced genes show a slow increase in expression during the light period , reaching a peak before lights-off in LD . Accordingly , they were defined as “slow light-induced genes” . This group includes key circadian genes ( e . g . cry2a/b , arntl1b , and nfil3-5 ) . The expression of dark-induced genes is repressed during the light period but increases in the dark period reaching peak level after lights-off . Among the dark-induced genes are melatonin synthesizing enzyme ( aanat1 ) , dopamine D4 receptor ( drd4a ) and two genes related to pigmentation ( mitfa and slc24a5 ) . In short , there are complex , dynamic responses to light entrainment during circadian gene expression . In another set of experiments , we collected larval zebrafish fertilized and raised under either DD or LL ( constant light ) conditions and exposed them to light or dark , respectively for 20 hours starting at 5 dpf . The expression of one representative gene from each of the previously described groups was measured by quantitative real-time PCR ( Text S1 ) : per2 from the fast light-induced genes , cry2a from the slow light-induced genes , aanat1 from the dark-induced genes and the clock gene ( Figure 2B , primers used for real-time PCR are in Table S5 ) . The first few hours after light/dark exposure were sampled more frequently . Without light/dark exposure , the circadian clock was desynchronized in animals kept in DD or LL conditions [11] , [12] . The expression levels of per2 and cry2a are significantly higher in LL than DD and aanat1 was expressed at a significantly lower level in LL than DD whereas the expression level of the clock gene showed no significant difference between LL and DD . After light exposure , the expression of per2 increases rapidly and reaches a peak after 3 hours whereas , after dark exposure , it decreases rapidly within 3 hours . In contrast , cry2a expression increases slowly after light exposure and decreases slowly after dark exposure only reaching its peak or trough about 12 hours after lights-on or lights-off , respectively . The expression of aanat1 increases quickly upon dark exposure and drops just as rapidly after light exposure . The expression of clock shows a slow increase of expression reaching its peak around 12 hours after light exposure and an even later increase around 20 hours after dark exposure . These results indicate that the endogenous circadian clock has been resynchronized after the exposures and may regulate the induction and repression of slow light-induced genes and dark-induced genes . To understand what drives the circadian oscillation of ZCOGs , we developed a comparative genomics pipeline to predict TF binding sites for zebrafish genes . Our pipeline uses orthologous promoter sequences from teleost fish genomes , including zebrafish , fugu , medaka , stickleback , and Tetraodon ( Materials and Methods ) . Our ZCOGs consist of 140 TFs , including novel circadian TFs in addition to the known core circadian TFs ( Table S6 ) . We focused on the 54 circadian TFs with known DNA-binding motifs in the TRANSFAC database [13] for promoter analysis . Because the circadian phases of ZCOGs show a bimodal distribution ( Figure 1B ) , we divided ZCOGs into two groups , the first with peak times within CT22-CT10 , and the second with peak times within CT10-CT22 . In the first group , we identified E-BOX as the most enriched promoter motif ( p value = 2 . 5×10−10 , odds ratio = 2 . 3 ) , while RRE is significantly enriched in promoters from the second group of genes ( p value = 1 . 3×10−5 , odds ratio = 2 . 0 ) . This is consistent with observations in mouse except that the enriched phases are also shifted by about 10 hours following the same phase-shift of homologous TFs in zebrafish and mouse [5] . We then used a sliding window approach to identify the specific time window when the circadian phases of circadian TF targets are enriched ( Table S7 ) . In addition to E-BOX and RRE , we also observed that targets of ppargc1b , yy1a/b , atf/creb , hnf1a , foxo3b , and myog were enriched at a specific time window ( p<0 . 001 , Figure 3 ) . There are varying delays between phases of TFs and enriched phases of their targets . Phases of tef , foxo3b , ppargc1b , and hnf1a are close to their targets , while arntl/clock , nfil3 , yy1a , and myog are nearly anti-phase to their targets . To assist sequence-based TF-target prediction , we generated co-expression groups of zebrafish genes using zebrafish microarray data available in the Gene Expression Omnibus ( GEO ) database ( Materials and Methods ) and mapped ZCOGs to these groups . By making use of the fact that co-expression often indicates co-regulation by TFs , we were able to identify the TF motifs enriched in the promoters of the co-expression groups and their associated circadian TFs ( Table S8 ) . A group consisting of mostly mitochondrial genes is enriched with ZCOGs and their circadian phases are around CT0 . The genes in this group are predicted to be regulated by yy1a . Consistent with our GO analysis , proteasome components form a co-expression group having circadian phases around CT18 , although no known TF motifs are enriched in the group . Other co-expression groups regulated by circadian TFs include heat shock proteins regulated by hsf2 , liver-specific genes regulated by hnf1a , and genes in visual processing regulated by crx ( core-rod homeobox ) . Hnf1a is known to be a liver-specific TF [14] and expression of crx is restricted to retinal photoreceptor cells and the epiphysis [15] , [16] . Using the spatial gene expression atlas in zebrafish , we searched for tissue-specific TFs among ZCOGs by requiring that the predicted targets of the TFs are predominantly expressed in the same tissue as the TFs themselves . We systematically identified 11 tissue-specific circadian TFs . They include several well-known tissue-specific TFs: mitfa in pigment cells [17] , cdx1b in the intestine [18] , ppargc1b in the liver , mef2a in heart and muscle tissue [19] , pax6a in nerve tissues [20] , smad1 [21] , smad3a [22] , and myog [23] in muscle tissue . A less well-known tissue-specific circadian TF , maf , is expressed in the lens vesicle [24] . There is evidence that maf plays an important role in lens development and in lens fiber cells in mice [25] . Interestingly , two key circadian TFs appear to be tissue-specific: nr1d1 and rorab are expressed in epiphysis and retina , while their predicted targets show enriched expression in the same tissues . Notably , nr1d1 plays an important role in photoreceptor development and function [26] . From this data we constructed a gene regulatory network of circadian TFs to depict how they regulate each other and how circadian oscillating genes in specific tissues are regulated ( Figure 4 ) . In this network , circadian-oscillating TFs were grouped by TF-motif and motif-target relationships , where TF-motif relationships were obtained from the TRANSFAC database , while motif-target relationships were predicted through promoter analysis in our comparative genomics pipeline ( see Materials and Methods ) . This approach successfully recapitulated the known core transcriptional feedback loop in circadian clock regulation: E-BOX binding TFs arntl/clock regulate nr1d1 and rorab , which in turn regulate arntl through RRE . Most circadian TFs are directly regulated by the circadian clock via known circadian cis-elements . For example , mef2a , hsf2 , tef , bhlhe40/41 ( dec1/2 ) , tfcp2l1 , foxo3b , and ahr1a via E-BOX , myog , nfil3 , atf4b1 via RRE and yy1a and smad3a via D-BOX . The remaining TFs are indirect targets of core circadian TFs via further transcriptional cascades . For example , maf and mitfa are regulated via cAMP responsive element ( CRE ) , hnf1a and creb3l3 via the cis-element of the PPAR family TFs and pax6a and nr3c1 via the cis-element of foxo3b . Our network analysis shows that clock is at the center of the regulatory network and all other circadian TFs are under its control either directly or indirectly . In order to validate this experimentally , we generated clock morpholino ( MO ) knock-down larvae . We measured circadian mRNA levels of bhlhe40/41 and per3 together with 11 circadian tissue-specific TFs by real-time PCR in 5 dpf clock morphants compared to WT or control morphants ( Figure 5 ) . In WT and controls , the circadian peak times of these genes show high consistency between real-time PCR and arrays . In clock morphants , bhlhe40/41 , per3 , smad3a , and hnf1a continue to oscillate but show significant reductions in oscillation amplitudes . The remaining genes show loss of rhythmicity except for crx in clock morphants . Furthermore , smad1 and myog show significant up-regulation ( p<0 . 001 ) , whereas ppargc1b , maf , mef2a show significant down-regulation in baseline levels in clock morphants compared to the WT and controls ( Primers used for real-time PCR are in Table S5 ) . In order to show that TF-mediated circadian control can have novel functional consequences , we focused on mitfa , one of our identified circadian TFs and the most enriched TF in the pigment cells . It is also a dark-induced gene . The mammalian homolog of mitfa is Mitf , which is a key TF controlling melanogenesis in mammals [27] . In zebrafish , mitfa is specifically expressed in pigment cells and retinal pigment epithelium ( RPE ) [28] . Notably , mitfb , a gene duplicate of mitfa , does not show circadian rhythm in our microarray result . To validate the function of mitfa in zebrafish , we generated mitfa MO knock-down zebrafish larvae . The mitfa morphants lost melanin production in melanocytes of the fish trunk and had smaller eyes compared to WT and control morphants ( Figure S2 ) . It has been reported that Mitf directly regulates Tyr , Tyrp1 , and Dct , genes encoding three key enzymes involved in melanin synthesis , in addition to other genes involved in melanogenesis in mouse and human [29] , [30] . In our study , tyrp1b ( the zebrafish homolog of Tyrp1 ) , slc24a5 ( another gene known to be involved in melanogenesis ) and dct all showed circadian peaks around CT20 in LD , reduced oscillation amplitudes and elevated troughs in DD , similar to mitfa ( Figure S3 ) . We then measured the mRNA levels of these genes in mitfa morphants by real-time PCR . All showed reduced expression in mitfa morphants compared to WT and control morphants , suggesting that they are also mitfa targets in zebrafish . We reasoned that the circadian oscillation of mitfa expression implies that melanin synthesis also oscillates on a daily basis . To experimentally demonstrate the circadian oscillation of melanin synthesis , we first measured the areas of melanocytes in the head regions of larval zebrafish under LD beginning at 4 dpf ( Figure 6A , Materials and Methods ) and found a significant circadian rhythm in pigmentation with a peak around CT0 ( p<0 . 001 , Fisher's g test , Figure 6B ) . Then we quantified total melanin concentration using a melanin assay in whole larval zebrafish . The melanin concentration showed a strong circadian rhythm with a circadian phase also around CT0 in LD ( p<0 . 002 , Fisher's g test after detrend , Figure 6C ) starting at 4 dpf , while the absolute amount of melanin increased during development . To determine the effect of clock on pigment synthesis , we examined the melanin concentration rhythm in clock morphants . The melanin concentration showed a significantly reduced circadian rhythm in LD compared to control morphants ( Figure 6D ) . Melanin oscillations were not significant in DD in either clock morphants or controls ( Figure 6E ) . Similarly , the amplitude of mitfa expression was significantly reduced by over 30% in clock morphants in LD compared to WT and controls , and oscillations are nearly absent in DD . From this we conclude that the circadian control of melanogenesis is mediated by mitfa in zebrafish larvae . The melanogenesis signaling pathway has been well-characterized in mammals . Alpha-melanocyte stimulating hormone ( α-MSH ) activates melanocortin receptor in pigment cells leading to the up-regulation of Mitf through cAMP signaling pathway [31] . Consistent with this , we identified a conserved CRE in the promoter of mitfa in zebrafish ( Figure S4 ) . In mammals corticotropin releasing hormone ( CRH ) secreted from the paraventricular nucleus ( PVN ) of the hypothalamus affects the secretion of α-MSH in the pituitary gland thus forming a hypothalamus-pituitary-melanocyte ( HPM ) axis . In our study , the mRNA level of crh shows circadian oscillation at CT9 in LD and reduced amplitude in DD . To examine if the HPM axis controls the expression of mitfa in zebrafish , we treated zebrafish larvae with the Crh receptor 1 antagonist antalarmin . We observed significant reduction of mitfa expression upon antalarmin treatment at CT0 ( Text S1 , Figure S5 ) . Thus , it appears as though the circadian oscillation of mitfa may be regulated by the crh signaling cascade through the HPM axis . The mitfa gene is involved in the early development of the neural crest from which pigment cells are derived [32] . Multiple neural cell types are also derived from the neural crest cells and mitfa's control of melanogenesis is likely to be only one aspect of its function . To investigate the broader regulatory functions of mitfa in zebrafish early development , we collected mitfa morphants together with WT and control morphants at 48 hpf ( hours post-fertilization ) and 50 hpf for microarray experiments ( Text S1 ) . We identified 555 down-regulated genes and 691 up-regulated genes in mitfa morphants , compared to WT and control larvae ( Table S9 ) . In addition to tyr , tyrp1b , dct , and slc24a5 , we also identified tcf7l2 , lef1 , camk2d2 , and slc45a2 which are also involved in melanogenesis , as being down-regulated in mitfa morphants . Mapping of the down-regulated genes onto the spatial gene expression atlas revealed that the retina is the most common tissue for their expression . This is consistent with the small eye phenotype of mitfa morphants . Genes involved in brain development and the wnt signaling pathway are enriched in the down-regulated genes . Of the mitfa-affected genes represented by ZCOGs , we found 87 ZCOGs down-regulated and 213 ZCOGs up-regulated in the mitfa knock-down . Among these , two TFs , six3a and vsx1 , are involved in retina development and both show circadian peaks at CT16 . We conclude that , in addition to mediating circadian melanogenesis , mitfa is also likely to mediate circadian control of early development in other tissues such as the retina .
Until now , studies of circadian rhythm have been limited to a handful of organisms such as mouse , fruitfly , Neurospora [33] , cyanobacteria [34] , and Arabidopsis [35] , species with large evolutionary distances separating them . Mouse is the only vertebrate species in which the circadian rhythm has been extensively studied . However , other vertebrate model systems have been largely unexplored . In recent years , zebrafish has become a model organism to study the vertebrate circadian rhythm , including at the molecular level [36] , [37] . Our results suggest that the molecular mechanism of the zebrafish circadian rhythm has many characteristics in common with the mammalian system . Following the duplication of many core circadian genes in zebrafish , the relative phase relationship of the duplicates has been largely conserved . Our genome-wide comparison of circadian gene expression between zebrafish and mouse identified novel evolutionarily conserved circadian genes in vertebrates that deserve further functional studies in both species . The transparent nature of larval zebrafish has also made it an excellent system to study light entrainment of the circadian rhythm . Our study shows that the amplitudes of many circadian oscillating genes are significantly reduced in DD compared to LD but their phases are still similar between LD and DD . Comparing gene expression under LD and DD conditions , we discovered that circadian gene expression was induced by light in a progressive manner . The fast light-induced genes showed the most dramatic up-regulation , within eight hours of light-onset . In comparison , expression of the slow light-induced genes rose slowly and reached their peak only near the time of light-off . The transcripts of slow light-induced genes may have longer degradation times than those of fast light-induced genes [6] . Gavriouchkina et al . ( 2010 ) identified a set of light-induced genes in zebrafish from a light exposure experiment in early embryos [38] . Among the 19 light-induced genes identified in their experiment , we detected 16 on our microarray . Twelve of these 16 light-induced genes were identified as fast light-induced genes except for cry2a and cry2b which were identified as slow light-induced genes in our study . The light-exposed samples in Gavriouchkina et al . 's experiment were collected after nine hours of light exposure which is close to the time to reach peak expression after light-onset among our fast light-induced genes . Weger et al . ( 2011 ) analyzed light-induced transcriptome change in zebrafish larvae , heart and cell cultures after one and three hours of light exposure [39] . Although very few genes changed their expression after one hour of light exposure , after three hours of light exposure 74 were up-regulated and 24 down-regulated in larvae . Among the up-regulated genes , we identified 32 as being fast light-induced genes and 8 as slow light-induced genes in our study ( Table S10 ) . Notably , none of the 24 down-regulated genes in Weger et al's study overlap with the dark-induced genes in our study . The distinct kinetics of light induction has also been observed in Neurospora [7] and Drosophila [40] . In Neurospora , early light response genes peak within minutes and late light response genes peak after 30 minutes or more , response times which are much faster than observed in zebrafish . In our study , tefa is a fast light-induced TF binding the D-BOX , whose expression spikes at CT4 under LD . Our promoter analysis also supports the possibility that D-BOX is enriched in the promoters of fast light-induced genes . Among ZCOGs , the targets of D-BOX were enriched around CT5 at which time fast light-induced genes reach their peaks ( Figure 2 ) . Thus we conclude that the up-regulation of fast light-induced genes is under the direct control of light via D-BOX . This observation is consistent with the results of Gavriouchkina et al ( 2010 ) , in which light-induced genes were mostly regulated by tef . Vatine et al . ( 2009 ) have shown that light entrains circadian rhythm via D-BOX in per2's promoter [41] . In Neurospora , the light responses largely depend on the white collar complex ( WCC ) , which consists of WC-1 and WC-2 , via an early light response element ( ELRE ) . D-BOX performs a similar role in fast light-induced genes in zebrafish . Dark-induced genes have been previously observed in both Arabidopsis [42] and cyanobacteria [43] . Fast light-induced genes in zebrafish include core clock genes such as per2 . Therefore dark-induced genes and slow light-induced genes are likely to be regulated by fast light-induced genes through clock regulatory mechanism . In support of this , the expression of clock gene increases during the DD to LL experiment , reaching the peak at CT12 in subjective time . Further studies are needed to separate the direct light-response from the indirect clock regulation . Among the dark-induced genes , aanat1 , a key enzyme that synthesizes melatonin at night , is known to be suppressed by light in the zebrafish retina [44] . In addition , the mouse homolog of dopamine D4 receptor ( drd4a ) has also been reported to be involved in dark sensing [45] . We also observed that two dark-induced genes , mitfa and slc24a5 , are involved in melanogenesis , which led us to investigate the circadian and photic controls of melanogenesis in more detail . On considering previous studies of circadian rhythm in mouse [5] , [46] , we find that our zebrafish network contains homologs of many known mouse circadian TFs . The basic core transcriptional feedback loops are similar in these two networks . However , the zebrafish network includes a number of novel TFs showing circadian rhythm in zebrafish larvae . Among them , only crx has been previously associated with circadian functions . It has been proposed that Crx and cAMP responsive TFs synergistically activate Aanat and Asmt , which encode the two enzymes synthesizing melatonin in rat pineal gland [47] . The mRNA level of Crx in rat pineal gland showed a circadian peak in the middle of night . In our result , the circadian phase of crx ( CT17 ) also occurs in the middle of night and just precedes the circadian phase of aanat1 ( CT20 ) , the zebrafish homolog of Aanat . Similar to mouse , zebrafish crx is also expressed in retina in addition to epiphysis [15] . Crx-deficient mice were affected in circadian entrainment as well as photoreceptor- and pineal-specific gene expression [48] . Among the Crx-targeted genes identified in ChIP-seq experiments in the mouse retina [49] , we identified 40 of them as having zebrafish homologs exhibiting circadian oscillation , including the retinal arrestin gene and prostaglandin synthases . The target genes of some circadian TFs tend to peak at a specific time of day . We predicted that yy1a , oscillating at CT12 , regulates a group of mitochondrial genes oscillating around CT0 . Support for this prediction comes from a Yy1 ChIP-seq assay in human cells that identified Yy1 targets as being enriched in mitochondrial genes [50] . The coordinated oscillation of genes in the mitochondrial respiratory chain has also been observed with mouse SCN [51] . This was suggested to provide the maximal metabolic output during the active phase of SCN neurons [52] . The time-lag between circadian TF peaks and that of their targets can vary from immediate up to 12 hours . This is likely to reflect differences between TFs with respect to time delays in their translation and the trans-activation of their target cis-regulatory elements . The circadian system in the whole zebrafish larva consists of a collection of peripheral clocks [37] , [53] . Although the mRNAs collected from zebrafish larvae are a mixture of transcripts originating from different tissues , we could associate different transcripts with tissues or cell types of expression using the spatial expression atlas in zebrafish . We observed that many circadian TFs show tissue- or cell-type specific expression . We postulated that they are effectors of tissue-specific circadian output from core circadian genes via transcriptional cascades , as implied by the reduction of their amplitude or loss of their rhythmicity in clock morphants . This may be the key to explaining why such a wide range of circadian functions can be driven by a handful of core circadian genes . We then focused on mitfa , a novel circadian TF governing melanogenesis and found not only that melanogenesis followed a circadian rhythm but that this was mediated by circadian oscillation of mitfa . Interestingly , a similar pathway mediating circadian control of melanogenesis has been elucidated for the circadian control of cortisol synthesis in adrenal gland along the hypothalamus-pituitary-adrenal ( HPA ) axis in mouse [54] . In the HPA axis , adrenocorticotropic hormone ( ACTH ) is released by the pituitary gland to stimulate cortisol release in the adrenal gland . In fact , ACTH and α-MSH are generated from the same protein precursor: proopiomelanocortin ( POMC ) in the pituitary gland . Thus , it is conceivable that the HPM axis governs the circadian synthesis of melanin in zebrafish melanocytes in the same way as the HPA axis governs circadian synthesis of cortisol in the mammalian adrenal gland . The dark-inducing or light-suppressing signals may be first transmitted to the hypothalamus from the retina and then relayed to melanocytes through the HPM axis . However , the circadian system in zebrafish has a very different organization from that of the mouse , as individual peripheral clocks in zebrafish are directly light responsive [53] , [55] . Our whole-larva study cannot distinguish the direct impact of light on peripheral clocks from the indirect systemic effects based on light input to specialized photoreceptor tissues such as the pineal complex and deep brain photoreceptors . It is possible that the light signals enter melanocytes and influence the local clock directly . Although melanogenesis is strongly activated by darkness , it is not entirely driven by the light/dark cycle because residual circadian oscillations in mitfa and slc24a5 still persist even under DD conditions . This is perhaps a consequence of the persistent oscillation of cAMP level governed by the local clock . In the zebrafish retina , the concentration of cAMP oscillated in a circadian manner and was regulated by the clock gene [56] . Furthermore , the circadian phase of cAMP concentration is in the subjective early morning , close to the peak time of mitfa . Another dark-induced gene , aanat1 , is also regulated by the cAMP signaling pathway [57] . In our study , two adenylate cyclases , adcy2b and adcy8l ( LOC560410 ) , showed circadian oscillation at CT16 under DD conditions , indicating that cAMP level is likely to be still oscillating under DD conditions in melanocytes . In combination , local clock and external photic signals may synergistically control melanogenesis mediated by mitfa . Organisms can use circadian rhythms to control their specific physiological behavior in adapting to their own ecological niche . In Neurospora , four genes encoding enzymes in the consecutive steps of the carotenoid biosynthesis pathway , which produces photoprotective pigments , are all early light-induced [7] . In contrast , in zebrafish larvae , melanin synthesis is dark-induced which may lead to skin color adaptation to their environment so that they can better evade predators . This may also help them to adjust their daily light-sensitivity in pigment cells . The melanin biosynthesis pathway regulated by mitfa conforms to a network motif common in gene regulatory networks , namely the single-input module ( SIM ) , a master TF controlling a group of target genes [58] . A SIM can generate temporal order of gene expression in metabolic pathways such as arginine biosynthesis in E . coli [59] . The wiring of such modules in the circadian pathway may have evolved to generate a daily change of pigmentation in larval zebrafish . Tissue-specific TFs such as mitfa mediate the circadian control of the central circadian clock by receiving inputs from the central clock and relaying signals to multiple targets , thus generating circadian-linked physiological changes such as circadian melanin synthesis . This phenomenon is not unique to zebrafish . In the mouse heart , the circadian cycle controls cardiac arrhythmogenesis through a Bmal1/Clock targeted TF , kruppel-like factor 15 ( Klf15 ) [60] . Klf15 then regulates the circadian expression of Kv channel-interacting protein 2 ( KChIP2 ) thus leading to a daily change of susceptibility to heart arrhythmia . Notably , TF-encoding genes are enriched among Bmal1 targets in a genome-wide ChIP-seq experiment [6] . Mouse homologs of zebrafish circadian TFs investigated in this study ( e . g . Mef2a , Foxo3 , Ppargc1b , Smad1 , and Maf ) have been identified as Bmal1 targets in mouse . Mef2a in heart , Smad3 in skeletal muscle , and Ppargc1b and Foxo3 in liver showed strong circadian oscillations in their respective tissues [5] . Their tissue-specific circadian functions merit further investigation . Our results suggest that the transcriptional cascade via TFs from a central clock is a universal mechanism that generates diverse circadian functions in all complex organisms .
WT AB strain adult zebrafish ( Danio rerio ) were obtained from the National Zebrafish Resources of China ( Shanghai Institutes for Biological Sciences ) . Fertilized embryos were collected in the morning shortly after lights-on . Zebrafish larvae were maintained in the incubator at 28°C under 14 h∶10 h light/dark cycle from birth . The light was turned on at 9:00 and turned off at 23:00 . The luminance during light exposure was determined to be about 1000 lux as measured by a digital luxmeter ( Model ZDS-10 , SHXL ) on the water surface . Zebrafish handling procedures were approved by the Institute of Neuroscience , Shanghai Institutes for Biological Sciences , Chinese Academy of Sciences . To examine the circadian gene expression of larval zebrafish , we continuously collected larval samples for microarray analysis starting at 5 dpf in both LD ( 14 h∶10 h light/dark ) and DD ( constant dark ) conditions , respectively . WT larvae were raised in 14 h∶10 h LD conditions from birth to 4 dpf . At 4 dpf , the larvae were transferred to 24 dishes ( 35 mm ) each containing 40 larvae . Before lights-on at 5 dpf , 12 dishes remained in the LD conditions and the other 12 dishes were placed in the DD conditions . A total of 40 larvae/dish in LD conditions and 40 larvae/dish in DD conditions were sampled simultaneously at 4 h intervals starting at CT0 ( CT0 = lights-on at 9:00 ) of 5 dpf for 12 time points . None of the larvae in the samples was found to have died during this process . The larvae were sucked into freezing tubes , removed of water , frozen immediately in liquid nitrogen , and stored at −80°C . The collection of samples under dark was performed under dim red light . Total RNA of larval sample was extracted using Trizol ( Invitrogen ) according to the manufacturer's instructions . The quantity and quality of the RNA samples were assessed with a NanoDrop ND-1000 spectrometer ( NanoDrop Technologies ) and an Agilent 2100 bioanalyzer ( Agilent ) . 12 LD RNA samples and 12 DD RNA samples were used for Agilent whole zebrafish 4x44K microarrays , consisting of 43 , 603 probes providing a whole-genome transcriptional profile . Purified total RNA of each sample was amplified and labeled with a fluorescent dye Cyanine 3 ( Cy3 ) using a low-RNA input linear amplification kit following the manufacturer's protocol ( 5184-3523 , Agilent ) . Cy3-labeled cRNA ( 800 ng each ) was hybridized to the zebrafish microarray ( G5219F , Agilent ) for 17 h at 65°C . The hybridized microarrays were then washed according to the manufacturer's protocol . Microarray results were extracted using Agilent G2565BA Scanner and Feature Extraction software ( v10 . 5 . 1 , Agilent ) , and subsequently analyzed by Gene-Spring GX software ( v11 . 0 . 1 , Agilent ) . Microarray data were quantile normalized . The probe sets detected in 75% of the samples were kept for downstream analyses . The annotations for Agilent probe sets including gene names , gene symbols , RefSeq accessions , Genbank accessions , UniGene IDs , Entrez Gene IDs , Ensembl IDs , and TIGR IDs were obtained from the Agilent whole zebrafish 4x44K Microarrays annotation website ( http://www . home . agilent . com/agilent/ ) . For the probe sets with no Gene IDs assigned , we used information contained in the NCBI ( Gene and RefSeq accessions ) and ZFIN databases to provide further annotation ( Text S1 ) . Gene descriptions are based on genome release Zv9 in the Ensembl database ( release 56 ) . The microarray data have been deposited in GEO under accession: GSE37332 . Time-series of microarray expression values were converted from time domain to frequency domain using discrete Fourier transform algorithm . The significance of the observed periodicity compared to random noise was estimated by Fisher's g-test [61] from the GeneCycle package [62] in R . We chose a dominant period of 24 h to identify circadian oscillating genes in our data . The false discovery rate ( FDR ) was estimated by performing Fisher's g-test on randomly permuted time-series data 1 , 000 times . The FDR for genes oscillating in both LD and DD ( ZCOGs ) was estimated from permutations of LD and DD data simultaneously . The selection criteria used were as follows: g-test p values less than 0 . 5 in LD and g-test p values less than 0 . 5 in DD with dominant period as 24 h each corresponded to an overall FDR less than 5% . Circadian phases were estimated by fitting the time-series data to a set of cosine curves with 24 h periods but shifting phases [5] . To identify the genes affected by light-entrainment , we searched for the DD amplitude-reduced genes reflecting either decreased peaks or elevated troughs among LD oscillating genes . We fitted the joint LD/DD gene expression values to a set of cosine curves with 24 h periods of shifting phases in LD but constant levels at either +1 ( elevated troughs ) or −1 ( decreased peaks ) in DD . A G-test p value less than 0 . 15 was used to select 3 , 677 LD oscillating genes ( FDR≤10% ) . A p value less than 0 . 0001 was used in the fit of joint LD/DD expression values . Among the peak-decreased genes , those with LD phases between CT2 and CT16 were defined as light-induced genes . A delay of 2 hours from light period was used here so that the gene expression increases after light onset . In addition , the light-induced genes with LD phases between CT2 and CT8 were defined as fast light-induced genes and those with LD phases between CT8 and CT16 were defined as slow light-induced genes . Among the trough-elevated genes , those with LD phases between CT16 and CT24 or between CT0 and CT2 were defined as dark-induced genes . To examine the expression of light- and dark-induced genes upon light or dark exposure , we raised WT larval zebrafish from birth to 4 dpf at 28°C in DD ( constant dark ) and LL ( constant light ) conditions respectively . At CT0 at 5dpf , DD-control larvae remained in DD and DD-LL larvae were transferred to LL conditions . Similarly , LL-control larvae remained in LL and LL-DD larvae were transferred to DD conditions . Twenty five larvae per sample in each group were collected at CT0 , CT1 , CT2 , CT3 , CT4 , CT6 , CT8 , CT12 , CT16 , and CT20 at 5dpf . None of the larvae in the samples was found to have died during this process . The larvae were sucked into freezing tubes , cleared of water , frozen immediately in liquid nitrogen , and stored at −80°C . The collection of samples under dark was performed under dim red light . Real-time PCR was performed to determine the expression of different types of light- and dark-induced genes . The gene annotations and repeat-masked genome sequences for the five teleost species zebrafish , fugu , medaka , stickleback , and Tetraodon were downloaded from ENSEMBL ( version 62 ) . Promoter sequences , defined as the region 1 kb upstream to 200 bp downstream of the transcriptional start site ( TSS ) , were extracted from each species using Perl Scripts . For each zebrafish gene , we obtained their orthologous gene information in the other four fish species using ENSEMBL homologs data ( version 62 ) . The Pscan program was applied to calculate the enrichment of TF motifs given by TRANSFAC vertebrate TF database for each group of orthologous fish genes [63] . The enriched TF motifs with a p value<0 . 01 and ranking at least in the top 20 were selected for each orthologous gene group . Fisher's exact test was further applied to calculate the enrichment of TF motifs in a given gene set . For each circadian TF , a sliding window approach described in our previous work [5] was applied to identify the specific time window when their putative targets are enriched . In order to identify co-expressed zebrafish genes , we downloaded all the zebrafish Affymetrix array data with raw data in CEL format deposited in GEO , comprising 57 experiments in total . The raw data were normalized by the rma method . Pearson's correlation coefficient was calculated for each pair of genes . A gene pair was considered to be co-expressed if their Pearson's correlation coefficient is larger than 0 . 6 and each gene is within the top 20 most correlated genes of the other member of the pair . Using this criterion , 9 , 926 genes can have at least one other co-expressed gene . The Qcut program [64] was applied to cluster the genes according to their Pearson's correlation coefficients . In total , we obtained 536 clusters of which 435 contained probe sets that can be mapped to the genes on our Agilent zebrafish microarray . For these 435 groups of genes , we calculated the enrichments of the circadian genes , circadian phase , and TF binding sites respectively using Fisher's exact test . The circadian gene regulatory network consisted of circadian TFs and their DNA-binding motifs . They were connected by TF-motif and motif-target relationships . Usually , several similar motifs corresponded to one TF . We grouped the motifs using the motif grouping information in TRANSFAC . The smallest p-value of motif-target predictions in the motif group was selected as the p-value of the motif group with the target . The motif-target prediction with p value<0 . 01 within the top five target predictions for a given motif was used as the criteria to select the motif-target pairs . Fisher's exact tests were used to calculate the enrichment of TF binding sites for the genes expressed in each tissue . The network was visualized using the Cytoscape program [65] . Morpholinos oligonucleotides ( MOs ) were purchased from Gene Tools . All MOs except standard control were designed to target the start codon region of the genes . The sequence of the clock MO was 5′-CAT CCC GGT CTA TGC TGG AGG TCA T-3′ as previously used by Li et al . [56] . The sequence of the mitfa ( nacre ) MO was 5′-CAT GTT CAA CTA TGT GTT AGC TTC A -3′ as previously described [66] . The sequence of the standard control MO was 5′-CCT CTT ACC TCA GTT ACA ATT TAT A-3′ . MOs were used at the following final doses: clock MO: 2 . 5 ng; mitfa MO: 9 ng or 13 ng ( microarray analysis ) . MOs were pressure-injected into 1- to 2-cell stage embryos at a volume of 1 nl using Picospritzer II injectors as previously described [67] . Each MO sample contained 40 individually treated larvae . WT larvae were raised in a 14 h∶10 h LD cycle from birth to 3 dpf at 28°C . From CT0 of 4 dpf , 10–15 larvae were collected under LD conditions at 4 h intervals for a 72 h period . Larvae were fixed with 4% paraformaldehyde ( PFA ) for 12 h at 4°C and were then embedded in 1% low melting point agarose . Images were taken using an Olympus microscope SZX16 equipped with a DP71 CCD camera controlled by DP Controller software . The covered area of melanocyte was measured in a 1 , 360×1 , 024 pixel frame in the head region from the pineal gland to the optic vesicles excluding the eyes using ImageJ 1 . 41 software . WT and MO-injected larvae were raised in 14 h∶10 h LD cycle from birth to 3 dpf at 28°C . WT larvae were collected in LD from CT0 of 4 dpf at 4 h intervals for 72 hours . Control MO and clock MO-injected larvae were collected under both LD and DD conditions simultaneously from CT0 of 4 dpf at 4 h intervals for 48 hours . Each time point consisted of two or three sample replicates and each sample contained 15 individual larvae . The melanin quantification assay protocol used was based on the published work by Maldonado et al . [68] with minor modifications . Briefly , 15 larvae per sample were placed in 300 µl buffer ( 20 mM Tris-HCl , 2 mM EGTA , 1 mM PMSF , pH 7 . 1 ) . They were immediately placed on ice to anesthetize the larvae and then stored at −80°C . The amount of melanin was measured as follows: whole bodies of larvae were homogenized with a disposable pestle ( T10 , IKA ) . Then 100 µl DMSO , 500 µl of 2M NaOH and 100 µl dH2O were added to each tube . The standard melanin was freshly made from the common cuttlefish ( Sepia officinalis ) ( M2649 , Sigma ) at 1 mg/ml in 1% hydrogen peroxide and diluted to a concentration gradient of 0 µg/ml , 50 µg/ml , 100 µg/ml , 200 µg/ml , 300 µg/ml , and 400 µg/ml . The larval samples and standards were then heated for 2 h at 80°C and centrifuged at 12 , 000 g for 10 min . The supernatant was collected and the absorbance analyzed at 350 nm using FlexStation 3 ( Molecular Devices Inc . ) . The melanin quantities were obtained by a linear fit to the standard curve using the OriginPro 8 software . | For most animals whose lives are dependent on the sun , circadian clocks govern their daily behaviors and physiology . In different animals , novel functions under the circadian clock's control can evolve as adaptations to their specific environment . A zebrafish demonstrates a remarkably high level of interplay between external light and its internal circadian clock due to its transparent nature . In a genome-wide study , we identified a large number of circadian oscillating genes as well as genes whose expression is highly sensitive to the light or dark in zebrafish . Our computational analysis of gene regulatory networks revealed a number of transcription factors ( TFs ) that mediate novel circadian functions . We investigated one example in depth , a key TF that relays the control of the circadian clock to the enzymes synthesizing melanin in a dark-induced pathway thus causing the daily change of pigmentation in zebrafish . This dark-induced circadian melanogenesis can lead to an anticipatory change in zebrafish skin color allowing zebrafish to adapt to its environment . This mechanism allows zebrafish to better evade predators and effectively adjust its daily light-sensitivity in the pigment cells . Our study provides an excellent example of how the circadian clock is adapted in a specific organism to control its behavior , thus enabling evolutionary adaptation to the organism's ecological niche . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"microarrays",
"genome",
"expression",
"analysis",
"genomics",
"gene",
"expression",
"genetics",
"molecular",
"genetics",
"biology",
"computational",
"biology",
"genetics",
"and",
"genomics",
"dna",
"transcription"
] | 2013 | Analysis of a Gene Regulatory Cascade Mediating Circadian Rhythm in Zebrafish |
The caterpillar of the moth Premolis semirufa ( Lepidoptera: Arctiidae ) , commonly named Pararama , is endemic of the Amazon basin . Accidental contact with these caterpillar bristles causes local symptoms such as intense heat , pain , edema and itching which last for three to seven days; however , after multiples contacts , it may induce joint-space narrowing and bone alteration , as well as degeneration of the articular cartilage and immobilization of the affected joints . Specific treatment for this disease does not exist , but corticosteroids are frequently administered . Despite of the public health hazard of Premolis semirufa caterpillar poisoning , little is known about the nature of the toxic components involved in the induction of the pathology . Here we have investigated the biological and immunochemical characteristics of the caterpillar's bristles components . Analysis of the bristles extract in in vitro assays revealed the presence of proteolytic and hyaluronidase activities but no phospholipase A2 activity . In vivo , it was observed that the bristles extract is not lethal but can induce an intense inflammatory process , characterized by the presence of neutrophils in the paw tissues of injected mice . Furthermore , the bristles components stimulated an intense and specific antibody response but autoantibodies such as anti-DNA or anti-collagen type II were not detected . The results suggest that Premolis semirufa caterpillar bristles secretion contains a mixture of different enzymes that may act together in the generation and development of the clinical manifestations of the Pararama envenomation . Moreover , the high immunogenicity of the caterpillar bristles components , as shown by the generation of high antibody titers , may also contribute to the induction and establishment of the inflammatory disease .
Moths and butterflies are insects of the Lepidoptera order , of which the young stage is called larva . The larval form of some families of moths containing urticating hairs is known as caterpillar . Although caterpillar venoms have not been analyzed as much as the venoms from snakes , spiders and scorpions , there are many reports on the characterization of bristles extracts from a variety of species . Coagulation disorders have been reported after contact with the Saturniidae caterpillars from Lonomia genus . Since 1989 , accidents involving Lonomia obliqua species were reported in South of Brazil , Argentine , Paraguay and Uruguay [1] , [2] . The physical contact with this caterpillar induces a toxic secretion from bristle , which promotes local and systemic symptoms in the victim between 6 to 72 hours after contact , such as burning sensation , intense hematuria , disseminated intravascular coagulation-like reactions ( severe depletion of the coagulation factors ) and secondary fibrinolysis [2] . Serious clinical complications , such as acute renal failure and intracranial hemorrhage may also occur [1] , [3] . The Brazilian caterpillar of Premolis semirufa usually called as Pararama , belongs to the Arctiidae family . The genus Premolis contains four species: P . semirufa , recorded in the Amazon region in Brazil , French Guiana , Ecuador , Peru and Panama; P . excavata found in Panama; P . rhyssa in Peru and P . amaryllis in French Guiana . Premolis semirufa feeds of Hevea brasiliensis , the rubber tree found in the Amazon forest ( Figure 1 ) . The tappers , when collecting the latex , can stick their fingers in the trunk of the rubber trees to facilitate the harvest and , at that time , may come into contact with the Pararama . Known as “Pararama associated phalangeal periarthritis” and due to its importance as an occupational disease , predominantly in the rubber tree areas of Pará , Brazil , this caterpillar envenomation was inserted into the “Manual of diagnosis and treatment of envenomations” , released by the Brazilian Ministry of Health in 1992 [4] . The contact with the bristles , in most cases , causes instantly an intense itching , followed by symptoms of the acute inflammation such as pain , heat and redness , which lasts up to seven days , after the first accident [5]–[11] . Chronic symptoms , which frequently occur in individuals after multiple accidents , are characterized by synovial membrane thickening , with joint deformities and chronic synovitis ( mono or oligoarticular ) , symptoms similar as those found in rheumatoid arthritis . So far , there is no effective treatment for the accidents with Pararama , since neither the toxic components of the bristles of the caterpillar nor the mode of action of the venom are known . However , systemic corticosteroids treatment has been used , in the belief that this would prevent the onset or attenuate the chronic disease [7] , [12] , [13] . In case of infection , due to the scratching and unhygienic conditions , the disease may progress to pyogenic arthritis [11] . Despite being a serious problem in occupational medicine and a social problem affecting the Brazilian Amazon region , since the rubber tappers can no longer return to their activities , which are the source of their livelihood , studies on the pathogenesis of Pararama are scarce . Thus , the aim of the present study was to analyze the biological and immunochemical characteristics of Premolis semirufa caterpillar's bristles crude extract .
Triton X-100 , Tween-20 , Hepes , bovine serum albumin ( BSA ) , cetyltrimethylammonium bromide ( CTAB ) , ortho-phenylenediamine ( OPD ) , phosphatidylcholine , bromothymol blue , gelatin , Coomassie Brilliant Blue R-250 , phenylmethylsulfonyl fluoride ( PMSF ) , 1 , 10-phenanthroline , hyaluronic acid , anti-mouse IgG horseradish labelled with peroxidase ( IgG-HRPO ) , native salmon sperm DNA and collagen from bovine tracheal cartilage were purchased from Sigma–Aldrich ( Missouri , USA ) . Anti-mouse IgM , -IgG1 , -IgG2a HRPO-conjugate and anti-mouse IgG2b , IgG3 biotin-conjugate were purchase from BD Bioscience ( California , USA ) . Anti-mouse IgG labelled with alkaline phosphatase ( IgG-AP ) , 5-bromo-4-chloro-3-indolyl-phosphate ( BCIP ) and nitroblue tetrazolium ( NBT ) were from Promega Corp . ( Wisconsin , USA ) . Brij-35 P was purchased from Fluka – BioChemika ( Werdenberg , Switzerland ) . Fluorescence Resonance Energy Transfer ( FRET ) substrates were synthesized and purified according to Araújo et al . [14] . Caterpillars from Premolis semirufa were collected in the city of São Francisco do Pará , Pará , Brazil , and maintained at the Immunochemistry Laboratory , Butantan Institute , SP , Brazil . The bristles extract was prepared after exposing the caterpillars to 4°C for few minutes; the bristles were cut off with scissors at their insertion in the tegument , avoiding any tegument incision and , then , suspended in cold phosphate-buffered saline - PBS ( 8 . 1 mM sodium phosphate , 1 . 5 mM potassium phosphate , 137 mM sodium chloride and 2 . 7 mM potassium chloride , pH 7 . 2 ) . This suspension was macerated with the aid of a glass stick , homogenized and centrifuged at 560×g for 20 min at 4°C . The supernatant was collected and its protein content was determined by using the BCA Protein Assay Kit ( Pierce Biotechnology , MA , USA ) . Supernatant aliquots were stored at −80°C until use . Venoms from Micrurus hemprichii and Bothrops jararaca snakes , which were used as positive controls in the assays for determination of PLA2 and hyaluronidase activities , respectively , were supplied by Herpetology Laboratory from Butantan Institute , SP , Brazil . The authorization to access the venoms of Premolis semirufa caterpillar , Bothrops jararaca and Micrurus hemprichii snakes were provided by the Brazilian Institute of Environment and Renewable Natural Resources - IBAMA - a Brazilian Ministry of the Environment's enforcement agency ( permission no . 01/2009 ) . BALB/c strain male mice aged 2 months and weighing 18–22 g were obtained from Central Animal Breeding from Butantan Institute , SP , Brazil . All the procedures involving animals were in accordance with the ethical principles in animal research adopted by the Brazilian Society of Animal Science and the National Brazilian Legislation no . 11 . 794/08 . Protocols were approved by Institutional Animal Care and Use Committee ( protocol approval number 413/07 ) . The caterpillar bristles extract ( 10 µg of protein ) was solubilized in sample buffer , using non-reducing and reducing conditions , and separated on 12% SDS-PAGE gel [15] . Molecular weight markers were included in all runs . Gels were stained with silver [16] . The Phospholipase A2 activity of Premolis semirufa's bristles extract was determined as described by Price III [17] , with some modifications . Samples of the extract ( 4 µg or 16 µg of protein ) , 20 µL HCl ( positive control ) , or 20 µL PBS ( negative control ) were mixed in 96-well microtitre plates . 180 µL of an assay mixture containing 10 mM Triton X-100 , 5 mM phosphatidylcholine , 1 . 5 mM HEPES , 10 mM calcium chloride , 0 . 9% sodium chloride and 0 . 03% ( wt . /vol . ) bromothymol blue dye in water , at pH 7 . 5 and 37°C , were added . The plate was analyzed at λ 620 nm in a spectrophotometer ( Multiskan EX , Labsystems , Finland ) after 5 min of incubation and the linearity of the reaction was verified by linear regression ( MSExcel 2007 ) . All enzymatic assays were performed in duplicate and expressed as specific activity ( nmol/min/µg ) . As positive control for PLA2 activity , venom of the snake Micrurus hemprichii ( 4 . 0 µg ) was used . Hyaluronidase activity was measured as described by Pukrittayakamee [18] , with slight modifications . In a microtitre plate , Premolis semirufa's bristles extract ( 8 . 0 µg of protein ) were mixed with 25 µL of the hyaluronic acid ( 0 . 5 mg/mL ) and acetate buffer ( 0 . 2 M sodium acetate-acetic acid , pH 6 . 0 , containing 0 . 15 M NaCl ) , in a final volume of 100 µL , and incubated for 30 min at 37°C . After incubation , 200 µL of CTAB 2 . 5% in NaOH 2% was added to the samples . The absorbances were measured at λ 405 nm in a spectrophotometer ( Multiskan EX , Labsystems , Finland ) against a blank containing hyaluronic acid , acetate buffer and 250 µL of CTAB . All assays were performed in duplicate . Results were expressed in units of turbidity reduction ( UTR ) per mg of extract . Bothrops jararaca snake venom ( 8 . 0 µg ) was used as positive control . Lethality was assessed by intraperitoneal injection of increasing amounts of bristles extract in 200 µL of PBS into male BALB/c strain of mice . Four animals were used for each dose and the LD50 was calculated by probit analysis of death occurring within 72 h after extract injection [20] . The possible edematogenic activity of the caterpillar bristles extract was evaluated by BALB/c mice intraplantar injection of 50 µL of sterile PBS containing 10 µg ( protein ) of the extract into the left hind footpad . As control group , mice received 50 µL of sterile PBS into the left hind footpad . The animals were injected , seven times , at intervals of two weeks . Before extract or PBS inoculations , the thickness of each left footpad ( Th0 ) was determined using a caliper measurement ( Mitutoyo , Sul American Ltda . ) . Subsequent readings of the thickness ( Tht ) after extract or PBS injections were carried out at 30 , 60 , 120 and 180 min , and compared to the initial readings . The edema ( E ) was calculated as follows: E [%] = [ ( Tht−Th0 ) /Th0]×100 . Where Tht is the thickness ( mm ) of the footpad at time “t” after the injection of the extract or PBS . Th0 is the thickness ( mm ) of the footpad before the injection of the extract or PBS . BALB/c mice , injected as described above , were euthanized 24 h after the 7th extract inoculation , their hind limbs removed and processed for histological analysis . The paws were immersed in 10% neutral buffered formaldehyde solution for 24 h . After decalcification , the tissues were embedded in paraffin , sectioned and stained with hematoxylin and eosin ( H&E ) . The H&E preparations were microscopically observed and examined for the presence of inflammatory cell infiltration . As control , paws injected with an equal volume of PBS , were collected , processed and analyzed as described above . All tissue sections were examined under a light microscope ( Leica DM2500; Wetzlar , Germany ) . The antiserum against Premolis semirufa's bristles extract was obtained from mice inoculated with 10 µg ( protein ) of the extract or PBS , into the left hind footpad , seven times at two weekly intervals , without adjuvant . Bleeding was carried out , by retro-orbital plexus with a Pasteur pipette , 48 h after the injection . The blood was allowed to clot at room temperature for 15 min and then was left at 4°C for 6 h . After centrifugation at 560×g for 15 min at 4°C , the serum was collected and immediately frozen at −20°C until use . Statistical analysis was performed by Students't-test using GraphPad Prism software . Differences were considered statistically significant when P values were P<0 . 05 , P<0 . 01 and P<0 . 0001 .
Extract samples collected from Pararama bristles were prepared in PBS and analyzed for protein composition using SDS-PAGE , under reducing and non-reducing conditions . Figure 2 shows that the electrophoretic profiles of the extract , analyzed under both conditions , were similar , showing components with Mrs between 20 and 200 kDa and the presence of an intense band with Mr around 82 kDa . In order to assess Premolis semirufa's bristles extract toxicity , the extract was tested using a variety of functional biochemical assays , to identify if it contained activities frequently found in animal venoms . The lethal toxicity of the bristles extract was determined in groups of BALB/c mice , after intraperitoneal injection of increasing protein concentrations of the extracts ( 1 . 2 mg/kg , 2 . 3 mg/kg and 6 . 8 mg/kg ) and no death was observed after 72 hours of the inoculation ( data not shown ) . Moreover , in this condition , no manifestation of discomfort was observed in any of the envenomated animals . The phospholipase A2 ( PLA2 ) activity of P . semirufa caterpillar bristles extract was assessed by a colorimetric method after incubating samples of 4 or 16 µg of the extract with phosphatidylcholine , the substrate of the PLA2 . Figure 3A shows that the venom of the caterpillar showed no PLA2 activity , while the Micrurus snake venom ( 4 µg ) , used as positive control , presented a high lipase activity on phosphatidylcholine . The hyaluronidase activity of the caterpillar bristles was measured incubating samples of the extract ( 8 µg of protein ) with hyaluronic acid , the substrate of the reaction . As positive control of the reaction , it was used the venom from Bothrops jararaca snake . Figure 3B shows that the bristles extract present significant hyaluronidase activity . The proteolytic activity of the bristles extract ( 1 µg ) was tested using the fluorescence resonance energy transfer ( FRET ) peptide Abz-FRSSRQ-EDDnp as substrate . Figure 3C shows that the extract efficiently hydrolyzed the FRET peptide , and that this activity was strongly inhibited by the serine protease inhibitor PMSF ( 88% ) and partially by the metalloprotease inhibitor phenanthroline ( 50% ) . Figure 3D shows that a 82 kDa component , the Mr corresponding the intense protein band observed after silver staining ( Figure 2 ) , has a high gelatinolytic activity , as measured by zymography ( Fig . 3 - line 1 ) . This activity was significantly inhibited by PMSF ( Fig . 3 - line 3 ) , a serineprotease inhibitor , and poorly blocked by phenanthroline ( Fig . 3 - line 2 ) , a metalloprotease inhibitor . BALB/c mice were injected seven times , at intervals of two weeks , with 10 µg of the extract proteins into the foot pad of the left hind leg . Controls animals were injected with PBS . The intraplantar injection of Premolis semirufa caterpillar bristles extract , caused discomfort to the animals ( pain ) and a significant increase in the paw volume , as compared to that induced by injection of the vehicle , i . e . , PBS ( Figure 4A and B ) . The edema induced by both , extract and PBS , was detected as early as 5 min post-injection and peaked at 30 min . The increase in paw volume was observed until 300 min after injection of the extract , while the increase induced by PBS was resolved 120 min after injection . The comparison of the edematogenic responses , along the seven inoculations of the extract or PBS , determined at the peak of the reaction , i . e . , at 30 min is shown in Figure 4C . The edematogenic responses were significant and successively more intense after the inoculations of the extract compared to PBS and reached a maximum after the 4th injection . BALB/c mice , injected as described above , were euthanized 24 h after the 7th extract or PBS inoculations , their hind limbs removed and processed for histological analysis . Figure 5B shows that the bristles extract injection resulted in the establishment of a pronounced inflammatory reaction , characterized by the presence of mixed inflammatory cellular infiltrate distributed throughout the tissue . Furthermore , the connective tissue was increased , partially occupying areas where , in normal tissues , structures such as sweat glands and fat tissue were found ( Figures 5A and C ) , and initiating a fibrotic process ( Figure 5D ) . The immunogenicity of Premolis semirufa caterpillar bristles extract was assessed by ELISA , using sera obtained from BALB/c mice subcutaneously inoculated with 10 µg of the extract proteins or PBS . Figure 6A shows that the repeated inoculations of the extract , but not of PBS , induced a high IgG antibody response . In addition , analysis of antibody classes and subclasses revealed that the sera obtained from animal injected with the extract presented higher IgG1 , IgG2a , IgG2b and IgM titers as compared to the sera collected from PBS injected mice . IgG1 sera titers , determined for envenomated animals , were higher than the others antibodies isoptype/sucblcasses and no IgG3 antibodies could be detected in these samples ( Figure 6B ) . The presence of anti-DNA or anti-Collagen type II IgM and IgG autoantibodies in sera from BALB/c mice inoculated with extract or PBS , was evaluated by ELISA . Anti-DNA or anti-Collagen type II IgG antibodies were not detected in sera from bristles extract inoculated animals , while high titers of these antibodies were detected in the serum of mice with systemic lupus erythematosus ( SLE ) autoimmune disease . Anti-DNA and anti-Collagen type II IgM antibodies were not detected in bristles extract injected or SLE mice ( data not shown ) .
We have investigated activities of the bristles extract of Pararama , a caterpillar responsible for the occupational disease ‘Pararama associated phalangeal periarthritis’ . Until now nothing was know about the composition of its venom . In this paper we present , for the first time , some of the biochemical and biological properties of Premolis semirufa's bristles extract . Electrophoretic analysis of the extract showed that it contained a great diversity of proteins , with Mr ranging from 20 to 200 kDa , with a major protein band of 82 kDa , contributing to over 90% of the protein content . No significant difference in the protein profile was observed in the extracts submitted to reducing or non-reducing conditions . We subjected the venom to a variety of functional biochemical assays , to identify if it contained activities frequently found in animal venoms . Hyaluronidase activity is present in many animals venoms and its activity potentiates the toxicity of the venom , promoting loss of extracellular matrix integrity of soft connective tissues , surrounding the blood vessels , increasing the systemic influx of toxins and , thus , facilitating the dispersion of the toxic components [22] . Premolis semirufa's bristles extract showed significant hyaluronidase activity , suggesting that this enzyme may participate in the genesis of the joint immobility , since hyaluronic acid is an abundant component of the intercellular matrix of the skin , cartilage and synovial fluid , playing an important role as stabilizer and lubricant of the joints [23] . The hyaluronic acid degradation may explain , in part , the changes in the joint and loss of the cartilage and bone structure , seen in the pararama induced disease . Zymography analysis showed that the 82 kDa component found in the caterpillar bristles extract possesses gelatinolytic activity . Gelatinases are capable of degrading types IV , V , VII and XI collagens , present in bone and articular cartilage , and may regulate their remodeling [24] . Using specific inhibitors for metallo- and serine- proteases we identified the gelatinase as a serine protease . The bristles extract also demonstrated high proteolytic activity towards the FRET peptide Abz-FRSSRQ-EDDnp . The use of PMSF showed that serineproteases were largely responsible for this while using the metalloproteases inhibitor phenanthroline demonstrated that metalloproteases were involved as well . Venom serineproteinases have a highly diverse pharmacological potential , including actions on proteins of the coagulation cascade , activation of factor V , activation of protein C , fibrinogenolysis , activation of the plasminogen and induction of platelet aggregation [25] . Thus , it is possible to propose that serineproteases , with gelatinolytic activity and other proteolytic activities , may be involved in the process of cartilage and joint degradation produced by contact with the bristles of Pararama [26] . Metalloproteinases , abundant molecules in snake venoms , are responsible for the development of local tissue injury and the occurrence of bleeding [27] , being able to degrade important components of the matrix , such as laminin and type IV collagen [28] . Thus , the presence of different classes of proteases in the Pararama bristles extract may contribute to the tissue injury seen in the caterpillar human accidents . In many animal venoms , Phospholipase A2 ( PLA2 ) is important for digestion and immobilization of the prey , as well as responsible for some pathologies observed in humans stung/bitten by bees , wasps , spiders and snakes [29]–[31] . Phospholipase A2 activity has also been described in crude bristles extract of the Euproctis ( Lymantriidae ) caterpillar [32] and more recently in bristles crude extract of Lonomia obliqua ( Saturniidae ) [33] . However , Premolis semirufa's bristles extract did not show phospholipase A2 activity . The present study also aimed to evaluate the toxicity of the bristles extract using a murine model and , under the experimental conditions used , no discomfort or death was observed . On the other hand , the intraplantar injection of the bristle extract , as used in histopatological/edema studies , caused a strong discomfort to the animals , suggesting that they were feeling pain . Further studies will be conducted in order to analyze the possible hyperalgesic properties of the pararama venom . Toxic venom proteins serve in a number of adaptive functions such as immobilizing , paralyzing , killing , liquefying prey and deterring competitors . Other proteins may act synergistically by enhancing the activity or spreading of toxins . In contrast to animals such as snakes and scorpions , which use venoms to immobilize prey and to facilitate its digestion , caterpillars feed on leaves; their venoms are used solely for defense [34] and , therefore , has not to be necessarily lethal . The first intraplantar injection of Premolis semirufa's bristles extract produced a swelling which was detected after 5 min , peaked at 30 min and disappeared within 300 min after injection , while the response to PBS disappeared within 120 min . The prolonged and increased induction of the edema upon 1st exposure was likely to be mediated by the action of the venom , being pro-inflammatory itself or by inducing inflammatory mediators , locally released or synthesized in the course of the envenomation , all of which would increase the permeability of the microvessels . The response induced 30 min after extract injection , gradually increased over the inoculations and reached a maximum after 4 injections . This response was significantly higher than the response to PBS . In addition , the multiple extract injections in mice footpads induced a pronounced inflammatory reaction , characterized by the presence of mixed inflammatory infiltrate , with increase of the conjunctive tissue and beginning of the fibrosis process . In a previous study using a rat model , Costa and collaborators [8] have shown that inflammation was induced by the injection of saline extract of pararama bristles , with the presence of a large number of inflammatory cells around the site of injury . Investigation on the P . semirufa's bristles extract immunogenicity revealed that the repeated inoculations of the extract , in the absence of adjuvants , induced a strong immune response , with high antibody titers . Moreover , data showed that the sera obtained from animal injected with the extract presented higher IgG1 titers than other IgG subclasses , indicating the predominance of a Th2 immune response , since this particular antibody subclass is mainly induced by the presence of Th2 cytokines such as IL-4 , IL-5 , IL-10 and IL-13 [35] . The disease caused by the contact with the Premolis semirufa's bristles shares many features with those found in patients with rheumatoid arthritis ( RA ) , a systemic and chronic illness , characterized by severe synovial inflammation and cartilage and/or bone destruction [36] . Autoantibodies , such as anti-collagen type II and anti-DNA , are found in the vast majority of patients [37] . Analysis of the presence of anti-DNA or anti-collagen type II antibodies revealed that these autoantibodies were not present in the sera obtained from mice inoculated with the Premolis semirufa's bristles extract . Together , these data show the existence , in the Premolis semirufa's bristles extract , of a mixture of different enzymes that may be acting together in the generation and development of clinical disease manifestations . Moreover , this study demonstrates the production of high antibody titers in mice inoculated with the extract , which may also contribute to genesis of inflammatory reactions observed in the envenomation . The absence of autoantibodies indicate that the molecular mechanisms causing disease after multiple contact with the Premolis semirufa's bristles differ from that observed in chronic synovitis , such as the rheumatoid arthritis . The bristles toxic action , high antibody response with the formation of immune complexes and complement activation may also play a role in the establishment of the disease . These aspects will be further investigated in future studies . | Pararama , the popular name of the larval form of the moth Premolis semirufa inhabits rubber plantations in the Amazon region and the accidental contact of the skin with the caterpillar's bristles or cocoons results in immediate and intense heat , pain , edema , and itching . In many cases a chronic inflammatory reaction with immobilization of the joints occurs . The current study has evaluated the biological and immunochemical characteristics of the Pararama caterpillar bristles extract . Electrophoretic analysis showed the presence of several components , including a very intense 82 kDa band . This latter component was endowed with intense gelatinolytic activity , as observed in zymography assays . Further analysis revealed that the extract also contained hyaluronidase activity but is devoid of phospholipase A2 activity . In vivo assays , using mice , showed that the extract was not lethal , but caused significant edema and induced intense infiltration of inflammatory cells to the envenomation site . The extract also induced high specific antibody titers , but no autoantibodies were detected . The data obtained , so far , demonstrate the existence of a mixture of different enzymes in the bristles of Premolis semirufa caterpillar , which can act together in the generation and development of the clinical manifestations of the Pararama envenomation . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"biochemistry",
"immunology",
"biology"
] | 2012 | Premolis semirufa (Walker, 1856) Envenomation, Disease Affecting Rubber Tappers of the Amazon: Searching for Caterpillar-Bristles Toxic Components |
We propose a novel biologically constrained three-phase model of the brain microstructure . Designing a realistic model is tantamount to a packing problem , and for this reason , a number of techniques from the theory of random heterogeneous materials can be brought to bear on this problem . Our analysis strongly suggests that previously developed two-phase models in which cells are packed in the extracellular space are insufficient representations of the brain microstructure . These models either do not preserve realistic geometric and topological features of brain tissue or preserve these properties while overestimating the brain's effective diffusivity , an average measure of the underlying microstructure . In light of the highly connected nature of three-dimensional space , which limits the minimum diffusivity of biologically constrained two-phase models , we explore the previously proposed hypothesis that the extracellular matrix is an important factor that contributes to the diffusivity of brain tissue . Using accurate first-passage-time techniques , we support this hypothesis by showing that the incorporation of the extracellular matrix as the third phase of a biologically constrained model gives the reduction in the diffusion coefficient necessary for the three-phase model to be a valid representation of the brain microstructure .
Brain tissue is naturally divided into two domains , the intracellular space ( ICS ) and the extracellular space ( ECS ) . The ICS is a tightly packed composite of neurons , glia and their cellular extensions , and the ECS is the microenvironment that separates brain cells [1] . The structural maintenance of the ECS is essential for normal brain functioning , as intercellular communication , nutrient transport and drug delivery all depend on ECS integrity . Many pathological brain conditions are associated with changes in ECS size and geometry , including ischemia , inflammation , and tumor progression [2] . Given that the brain is naturally divided into these distinct regions , its microstructure can be well-described as a random heterogeneous material , a medium that is composed of randomly arranged domains of different phases [3] . Brain tissue has conventionally been modeled as a two-phase material , where the two domains are the ICS and the ECS . Despite the seemingly natural classification of the brain as a two-phase material , it is important to note that the ECS is actually a heterogeneous composite of ions , transmitters , metabolites , peptides , neurohormones and molecules of the extracellular matrix ( ECM ) [2] . Using current imaging techniques , the ECS can be visualized in two dimensions , but not in three dimensions . An electronmicrograph done by Dr C . B . Jaeger of a small region of rat cortex can be found in [4] . According to the theory of random heterogeneous materials , macroscopic properties of a medium provide an average measure of the underlying microstructure [3] . This is particularly useful in the case of brain tissue , since detailed three-dimensional ( 3D ) microstructural images do not exist , but macroscopic properties can easily be measured . The ECS volume fraction φ1 and effective diffusion coefficient De are the two macroscopic parameters commonly employed to give an average description of brain's microstructure [2] . In fact , diffusion analysis can be applied to study the the microstructure of any tissue type [5] . For brain tissue in particular , the real-time iontophoretic ( RTI ) method has been applied to determine macroscopic properties . Using tetramethylammonium ( TMA+ ) as the tracer , it has been measured that φ1 = 0 . 2 and De = 4 . 8×10−6 cm2 s−1 [1] . More commonly , one represents the effective diffusion coefficient using one of two dimensionless quantities: the dimensionless effective diffusivity , defined as D* = De/D1 , or tortuosity , defined as λ = ( D1/De ) 1/2 , where D1 is taken to be the diffusion coefficient of TMA+ in agarose [1] , [2] . We note here that in the field of random heterogeneous materials , tortuosity is defined as λ = D1/De [3] . The definition used here is consistent with that used by others studying diffusion in brain tissue . It has been measured that D1 = 1 . 2×10−5 cm2 s−1 , giving a brain tortuosity of λ≈1 . 6 and a dimensionless effective diffusivity of D*≈0 . 40 . In this paper , all results will be presented in terms of D* . Volume fraction and diffusion measurements that differ significantly from these average values are hallmarks of pathological brain states , highlighting that these macroscopic parameters capture significant microstructural information . Nonetheless , these parameters cannot fully describe the underlying microstructure , as this can only be done via an infinite set of n-point correlation functions [3] . Armed with information about the brain's macroscopic properties , theoretical models of the microstructure can be developed . Previous attempts have been made to model the brain microstructure as a two-phase isotropic material composed of uniformly spaced closely packed convex cells [6]–[10] . In the studies that simply treat brain cells as homogeneous impenetrable obstacles , it has been predicted that the effective diffusivity of brain tissue is well-approximated by the two-phase Hashin–Shtrikman ( HS ) upper bound ( 1 ) where d is the spatial dimension [3] , [6]–[9] . At ECS volume fraction φ1 = 0 . 2 , the 3D HS bound predicts that D*≈0 . 71 , a diffusivity significantly larger than that measured in brain tissue . For this reason , it is clear that two-phase models composed of uniformly spaced closely packed convex cells lack some key features of brain tissue . Recent work in ischemic brain tissue suggests that dead-end microdomains are a major determinant of extracellular tortuosity , although it is currently unknown if such voids and dead-ends exist in normal brain tissue [11] . Since dead-ends have been implicated in ischemia , several approaches have been taken to incorporate dead-ends into two-phase models of healthy brain tissue . In one approach , dead-ends arise because cubic brain cells are allowed to overlap . An unrealistically large number of concavities are found when cubic cells can overlap , giving diffusivities significantly below that measured in brain tissue [12] . In another approach [6] , [13] , rectangular dead-end cavities ( open at one end to the ECS ) are punched into convex cellular elements . While the technique can yield a model with the measured diffusivity of brain tissue , brain cell bodies are generally described as being convex objects with fine cellular extensions emanating from the body [14] , not a geometry consistent with representing brain cells in this manner . While the notion of cellular convexity has not been verified for all cells , there is certainly no biological evidence suggesting that concavities are found in all brain cells . Thus , given the known features of brain cells , including the lack of evidence that an abundance of concavities exist in brain cell bodies , it is reasonable to assume that concavities play a role in , but are not the sole factor responsible for the diffusivity of brain tissue being significantly smaller than that predicted by models of uniformly spaced convex cells . If we work under the assumption that the majority of brain cell bodies are convex , alternate mechanisms need to be implemented to develop a realistic microstructural model that preserves the topological and geometric features of the ICS and ECS while simultaneously having the correct diffusion properties . Designing an appropriate microstructural model for tissue in general is tantamount to a packing problem; i . e . , dense aggregates of cells or “particles” [3] , [15]–[19] . Approaching model development from this perspective lends further support to the notion that most brain cells are convex , as tightly packing these cells in space is a densification procedure , and such processes tend to drive the shape of the object being packed towards convexity . Given the low porosity and diffusivity of brain tissue , the nonoverlapping mostly convex cells of the ICS have no choice but to pack tightly . With the appropriately chosen packing procedure , realistic geometric features of brain tissue will naturally emerge . However , it is important to note that when nonoverlapping convex cells are packed in 3D space , very few dead-ends actually arise . This is because topological connectedness increases with dimension [3] , and although packings of convex cells in 2D can result in a significant amount of dead-end space , this is not the case in 3D . Based on this observation , we conclude that an insufficient amount of dead-end space naturally arises because of the size , shape and distribution of brain cells . This does not rule out the possibility that other factors , such as glial cell processes or ECM macromolecules , can lead to the formation of dead-end microdomains in the brain , but it does highlight that current models are not appropriately accounting for a dominant mechanism that contributes to brain tissue tortuosity . In this paper , we propose a novel three-phase model of the brain microstructure that obeys known properties of the ECS and ICS , naturally develops a small number of dead-end microdomains due to cell shape and position , and accounts for diffusion hindrance by ECM macromolecules . In light of the highly connected nature of 3D space , as well as the previously mentioned fact that the ECS , unlike water , is a restricted medium for diffusion [20] , we hypothesize that the inclusion of the ECM as the third phase in our proposed model can give the decrease in the diffusion coefficient necessary for the model to have the same diffusivity as brain tissue . To confirm that the model is a realistic representation of the brain microstructure , diffusion properties of the medium both with the ECM ( three-phase model ) and without the ECM ( two-phase model ) are studied using an accurate state-of-the-art technique borrowed from the field of random heterogeneous materials: a first-passage-time Monte Carlo simulation [21] . We find that by adding the ECM to our two-phase model ( which acts to reduce the free diffusion coefficient of the ECS in accordance with previous experimental observations [20] ) the model can achieve an order of magnitude decrease in the diffusion coefficient and , at the appropriate ECM concentration , conforms to the diffusion properties measured in brain tissue . From this study , we argue that two-phase media subject to a proposed set of biological constraints , which includes limiting cells to be mostly convex bodies , cannot achieve the diffusion parameters of brain tissue . We have shown that , as suggested from experimental data , the addition of the ECM gives the decrease in the diffusion coefficient necessary for the model to conform to the macroscopic properties of brain tissue . It is plausible that the novel arrangement of cells proposed herein , along with the implementation of the ECM , can further benefit by incorporating more dead-ends . Nonetheless , the contribution of this work is to highlight the importance of , and suggest a method to implement ECS heterogeneity in realistic brain microstructural models .
To accomplish our goal of developing a realistic microstructural model , we have compiled a list of brain tissue features: Despite the semipermeable nature of cell membranes , in our model we assume that all cells of the ICS are impenetrable . The development of a realistic microstructural representation of brain tissue is independent of the permeable nature of the ICS , and therefore this simplification is justified . Another important feature of the brain microstructure is the cellular processes that emanate from cell bodies . Axons and dendrites , the processes that arise from neuronal cell bodies , are generally convex structures with vastly different morphologies . Axons are typically a single long cylindrical structure , whereas dendrites are branching cylindrical structures [1] . Glial cell processes are thin sheet-like structures that exhibit a wide range of morphological variability . While these processes are certainly important in the brain , the complexity of their structures makes it very difficult to incorporate them into a brain microstructural model . It is certainly plausible that allowing some cellular concavities and dead-ends to persist in the model may grossly account for these features , as has previously been tackled [6] , [13] . Without denying the validity of this approach , our goal here is to limit the number of concavities and dead-ends ( sticking with the assumption of mostly convex cellular bodies ) and yet develop a realistic microstructural model with the correct diffusion properties . In order to develop our three-phase model , we begin by proposing a novel two-phase model that accounts for the four properties of brain tissue . In lieu of property P1 , the intracellular space must be composed of mostly convex objects . Previous theoretical work has concluded that convex cells of different shapes arranged comparably give rise to the same medium tortuosity [3] , [9] . In particular , provided that the cells are compact convex shapes of high symmetry and have the same spatial arrangements , one can be certain the diffusion properties will be comparable even if the shapes are different [3] . If one is not careful and chooses shapes without high symmetry and then also uses a different spatial arrangement , it is then the case that the shape of the cell can influence the diffusivity of the medium . Since biological cells are higly symmetric , in order to achieve the desired porosity in our model , brain cells are represented by the most basic convex shapes: squares in 2D and cubes in 3D . We will generally use the term cube to describe both squares and cubes for succinctness . Ordered configurations of uniformly spaced cubes on a lattice have already proven to be an unsatisfactory model of the brain microstructure , as the medium is not sufficiently tortuous [9] . In an effort to develop a more tortuous model , we propose a packing construction that exhibits both brain cell size and shape variation ( P2 ) , as well as nonuniformity of spacing between brain cells ( P3 ) . To develop our novel model of nonoverlapping and nonuniformly spaced cubes , begin by dividing space into N×N squares ( 2D ) or N×N×N cubes ( 3D ) . In order to balance computational restrictions with the desire to simulate a large number of cells , N was taken to be 30 in 2D ( giving 900 cells ) and 7 in 3D ( giving 343 cells ) . Furthermore , in 2D each square element was divided into 30×30 pixels , and in 3D each cubic element was divided into 70×70×70 voxels . These cubes can be nonstaggered , staggered in one direction , or staggered in two directions in 3D space . Within each of these regions , a “target volume” is defined . Each region is then populated with a cubic obstacle that occupies 80% of the region and has its center coordinate randomly placed in the target volume . The approach described here can be generalized to allow more variation in cell shape and size by permitting the placement of both cubical and cuboidal cellular obstacles . The resulting geometric representation of the brain microstructure ( in the nonstaggered case ) can be seen in Figure 1 . In 3D , we are modeling approximately a 2 . 16×10−3 mm3 volume of brain tissue . The model will be analyzed using periodic boundary conditions to minimize boundary effects . Properties P1–P4 of brain tissue are satisfied by the proposed two-phase model . In particular , the model is composed of densely packed mostly convex cells and the ECS occupies 20% of space . While each obstacle placed into the system has a fixed size , the placement of some obstacles results in the formation of elongated cells , some of which are oddly shaped and not convex . This feature is desirable , as not all brain cells are convex and , as property P2 states , brain cells can vary in size . Each cellular object is surrounded by ECS , and the ECS surrounding each cell is not uniform in width . Finally , because obstacles placed in each region can touch neighboring obstacles , a small number of dead-end regions naturally arise in this geometry . It is essential that we evaluate the limitations of two-phase models to justify the development of a three-phase model . Previously developed two-phase models either overestimate the brain's effective diffusivity [9] , or achieve the diffusivity by introducing a large number of cellular concavities [6] , [12] , [13] . Using the discrete first-passage-time algorithm ( see Methods section ) , we wanted to determine if the proposed two-phase model has a diffusion coefficient comparable to that measured in brain tissue . The model was analyzed in 2D and 3D to determine the impact dimensionality has on the results . We found that the 2D model is a successful representation of the brain microstructure: the geometry proposed is subject to the same set of biological constraints as brain tissue and has similar diffusion properties ( data not shown ) . While the 2D results are promising , the brain is a 3D structure . For this reason , we next applied the first-passage-time technique to test if the 3D geometry successfully reconstructs the brain microstructure ( Table 1 ) . While all 3D media created satisfy the two-phase HS bound , the lowest diffusivity obtained ( D* = 0 . 63 in both the staggered and nonstaggered case ) is significantly larger than the effective diffusivity measured in brain tissue ( D* = 0 . 4 ) . Even though the proposed two-phase model has a lower diffusivity than models of uniformly spaced convex cells , the 3D model , unlike its 2D analog , does not have a low enough diffusivity to be a valid representation of the brain microstructure . In order to develop a biologically constrained model of the brain microstructure with the expected diffusion properties , we return to the experimental observation that the ECS is not a homogeneous solution , but is instead a heterogeneous composite , and the largest components in this composite are the macromolecules of the extracellular matrix [2] . By definition , the extracellular matrix is an intricate network of macromolecules that assemble into an organized meshwork in close association with the surface cell that produces them ( Figure 2A ) [24] . We propose that the ECM be treated as an independent third phase of the brain microstructure . While the limitations of two-phase models discussed in the previous section lead us to deviate from the conventional two-phase modeling approach , it is important to reiterate that the novelty of this work is the direct inclusion of ECS heterogeneity into a model of the brain microstructure . Although this model is thus not proposing a new biological mechanism , it is certainly guided by experimental evidence that suggests that a three-phase model fits the task at hand . Firstly it has been shown that molecular changes in ECM content occur during normal and pathological processes that are characterized by altered brain diffusion properties [2] . This observation provides evidence that there is a correlation between changes in ECM content and the diffusion of small tracers in the brain , although no causative relation has been proven . Secondly , it has been speculated that the transport of positively charged molecules ( such as the tracers used in RTI experiments ) is hindered by the negative charge associated with the ECM [11] , lending further support to the theory that the ECM does impact the diffusion of small ions in the brain . Given the theoretical evidence presented against conventional two-phase models and the biological evidence which suggests that the ECM may regulate brain diffusion properties , we turned the two-phase model that obeys properties P1–P4 of brain tissue into a three-phase model by introducing the ECM ( the third phase ) into the ECS . Introducing the ECM into the model necessitates some a priori knowledge on the concentration , diffusion properties and precise structure of the ECM in the brain ECS . The unavailability of this information [11] necessitated a minimalistic modeling approach . Given the aforementioned definition of the ECM , we can envision the ECM as a low volume fraction mesh-like network that surrounds each cell in the brain . Thus , from a modeling perspective , a logical minimalistic first assumption is that the ECM forms a “shell” around each cell ( Figure 2B ) . This shell can act as either a barrier to diffusion , or more likely , can act to slow down diffusion near cell boundaries by trapping diffusing particles in the ECM mesh . Since it is unclear how to approach modeling this mesh-like structure and its altered diffusion properties with any accuracy , we will consider a first-order approximation to this situation . We propose that the ECM can be modeled as having the net effect of excluding a diffusing ion from some volume about the cell . We emphasize that volume exclusion is a net effect of the ECM , because it is more likely that the ECM reduces the diffusion coefficient in the area surrounding cells rather than excluding diffusion all together . If there were no other macromolecules in the ECS , this shell would just be a first-order approximation of the ECM mesh . However , there are other macromolecules that float around the ECS that are not formed in close association with a cell . If we want to also consider the effects of these molecules in our model , we need a computational technique that can predict the average influence of all of these molecules; that is , both those that are in close association with a cell , such as the ECM , and other freely suspended molecules . A technique that allows us to treat the ECM as a mesh-like shell around each cell while also accounting for the effects of those ECS molecules that are not associated with a cell is to use a finite-sized diffusing particle in our first-passage-time simulations . To explain why this implicit representation of the ECM plus other ECS molecules is reasonable , consider what happens when we only consider the ECM without any other molecules in the ECS . When we allow a diffusing tracer to take on a finite size , the tracer is excluded from a larger volume fraction than dictated by cell size , and this exclusion volume is nothing more than the “shell” we defined earlier to represent the ECM . Of course , this analogy only applies if the shell fully inhibits the diffusion of small ions , which is unexpected . Thus , if we were ignoring the effects of other ECS molecules and we just focused on the ECM , this approach gives us a first-order approximation on the net effect the ECM has on diffusion . We do not claim that the shell is of the proper concentration or is modeled with the correct diffusion coefficient , just that it has the same effect as the mesh-like network with the correct volume fraction and diffusivity . Since the ECM molecules are not the only compounds found in the ECS , there is no reason this first-order approximation has to only account for the effects of the ECM . The first-order approximation we propose here actually models the net effect of both the ECM and other molecules that are found free-floating in the ECS . In our first-order approximation , one key parameter , the radius of the diffusing particle used in simulations , will measure the exclusion-volume effects caused by the ECM plus other ECS molecules [25] . The larger this parameter , the more hindrance a particle encounters or the more time a particle is trapped as it diffuses through the ECS . It is important to note here that the finite-sized diffusing tracer is used to implicitly represent the presence of the ECM plus other ECS molecules; it is not related to the size of the actual tracer used in RTI experiments ! In order to quantify the effects that this hard-shell approximation of the ECM has on the proposed microstructural model , we have studied how both the average gap width and the fraction of concave cells changes as a function of the diffusing particle radius , and these results are summarized in Figure 3A . We have found that , as expected , the average gap width in the model decreases as the particle radius increases . More importantly , we have quantified how the fraction of concave cells increases in our model as a function of particle radius . As seen in Figure 3A , we have found that slightly less than 15% of the cells in the two-phase model ( particle radius equals zero ) are concave . If the diffusing particle is allowed to have a radius of 1 voxel ( which corresponds to 0 . 31 µm in our model ) , the percent of concave cells increases to 23% . Further increasing the radius to 2 voxels increases the percent of concave cells to 63% . Thus , our first-order approximation of the ECM has the net effect of decreasing the average gap width in the model and increasing the percent of concave cells . The 3D first-passage-time algorithm was applied to the proposed three-phase model ( all cubes; nonstaggered case ) , using a finite-sized diffusing particle to represent ECS heterogeneity ( the presence of the ECM plus other ECS molecules ) . Simulations were conducted for various values of the diffusing particle radius to probe the effects of a wide concentration of ECS molecules ( Figure 3B ) . When the radius of the diffusing particle is approximately 0 . 255 µm ( which is equivalent to 83% of the length of a voxel element in our model ) , the three-phase medium achieves an effective diffusivity comparable to that observed in brain tissue . At this particle radius , the net effect of the ECM plus other ECS molecules is to decrease the fraction of space available to the diffusing tracer from 0 . 2 to 0 . 140 . Importantly , this does not mean that the ECM is a hard shell that occupies 30% of pore space . Instead , it does mean that the hindrance to diffusion caused by both the ECM and other free-floating ECS molecules must have the same effect on the diffusion of small ions that is had by restricting a diffusing particle from 30% of pore space . When the particle radius is 0 . 255 µm , the average gap width in the model is 1 . 47 µm . This width is significantly larger than that reported in property P3 , and this discrepancy will be explored in the Discussion and Conclusions section . Further , 21 . 5% of the cells in the three-phase model are concave . This percent should be compared to those models that directly incorporate concavities by punching dead-ends into convex cells [6] , [13] . In these models , 100% of the cells must contain concavities to achieve the diffusivity measured in brain tissue . Further , for one of the proposed models with a lower diffusion coefficient , the effect that must be exerted by the ECM and other ECS molecules would be even smaller . For example , if we consider the model that includes short cuboids and hence has more variation is cell shape and size , we find that the fraction of space available to the diffusing tracer decreases from 0 . 2 to 0 . 145 , meaning that the hindrance imposed by the ECM and other ECS molecules must have the same effect on the diffusion of small ions that is had by restricting diffusion from 27 . 5% of pore space . Even the 27 . 5% proposed here is an upper bound , as will be explored in the Discussion section . Further , it is important to note that the net effects predicted by the model do not allow us to tease out the properties of the ECM , such as concentration and diffusivity , that are responsible for the decrease in the diffusion coefficient . With more biological data , the model can be moved from this first-order approximation to a more realistic representation of the third phase . Even with this first-order approximation , these results strongly suggest that ECS heterogeneity is an important contributor to the low effective diffusivity of brain tissue . Our simulation results are compared to the following two-point bounds for 3D , three-phase isotropic media: ( 2 ) where Dmax and Dmin denote the largest and smallest diffusivities amongst the three phases , respectively [3] . For the example at hand , the three-phase bounds can be greatly simplified . If we let phase 2 be the ICS , then we know that φ2 = 0 . 8 and Dmin = D2 = 0 cm2 s−1 . If phase 3 is the ECM , we know that φ3 = 0 . 2−φ1 . Moreover , since we are assuming that the ECM hinders diffusion relative to free diffusion in the ECS , we have that Dmax = D1 and that D3 = aD1 , where 0≤a≤1 . For this situation , the bound in ( 2 ) becomes ( 3 ) The upper bound given in Equation 3 is maximized ( for any 0≤φ1≤0 . 2 ) at a = 1 , that is , when the ECM ( plus other ECS molecules ) phase behaves exactly as the ECS and does not act as a hindrance to diffusion . For the special case of a = 1 , the bound reduces to the two-phase HS upper bound evaluated at φ1 = 0 . 2: 0≤D*≤0 . 71 , i . e . , the diffusion coefficient of the proposed three-phase medium obeys the same upper bound as any isotropic two-phase medium with the same porosity .
We have demonstrated that including the extracellular matrix in a novel brain tissue model composed of nonuniformly spaced mostly convex cells can give the decrease in the diffusion coefficient necessary for the proposed model to conform to the macroscopic properties of brain tissue . This is strong evidence that realistic microstructural models of the brain must account for the effects of the ECM . While the role of dead-ends is minimized in our model , our work does not contradict models that emphasize the importance of dead-ends . Instead , each model probably offers part of the picture , and it is probably some combination of both models that best represents the actual structure of the brain . The contribution of the present work is to give a novel way to consider the effects of the ECM and other ECS components , as well as to propose a novel packing procedure for cells in the brain . It has been demonstrated rigorously that diffusion properties of a heterogeneous medium can be linked to seemingly different properties of the same medium , including the elastic moduli [29] , [30] , electrical conductivity [31] , and fluid permeability [32] , [33] . In a future work , we will examine such cross-property relations [3] for model brain microstructures .
Consider a Brownian particle diffusing in a two-phase digitized composite material consisting of pixels ( 2D ) or voxels ( 3D ) which either have finite diffusivity D1>0 ( representing the ECS ) or D2 = 0 ( representing the ICS ) . In order to cope with the computational limitations of modeling a very large region of brain tissue , the algorithm employs periodic boundary conditions . For digitized images , the natural first-passage region is determined by the shape of a pixel and voxel; that is , squares are used in 2D and cubes are used in 3D [39] . In order to simulate the diffusion of a particle in a 3D digitized composite , the following set of rules [39] is applied: The details of the analogous 2D algorithm can be found elsewhere [39] . In any dimension d , the dimensionless effective diffusion tensor of the medium , is given by [38] ( 8 ) where the values of Xw ( displacement in the wth direction ) and τ ( time taken to hit the surface of a bounding region ) are calculated using the first-passage-time algorithm described above . The summation over the subscript k denotes Brownian paths in phase 1 , over l denotes paths in phase 2 , and over m denotes paths at the two-phase interface . | The goal of the present work is to develop a biologically constrained three-dimensional model of the brain microstructure . This is an important task because the brain's three-dimensional microstructure cannot be directly visualized , yet a knowledge of its structure is essential for understanding normal brain functioning . We first explore the shortcomings of the conventional modeling approach that treats brain tissue as a two-phase material . These models either do not preserve realistic features of brain tissue or preserve these properties while overestimating the brain's effective diffusivity , an average measure of the underlying microstructure . We thus developed a biologically constrained two-phase model that , upon analysis , achieves a lower diffusion coefficient than other constrained models yet proves to not have a low enough diffusion coefficient to be a valid representation of the brain microstructure . We then show that if the extracellular matrix is incorporated as a third phase in this model , then the reduction in the diffusion coefficient achieved allows the proposed model to be a valid representation of the brain microstructure . Using this model , we can test the impact that microstructural changes have on the transport of nutrients and signaling molecules in the brain . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"biophysics/theory",
"and",
"simulation",
"computational",
"biology"
] | 2008 | A Novel Three-Phase Model of Brain Tissue Microstructure |
NF-κB plays a central role in modulating innate immune responses to bacterial infections . Therefore , many bacterial pathogens deploy multiple mechanisms to counteract NF-κB activation . The invasion of and subsequent replication of Shigella within epithelial cells is recognized by various pathogen recognition receptors as pathogen-associated molecular patterns . These receptors trigger innate defense mechanisms via the activation of the NF-κB signaling pathway . Here , we show the inhibition of the NF-κB activation by the delivery of the IpaH E3 ubiquitin ligase family member IpaH0722 using Shigella's type III secretion system . IpaH0722 dampens the acute inflammatory response by preferentially inhibiting the PKC-mediated activation of NF-κB by ubiquitinating TRAF2 , a molecule downstream of PKC , and by promoting its proteasome-dependent degradation .
The intestinal epithelium deploys multiple sensors and defense systems against microbial intrusion . Microbial components and infection-associated cellular damage are recognized as pathogen-associated molecular patterns ( PAMPs ) and as danger-associated molecular patterns ( DAMPs ) , respectively . Pattern recognition receptors ( PRRs ) recognize PAMPs and DAMPs , thus activating the immune system to clear the bacteria and initiate the repair of the injured epithelial lining [1] , [2] . Nevertheless , many bacterial pathogens such as Shigella , Salmonella , Yersinia , enteropathogenic Escherichia coli ( EPEC ) , and enterohemorrhagic E . coli ( EHEC ) , are able to efficiently colonize the intestinal epithelium by utilizing highly evolved mechanisms to counteract host innate defense mechanisms [3] . Previous studies have reported that enteric bacterial pathogens possess distinctive mechanisms to attenuate host inflammatory responses , which is a prerequisite for promoting intracellular and extracellular bacterial survival [4] , [5] . For example , during the invasion of the intestinal epithelium by Shigella , the Toll-like receptors ( TLRs ) and Nod-like receptors ( NLRs ) are the PRRs that recognize PAMPs and DAMPs [6]–[9] . These receptors stimulate host inflammatory signaling pathways , including nuclear factor κB ( NF-κB ) and mitogen activated protein kinases ( MAPK ) , which culminate in the expression of proinflammatory chemokines and cytokines [6]–[9] . Shigella counteract innate immune responses by delivering effector molecules using its type III secretion system ( T3SS ) [8]–[10] . The targeting molecules and mechanisms of inhibition of the NF-κB pathway are specific to each of the effectors , and Shigella are known to deliver OspG [11] , OspI [12] , OspZ [13] , and IpaH9 . 8 [14]–[17] , to efficiently attenuate NF-κB activation thus allowing replication within the intestinal epithelium [8] , [9] . IpaH9 . 8 belongs to IpaH E3 ubiquitin ligase family [16] and it is also called novel E3 ligase ( NEL ) [18] . The ipaH genes , which are originally identified in Shigella large virulence plasmid , are conserved by Gram-negative bacterial pathogens , including Shigella , Salmonella , Yersinia , Edwardshiella ictluri , Bradyrhizobium japonica , Rhisobium sp . strain NGR234 , Pseudomonas putida , P . entomophila , P . fluorescens , and P . syringae [16] , [19] , [20] . The IpaH family proteins share common structural and functional properties; they contain an N-terminus leucine rich repeat ( LRR ) and a highly conserved C-terminal region ( CTR ) [21] , [22] . Within the conserved CTR there is a Cys residue that is critical for its E3 ubiquitin ligase activity [16] . Each of the IpaH family effectors , including IpaH9 . 8 ( Shigella ) , SlrP ( Salmonella ) , SspH1 ( Salmonella ) , SspH2 ( Salmonella ) , YopM ( Yersinia ) , Y4fR ( Rizobium ) , and BIpM ( Yersinia ) , has been shown to target specific host proteins in a variety of cell types; some of them act as effectors that attenuate host inflammatory responses , while others modulate host defense responses in plants [17] , [23]–[25] . The existence of multiple effectors with E3 ligase activity suggests that a divergent array of E3 ligases is required for promoting bacterial infection and antagonizing host innate defense responses . The Shigella flexneri strain , YSH6000 , has three ipaH genes ( ipaH9 . 8 , ipaH7 . 8 , and ipaH4 . 5 ) on its large virulence plasmid and seven ipaH genes on its chromosome ( ipaH0722 , ipaH0887 , ipaH1383 , ipaH1880 , ipaH2022 , ipaH2202 , and ipaH2610 ) [26]–[28] . We previously showed that all of these IpaH effector proteins are secreted via the T3SS [28] , [29] , suggesting they have the potential to act as E3 ubiquitin ligase effectors during infection , although their exact roles and host targets during infection , with the exception of IpaH9 . 8 , remain largely unclear [17] . In the mouse lung infection model , we showed that mice infected with the ΔipaH-chromosome mutant , which lacks all seven chromosomal ipaH genes , had a more severe inflammatory response and less colonizing bacteria compared to the WT strain [28] , suggesting that modulation of the host inflammatory response by chromosomal IpaH proteins promotes bacterial infection . Therefore , to gain further insight into the strategies employed by Shigella to counteract host innate defence systems during infection , we sought to characterize the chromosomally encoded IpaH effectors . We found that IpaH0722 targets TRAF2 and plays an important role in dampening the PKC–NF-κB pathway , in response to membrane damage generated during Shigella invasion of epithelial cells .
The infection of mice using the ΔipaH-chromosome mutant , which lacks all of the chromosomal ipaH genes , leads to an increased production of the NF-κB responsive gene MIP-2 [28] . In this current study , we used HeLa cells to determine the effect of the ΔipaH-chromosome mutant on the NF-κB activation by measuring the kinetics of IκBα degradation . HeLa cells were infected with YSH6000 ( S . flexneri WT ) or ΔipaH-chromosome mutant , and then whole cell lysates were harvested at 20 , 40 , 60 , and 80 min post-infection for the detection of IκBα . The degradation rate of IκBα in cells infected with the ΔipaH-chromosome mutant was higher than that of WT , indicating that one or more of the chromosomal IpaH proteins contributed to the dampening of IκBα degradation ( Fig . 1A , left ) . To identify the IpaH proteins that were involved in suppressing NF-κB activation , we generated deletions mutants for each of the seven chromosomal ipaH genes . IpaH0722 was critical for the inhibition of NF-κB activation . The degradation rate of IκBα was higher in ΔipaH0722 infection compared to WT infection ( Fig . 1A , right ) . IpaH0722 , which corresponds to ORF SF0722 of Shigella flexneri 2a Sf301 strain , has a Cys residue in its C-terminal region that is required for E3 ubiquitin ligase activity [16] , [28] , [29] . In Cos-7 and HeLa cells , ectopically-expressed IpaH0722 localized to the cell membrane ( Fig . 1B and S1A ) . A mouse lung infection model was used to evaluate the role of IpaH0722 in the pathogenesis of Shigella . The survival rates of mice infected with the ΔipaH0722 mutant Shigella were increased compared to mice infected with WT Shigella ( Fig . 1C ) . Next , to investigate whether the Cys379 residue in the IpaH0722 C-terminal region and its E3 ubiquitin ligase activity contributed to the pathogenesis of Shigella , we substituted Cys379 residue ( IpaH0722CA ) . When 293T cells expressing IpaH0722 or IpaH0722CA were infected with Shigella , the activation of NF-κB was decreased in the presence of IpaH0722 but not IpaH0722CA ( Fig . 1D ) . In addition , to confirm the importance of the E3 ubiquitin ligase activity of IpaH0722 in Shigella infection , HeLa cells were infected with Shigella ΔipaH0722 harboring ipaH0722 ( ΔipaH0722/ipaH0722 ) or ipaH0722CA ( ΔipaH0722/ipaH0722CA ) . In cells infected with the WT strain or ΔipaH0722/ipaH0722 strain , the degradation rate of IκBα was reduced at 60 and 80 min after infection compared to the ΔipaH0722 or ΔipaH0722/ipaH0722CA strains ( Fig . 1E ) . It is important to note that IpaH0722 had no effect on Elk-1 and JNK activation ( Fig . 1D and E ) . We speculated that IpaH0722 plays a role in the inhibition of Shigella-induced NF-κB activation in an E3 ubiquitin ligase-dependent manner . NF-κB activity can be stimulated by multiple signaling pathways that are triggered by various receptors in response to exogenous and endogenous stimuli , including TNFα-TNFR , TLRs , and NLRs [30] . To characterize IpaH0722-dependent dampening of NF-κB activation , we measured the levels of NF-κB activation in IpaH0722- or IpaH0722CA-expressing 293T cells that were stimulated with PMA , TNF-α , lipopolysacharide ( LPS ) , or IL-1βusing NF-κB-luciferase reporter assays . IpaH0722 markedly impaired PMA-dependent , but not TNF-α- , LPS- , and IL-1β-dependent NF-κB activation in an E3 ubiquitin ligase activity-dependent manner ( Fig . 2A ) . To confirm the specificity of IpaH0722-mediated induction of NF-κB activation , we investigated the expression of IL-8 in 293T cells stimulated with PMA using an IL-8 reporter assay . Consistent with the NF-κB findings , IpaH0722 preferentially inhibited the PMA-dependent IL-8 expression ( Fig . 2B ) . Moreover , the preferential targeting of PMA- but not TNF-α-induced NF-κB activation was also demonstrated by ELISA assays ( Fig . 2C ) . Although PMA was also known to induce MAPK activation , IpaH0722 failed to inhibit PMA-dependent AP-1 and Elk-1 activation ( Fig . 2D ) . Taken together these results clearly indicated that IpaH0722 preferentially inhibited PMA-induced NF-κB activation by targeting factors that modulated the NF-κB signaling pathway . It has recently reported that Salmonella SspH2 , an IpaH cognate protein , localizes to the host cell membrane through its modification by cellular palmitoyl transferases . SspH2 has a putative S-palmitoylation motif in its N-terminal and undergoes modification by host cell palmitoyltransferases , resulting in its localization to the host cell membrane [31] . Since we found a putative S-palmitoylated motif in IpaH0722 N-terminal portion , we investigated whether this motif played a role in IpaH0722 localization to the cell membrane . In Cos-7 and HeLa cells , ectopically-expressed IpaH0722 localized to host cell membrane ( Fig . 1B and S1A ) . IpaH0722 has two Cys residues in its palmitoylation motif ( Cys14 and Cys18 ) , we generated constructs in which the Cys residues were substituted with Ser singly and in tandem ( IpaH0722-C14S , -C18S , and -C14S/C18S ) . These constructs were ectopically expressed in HeLa cells to evaluate the subcellular localization of IpaH0722 . IpaH0722 WT and IpaH0722CA localized to the cell membrane , whereas IpaH0722-C14S , -C18S , and -C14S/C18S accumulated within the cytosol . Therefore , palmitoylation at Cys14 and Cys18 residues of IpaH0722 are critical for its localization to the plasma membrane ( Fig . S1A ) . We next investigated whether membrane localization of IpaH0722 was required for its inhibition of NF-κB . The various IpaH0722 constructs were ectopically expressed in 293T cells , which were stimulated with PMA , NF-κB activity was assessed using NF-κB-luciferase reporter assays . IpaH0722 , but not IpaH0722CA , inhibited NF-κB activation ( Fig . S1B ) . Moreover , since the disruption of the putative palmitoylation sites in IpaH0722 mutants did not alter their ability to reduce PMA-dependent NF-κB activation , we presumed that the localization of IpaH0722 to the plasma membrane was not required for NF-κB inhibition in our experimental setting ( Fig . S1B ) . PMA mimics the role of diacylglycerol ( DAG ) in the activation of the protein kinase C ( PKC ) -NF-κB pathway [32] . Therefore , we hypothesized that IpaH0722 selectively targeted the DAG–PKC–NF-κB pathway , which was likely due to the membrane localization of IpaH0722 once it was secreted by invading Shigella into epithelial cells . We sought to determine which of the PKC isoforms were involved in the activation of NF-κB . The PKC family of proteins is composed of 12 isoforms that act as lipid-activated Ser/Thr kinases [33] . The PKC family can be divided into four functional protein classes: conventional PKC ( PKCα , β , and γ ) , novel PKC ( PKCδ , ε , η , and θ ) , atypical PKC ( PKCζ , λ/ι ) and PKC related kinase ( PKCμ ) . Conventional and novel PKCs have a DAG binding domain . Conventional PKCs require DAG and Ca2+ for their activation , whereas novel PKCs require only DAG for their activation . Atypical PKCs do not depend on DAG and Ca2+ for their activation . Since the activity of PKC is regulated by phosphorylation and its recruitment to the cell membrane , we investigated the levels of phosphorylated PKC during Shigella infection of HeLa cells . HeLa cells were infected with Shigella WT and cell lysates were harvested at 10 , 20 , 40 , and 60 min after infection for the analysis of phosphorylated PKC by immunoblotting . As shown in Fig . 3A , Shigella infection augmented the phosphorylation of conventional or novel PKCs , such as PKCδ , at 10 and 20 min , and PKCμ at 20 , 40 , and 60 min post-infection . To investigate whether PKC activation triggers NF-κB activation in Shigella infection , we exploited dominant negative forms ( DN ) of PKC and siRNA that targeted PKC . To investigate the role of PKC in the activation of NF-κB during Shigella infection , DN-PKCα and DN-PKCδ , ε , θ , were ectopically expressed in 293T cells . The DN-PKCδ , ε , θ , but not DN-PKCα , significantly decreased Shigella-induced NF-κB activation ( Fig . 3B ) suggesting that the PKCδ-NF-κB pathway plays a critical role during Shigella infection . Similarly , siRNA-mediated knockdown of PKCδ in 293T cells infected with Shigella decreased NF-κB activity to less than half of the control levels ( Fig . 3C ) . The invasion of epithelial by Shigella produces membrane ruffles through the remodeling of the F-actin cytoskeleton [34]–[36] . Immediately following bacterial invasion , the bacteria are rapidly surrounded by a vacuolar membrane that the bacteria disrupt to facilitate dissemination into the cytoplasm by inducing actin polymerization [34]–[36] . To further understand the mechanism of PKC activation by Shigella invasion , we examined PKC phosphorylation during Shigella infection . First we confirmed the importance of Shigella invasiveness . When HeLa cells were infected with Shigella WT or the T3SS deficient mutant S325 , Shigella WT infection triggered the phosphorylation of PKCδ , however the S325 mutant did not ( Fig . 3D ) . We next sought to determine whether vacuolar membrane disruption potentiated DAG–PKC–NF-κB signaling . We previously created an ΔipaB/inv mutant by introducing Yersinia invasin gene into the ΔipaB mutant [37] . The resulting ΔipaB/inv mutant was internalized into the endocytic vacuole and it was unable to disrupt the vacuolar membrane for dissemination into the cytoplasm ( Fig . 3D ) . When HeLa cells were infected with Shigella WT or the ΔipaB/inv mutant , Shigella WT infection triggered the phosphorylation of PKCδ , however the ΔipaB/inv mutant did not ( Fig . 3D ) . In contrast , both WT and ΔvirG mutant , which is unable to support intra- and inter-cellular movement , induced PKCδ phosphorylation ( Fig . 3D ) . To further confirm the importance of vacuolar membrane disruption , we investigated PKC phosphorylation during Salmonella infection using a strain that triggers membrane ruffling but remain sequestered in the phagosome . When HeLa cells were infected with Shigella WT or the Salmonella WT , Shigella WT but not Salmonella WT infection triggered the phosphorylation of PKCδ ( Fig . 3D ) . These data supported the notion that membrane rupture by Shigella in infected epithelial cells triggers the activation of the DAG-PKC-NF-κB pathway . Because Shigella invasion of epithelial cells triggers the DAG–PKC–NF-κB signaling , we sought to determine the host factors in the PKC–NF-κB signaling pathway that were targeted by IpaH0722 E3 ubiquitin ligase . First , we determined which of the steps during the activation of PKC–NF-κB was targeted by IpaH0722 . The NF-κB activity induced by PKC signaling factors was measured in 293T cells with ectopic expression of IpaH0722 or IpaH0722CA . The transiently transfected cells were stimulated with a subset of constitutively active PKC isoforms ( PKCα , -δ , -ε , and θ ) and NF-κB activity was determined . IpaH0722 , but not IpaH0722CA , inhibited all PKC isoforms ( PKCα , -δ , -ε , and -θ ) and NF-κB activation , indicating that IpaH0722 targeted PKC itself or factors that lie directly downstream of PKC ( Fig . 4A ) . In HeLa cells , similar levels of phosphorylated PKCδ were detected in response to Shigella WT and ΔipaH0722 infection ( Fig . 4B ) . Moreover , no differences were detected in PKC phosphorylation in PMA plus ionomycin stimulated 293T cells expressing IpaH0722 or IpaH0722CA ( Fig . 4C ) . In addition , IpaH0722 did not bind to PKC , and had no effect on protein stability of PKC , suggesting that IpaH0722 targets downstream of PKC rather than PKC itself ( Fig . S2 ) . We therefore focused on the CARMA-Bcl10-MALT1 ( CBM ) complex because the phosphorylation of CARMA1 , 2 , and 3 by PKC induces a conformational change that recruits Bcl10-MALT1 to CARMA ( CBM signalosome ) , which is essential for the activation of downstream signaling [38] , [39] . Therefore , we tested the effect of IpaH0722 on CBM complex formation , and found that IpaH0722 failed to inhibit CARMA1- , CARMA2- , CARMA3- , or Bcl10-induced NF-κB activity ( Fig . 4D ) . Indeed , IpaH0722 did not bind to CARMA1- , CARMA2- , CARMA3- , or Bcl10 , and had no effect on their protein stability ( Fig . S3 ) . Furthermore , since IpaH0722 did not affect PKC-mediated CARMA phosphorylation ( Fig . 4E ) , we believe that IpaH0722 inhibited the PKC–NF-κB pathway but not via the CBM complex . We subsequently investigated other NF-κB signaling factors , namely TRAF2 , TRAF5 , TRAF6 , NIK , IKKα , IKKβ , or p65 as potential IpaH0722 targets [40] . To this end , 293T cells expressing NF-κB-luciferase reporter gene together with TRAF2 , TRAF5 , TRAF6 , NIK , IKKα , IKKβ , or p65 were transfected with a vector encoding IpaH0722 or IpaH0722CA and luciferase activity was measured ( Fig . 5A ) . IpaH0722 , but not IpaH0722CA , inhibited NF-κB activity when the cells that expressed TRAF2 suggesting that TRAF2 is a target for IpaH0722 E3 ubiquitin ligase ( Fig . 5A ) . In previous reports , PKC-mediated phosphorylation of TRAF2 was a prerequisite for NF-κB activation in epithelial cells [41] , [42] . PKCδ and PKCε phosphorylated TRAF2 at residue Thr117 resulting in IKK recruitment and ultimately NF-κB activation [41] , [42] . We therefore used siRNA-mediated TRAF2 knockdown cells to measure the effect of TRAF2 on activation of the PKC-NF-κB pathway . The results showed that TRAF2 knockdown decreased NF-κB activation in response to PMA stimulation and Shigella infection ( Fig . S4A ) . To further confirm the effect of TRAF2 knockdown on NF-κB activation we utilized Traf2-knockout mouse embryonic fibroblasts ( MEF ) [43] . We transduced Traf2 into the Traf2-knockout MEFs by retrovirus infection to generate stably expressing Traf2 cell lines . The degradation of IκBα in Traf2 stably expressing MEF cells and Traf2-knockout MEF cells during Shigella infection was determined . As shown in Fig . S4B , the degradation rate of IκBα in Traf2 stably expressing MEFs was higher than that of Traf2-knockout MEFs at 20 and 40 min after infection suggesting that TRAF2 plays a role in Shigella-induced NF-κB activation . Having shown that IpaH0722 targets TRAF2 , we performed immunoprecipitation assays to determine whether IpaH0722 interacted with TRAF2 . Cell lysates were harvested from 293T cells that expressed IpaH0722CA and TRAF2 , TRAF5 , TRAF6 , NIK , IKKα , IKKβ , or p65 . IpaH0722 precipitated with TRAF2 , but not TRAF5 , TRAF6 , NIK , IKKα , IKKβ , or p65 ( Fig . 5B ) . The interaction between IpaH0722 and TRAF2 was further confirmed by GST-pull down assay . Using whole cell lysates of HeLa and GST-IpaH0722 , we were able to pull down endogenous TRAF2 , but not TRAF6 ( Fig . 5C ) . In previous reports , structural analysis of IpaH proteins using IpaH1 . 4 or IpaH3 revealed that the LRR region of the IpaH family is required for substrate recognition , while the CTR region is essential for its E3 ubiquitin ligase activity [44] , [45] . IpaH proteins share characteristic domains: an N-terminal 60–70 amino acid stretch , a LRR region , and an intervening region flanked by the LRR and the conserved CTR ( Fig . S5A ) . We constructed a series of GST-tagged IpaH0722 truncations and performed GST-pull down assays using HeLa whole cell lysates . IpaH0722 truncations that contained the LRR regions bound to TRAF2; however , IpaH0722 truncations that contained the N-terminal , intervening region , or CTR failed to bind to TRAF2 ( Fig . S5B ) . Since IpaH0722 inhibited NF-κB activation in an E3 ubiquitin ligase-dependent manner , we tested whether IpaH0722 ubiquitinated TRAF2 using an in vitro ubiquitination assay . TRAF2 , IpaH0722 , or IpaH0722CA ( purified from E . coli ) were combined with E1 , ATP , UbcH5b ( an E2 ubiqitin conjugating enzyme ) in assay medium and subjected to immunoblotting . TRAF2 was ubiquitinated by IpaH0722 but not IpaH0722CA ( Fig . 6A ) . To confirm the fate of ubiquitinated TRAF2 , we measured the stability of TRAF2 in 293T cells ectopically expressing IpaH0722 or IpaH0722CA and treated with cycloheximide ( CHX; a protein synthesis inhibitor ) for 2 , 4 , and 6 h . TRAF2 degradation was faster in IpaH0722 expressing cells compared to IpaH0722CA expressing cells , which confirmed that IpaH0722 targeted TRAF2 for ubiquitin-mediated protein degradation ( Fig . 6B ) . We also measured TRAF2 degradation in 293T cells that were co-transfected with TRAF2 plus IpaH0722 or IpaH0722CA , and treated with MG132 ( a proteasome inhibitor ) or E64D plus pepstain A ( lysosome inhibitors ) . As shown in Fig . 6C , MG132 treatment prevented TRAF2 degradation in the presence of IpaH0722 , whereas E64D plus pepstatin A treatment did not prevent TRAF2 degradation . These data confirm that IpaH0722 targets TRAF2 for ubiquitination and proteasomal degradation . Previous studies indicated that TRAF5 compensates for TRAF2 function [43] , so we exploited Traf2/Traf5 double knockout MEFs to clarify the requirement of TRAF2 on NF-κB activation . We introduced Traf2 into the Traf2/Traf5 double knockout MEFs by retrovirus transduction . Under puromycin selection , we obtained stable TRAF2-expressing Traf2/Traf5 double knockout MEFs . Traf2/Traf5 double knockout MEFs and stable Traf2-expressing Traf2/Traf5 double knockout MEFs were infected with Shigella WT or ΔipaH0722 , and IκBα degradation was measured over time . The degradation rate of IκBα in ΔipaH0722-infected stable TRAF2-expressing Traf2/Traf5 double knockout MEFs was higher compared to WT Shigella infection ( Fig . 6D ) . In contrast , the degradation rates of IκBα in WT and ΔipaH0722-infected double knockout MEFs were almost equal ( Fig . 6D ) . To evaluate the importance of the in vivo role of E3 ligase activity of IpaH0722 for Shigella virulence , we infected mice with a sublethal dose of Shigella WT , ΔipaH0722 , ΔipaH0722/0722 , or ΔipaH0722/0722CA . Mice infected with ΔipaH0722 or ΔipaH0722/0722CA mutants showed significantly reduced bacterial colonization when compared to that of WT or ΔipaH0722/0722 infection ( Fig . S6A ) . Furthermore , histopathology showed that inflammatory responses , such as suppurative masses , neutrophil infiltration , and macrophage infiltration , were significantly higher in mice infected with ΔipaH0722 or ΔipaH0722/0722CA mutants when compared to that of WT or ΔipaH0722/0722 infections ( Fig . S6B ) . These results supported our notion that the interaction between IpaH0722 and TRAF2 is an important mechanism for dampening NF-κB activation during Shigella infection of epithelial cells .
In this study we show that IpaH0722 , a chromosomally-encoded IpaH E3 ligase , specifically dampens the PKC-dependent activation of NF-κB in response to Shigella invasion of epithelial cells . We find the IpaH0722 exerts its effect by targeting TRAF2 for proteasomal-dependent degradation ( Fig . 6E ) . The inhibitory activity possessed by IpaH0722 appears to be an important mechanism for the downregulation of the acute inflammatory response by blocking the activation of PKC-NF-κB pathway . This is a critical step since cell membrane rupture is a part of the process of Shigella invasion into epithelial cells , and the recognition of this as a DAMP triggers the recruitment and activation PKC , and ultimately the activation of the innate immune response via PKC-NF-κB pathway . This notion was supported by the experiments using the ΔipaB/inv mutant , which was internalized by the vacuolar membrane but could not rupture the vacuole [36] , we showed that Shigella WT infection , but not ΔipaB/inv mutant infection , caused in the phosphorylation of PKCδ ( Fig . 3E ) . In fact , Dupont et al . recently reported that the vacuolar membrane remnants generated by Shigella invasion into epithelial cells undergo ubiquitination and provided a cue to stimulate autophagy and inflammatory responses [46] . Moreover , Shahnazari et al . showed that the DAG generated in Salmonella containing vacuole ( SCV ) membranes serves as a specific signal to induce autophagy through recruitment of PKC into the SCV [47] . Tattoli et al . showed that vacuolar membrane damage triggers amino acid starvation and the downregulation of mTOR activity , thereby inducing autophagy against intracellular Shigella and Salmonella [48] . Phagosomal membrane disruption by Listeria monocytogenes LLO in macrophages , which is known as a cholesterol-dependent cytolysin , triggers the recruitment of PKC to the membrane and its activation [49] . After disruption of the Listeria phagosome the membrane damage was recognized by PKC . These data suggest that PKC functions as a part of the repair mechanism of host cells , and PKC recruitment by the membrane rupture is recognized as a DAMP by the innate immune system [49] . These studies also imply that the membrane rupture , which is created by bacterial infection , acts as a double-edged sword for the pathogens; on one hand it benefits the bacteria by promoting further dissemination and multiplication , and on the other it hinders the bacteria by acting as a DAMP that stimulates the host innate immune response . In line with this , we identified IpaH0722 as a Shigella effector that plays an important pathogenic role because it counteracts DAMPs-mediated innate immune responses . Recently we reported that another Shigella effector , OspI targets UBC13 and dampens the DAG–CBM–TRAF6–NF-κB pathway , which is triggered by membrane ruffles around the Shigella entry sites into epithelial cells [12] . OspI acts as a glutamine deamidase for UBC13 to convert Q100 to E100 , thus inactivating UBC13 E2 ubiquitin conjugating enzymatic activity that is required for TRAF6 activation [12] . Currently , although the reason underlying why Shigella deliver IpH0722 in addition to OspI during the invasion of epithelial cells remains partly unclear , we believe that it is due to the fact that the two different signaling pathways are required to neutralize the acute inflammatory response and are triggered by distinct bacteria-induced cellular events: DAG-CBM-TRAF6-NF-κB ( membrane ruffles ) and DAG-PKC-TRAF2-NF-κB ( membrane rupture ) [12; this study] . TRAF2 and TRAF5 have been extensively studied in TNFR–NF-κB signaling in immune cells , and it has also been reported that TRAF2 is essential in Nod1– , PKC– , and non-canonical NF-κB signaling in non-myeloid cells [42] , [50] , [51] . In previous studies , PKC-mediated phosphorylation of TRAF2 at Ser11 , Ser55 , and Thr117 was required for prolonged NF-κB activation , but not JNK activation [42] , [52]–[54] . Consistent with these observations , our data showed that treatment of siRNA-mediated knockdown of Traf2 decreased PMA-induced NF-κB activation , suggesting that TRAF2 is involved in PKC–NF-κB signaling ( Fig . S4 ) . According to Hasegawa et al . , Traf2/Traf5 double knockout MEFs but not TRAF6 knockout MEFs had decreased NF-κB activation upon Nod1-RIP2 stimulation , suggesting that TRAF2 was also important in the Nod1–NF-κB pathway [51] . Since Nod1 recognizes the peptidoglycan that is released from Shigella at the entry site and in the cytoplasm , and plays an important role in Shigella-induced NF-κB activation , IpaH0722 likely inhibits both the PKC–NF-κB pathway and the Nod1–RIP2–TRAF2–NF-κB pathway during Shigella infection [55]–[57] . In fact , we see that IpaH0722 also inhibited Nod1-stimulated NF-κB activation in our reporter assay ( data not shown ) . We have provided evidence that IpaH0722 interacted with TRAF2 via its substrate recognition site , LRR , and modified its ubiquitination ( Fig . 6 and Fig . S5 ) . The ubiquitination of TRAF2 by IpaH0722 led to its proteasomal-dependent degradation and inhibited NF-κB activation ( Fig . 6 ) . Though not yet fully understood , IpaH0722 does not inhibit NF-κB in response to TNF-α; unidentified conformational changes or signaling pathways involving TRAF2 likely occur during Shigella infection or PMA stimulation . Herein , we provide the first evidence that Shigella invasion induced membrane rupture of epithelial cells , which stimulates PKC-NF-κB signaling . Moreover , Shigella used the E3 ubiqitin ligase effector , IpaH0722 , to target TRAF2 for proteasomal-dependent degradation ( Fig . 6E ) . This is the first report IpaH0722 as a non-plasmid-encoded T3SS effector protein in Shigella .
Shigella flexneri strain YSH6000 was used as the wild type and S325 ( mxiA::Tn5 ) was used as a T3SS-deficient negative control . Salmonella typhimurium SB300 strain was used as the wild-type strain [58] . Construction of the non-polar mutant of ipaH-chromosome , ipaH0722 , ipaB , and virG of S . flexneri YSH6000 were carried out as described previously [28] , [37] . The ipaH0722-FLAG or ipaH0722CA-FLAG were cloned into pWKS130 to yield p-ipaH0722-FLAG and p-ipaH0722CA-FLAG . The resultant plasmids were introduced into the ΔipaH0722 strain . The ipaH0722 or ipaH0722CA coding sequences were amplified by PCR and cloned into pCMV-FLAG , pEGFP , pGEX-4T-1 , pcDNA-Myc6 ( 6× Myc ) , and pcDL-SRα-Myc vectors . cDNAs for human Traf2 , Traf5 , Traf6 , Carma1 , Carma2 , Carma3 , Bcl10 , IKKα , IKKβ , p65 , and a series of Pkc mutants were cloned into pCMV-FLAG , pEGFP , pcDNA-Myc6 , pcDL-SRα-Myc , and pcDNA-FLAG vectors . cDNA for mouse Traf2 was cloned into pGEX 4T-1 . Site direct mutagenesis of ipaH0722 or Pkc was performed using the QuickChange site directed mutagenesis kit ( Stratagene ) . The anti-M2 FLAG monoclonal FLAG antibody ( Sigma-Aldrich ) , anti-Myc 9B11 , phospho-PKC ( pan ) ( βII Ser660 ) , phospho-PKD/PKCμ ( Ser744/748 ) , phospho- ( Ser ) PKC substrate , phospho-JNK antibody ( Cell signaling ) , anti-actin ( MILLIPORE ) , anti-PKCδ ( C-20 ) , PKCμ ( C-20 ) , TRAF2 , TRAF6 antibody ( Santa Cruz Biotechnology ) , and anti-IκBα ( BD transduction ) were obtained commercially . The anti-IpaH antibody was described previously [29] . PMA , LPS , ionomycin , E64D , and pepstatin A were obtained from Sigma . IL-1β and TNF-α were obtained from Peprotech . MG132 was obtained from Peptide Inst . HeLa cells were cultured in Eagle's minimal essential medium ( Sigma ) supplemented with 10% fetal calf serum . Cos-7 and 293T cells were cultured in Dulbecco's modified Eagle medium ( Sigma ) supplemented with 10% fetal calf serum . The Traf2-deficient or Traf2/Traf5-deficient MEFs were maintained as described previously [43] . To construct Traf2−/− or Traf2/Traf5−/−cells stably expressing Traf2 , cDNAs encoding these genes were subcloned into pMX-puro retroviral expression vectors . Retroviral supernatants were produced in Plat-E cells . Target cells were transduced with supernatants in the presence of DO-TAP ( Roche ) and then cloned under puromycin selection . HeLa cells were infected with various strains of Shigella at a multiplicity of infection ( moi ) of 100 . In the case of the Shigella strains expressing afimbrial adhesin ( Afa ) , cells were infected at moi of 10 . To adjust the intracellular bacterial number , ΔipaB/inv were set to 3 times moi of WT . Infection was initiated by centrifuging the plate at 700× g for 10 min . After incubation for 20 min at 37°C , the plates were washed three times with PBS , transferred into fresh medium containing gentamicin ( 100 µg/ml ) and kanamycin ( 60 µg/ml ) to kill extracellular bacteria . Salmonella infection was conducted as described previously [58] . After incubation for indicated time , the cells were washed with PBS and harvested into 2× Laemmli's sample buffer for immunoblotting . The density of each band was quantified by measuring the mean intensity using NIH image software version 1 . 63 and the expression levels were normalized to the levels of β-actin . All immunoblotting experiments were repeated multiple times and representative data of similar results are presented . 293T cells were seeded into 24-well plates . After 24 h , the cells were transfected with reporter plasmids ( pNF-κB-luc , pIL-8-luc , pElK-1-luc , or pAP-1-luc ) , Renilla luciferase constructs ( phRL-TK , Promega ) using FuGENE6 transfection reagent ( Roche ) . The IL-8 reporter plasmid was constructed by amplifying the 181 bp human IL-8 promoter sequence from the genomic DNA of HeLa cells , which was then inserted into the pMet-luc reporter plasmid ( Clontech ) . Equal amounts of empty vector were used to control for the transfection process . After 24 h , cells were infected with Shigella WT ( moi = 30 ) or treated with PMA ( 25 nM ) , TNF-α ( 10 ng/ml ) or LPS ( 100 ng/ml ) for 3 h . To investigate LPS stimulation , TLR4 and MD2 vectors were also transfected into the cells . Cell extracts were prepared and reporter activity was determined using the luciferase assay system ( Promega ) . Results are presented as fold-change relative to the activity of uninfected or unstimulated cells . Data are representative of three independent experiments . For GST fusion proteins , E . coli BL21 ( DE3 ) strain harboring pGEX4T-1 derivatives were cultivated in L-broth supplemented with ampicillin ( 50 µg/ml ) 3 h at 30°C . Expression was induced by the addition of 1 mM IPTG and incubation for 3 h at 30°C . Bacteria were disrupted by sonication and lysozyme treatment . Purification of the GST fusion proteins with glutathione-Sepharose 4B ( GE Healthcare ) was performed according to the manufacturer's protocol . GST-IpaH0722 bound to glutathione sepharose 4B beads was mixed with HeLa cell lysates for 2 h at 4°C . After centrifugation , the beads were washed five times with 0 . 5% Triton X-100-PBS and subjected to immunoblotting . The immunoblotting experiments were repeated multiple times and representative data of similar results are presented . 293T cells were transiently transfected using FuGENE 6 ( Roche ) . Cells were washed with PBS and lysed for 30 min at 4°C in lysis buffer containing 150 mM NaCl , 50 mM HEPES pH 7 . 5 , 1 mM EDTA , 0 . 5% NP-40 , and Complete protease inhibitor cocktail ( Roche ) . Lysates were cleared by centrifugation and proteins were immunoprecipitated 2 h with anti-Myc ( 9B11 ) or anti-M2 FLAG and Protein G beads ( Sigma ) at 4°C . Immunoprecipitates were washed five times with lysis buffer and subjected to immunoblotting . The immunoblotting experiments were repeated multiple times and representative data of similar results are presented . In vitro TRAF2 ubiquitination assays were performed in 40 µl reaction mixture containing reaction buffer ( 25 mM Tris-HCl [pH 7 . 5] , 50 mM NaCl , 5 mM ATP , 10 mM MgCl2 , and 0 . 1 mM DTT ) , 1 µg TRAF2 , 0 . 5 µg E1 , 2 µg UbcH5b , and 2 µg ubiquitin purified from E . coli in the presence and absence of GST-IpaH0722 or GST-IpaH0722CA . Reactions were incubated at 37°C for 1 h and stopped by the addition of 5× Laemmli sample buffer . All immunoblotting experiments were repeated multiple times and representative data of similar results are presented . At 24 h after transfection , 293T cells were treated with cycloheximide ( 50 µg/ml ) ( Wako ) for the indicated times and cell lysates were harvested for immunoblotting . The density of each band was quantified by measuring the mean intensity using NIH image software version 1 . 63 . The expression levels were normalized to the levels of β-actin . All immunoblotting experiments were repeated multiple times and representative data of similar results are presented . Human PKCδ- or human TRAF2-specific siRNA were prepared by Sigma as follows: 5′-CGUGUGGACACGCCACAUUAU-3′ and 5′-AAUGUGGCGUGUCCACACGGA-3′ ( PKCδ ) or 5′-GGCCAGUCAACGACAUGAACA-3′ and 5′-UUCAUGUCGUUGACUGGCCUC-3′ ( TRAF2 ) . Cells were transfected using RNAiMax ( Invitrogen ) . siRNA treated cells were utilized after 72 h for further analyses . The pulmonary infection model in the mice was described previously [28] . In brief , five-week-old female Balb/c mice ( CREA Japan ) were housed in the animal facility of the Institute of Medical Science , University of Tokyo , in accordance with University guidelines . S . flexneri were suspended in sterile saline , and 20 µl of the bacterial suspension was administered intranasally at 1×108 cfu ( survival assay ) or 1×107 cfu ( colonized bacterial number assay ) . For histological analysis , mice were sacrificed at 24 h after infection , and their lungs were harvested , fixed in 4% paraformaldehyde in PBS , and frozen in liquid nitrogen for sectioning . The sections were stained with hematoxylin and eosin and examined under a microscope . Viable bacteria in the lung tissue were counted by culturing homogenized tissue for 18 h on LB agar plates . Each data point is the mean of the values for 6 infected mice in each group . All animal experiments were carried out in strict accordance with the University of Tokyo's regulations for Animal Care and Use protocol , which was approved by the Animal Experiment Committee of the Institute of Medical Science , the University of Tokyo ( approval number; PA11-92 ) . The committee acknowledged and accepted both the legal and ethical responsibility for the animals , as specified in the Fundamental Guidelines for Proper Conduct of Animal Experiment and Related Activities in Academic Research Institutions under the jurisdiction of the Ministry of Education , Culture , Sports , Science and Technology , 2006 ( Japan ) . All surgery was performed under carbon dioxide euthanasia , and all efforts were made to minimize suffering . Values are reported as means ± standard deviation ( SD ) or means ± standard error ( SEM ) of data obtained for independent experiments . Statistical analysis was performed Student's t-test or a one-way ANOVA . Kaplan-Meier survivor curves were generated . p-values<0 . 05 were considered significant . | In response to bacterial infection , host cells induce a plethora of innate immune responses to combat the infection . However , many bacterial pathogens have developed sophisticated mechanisms to evade the host's immune system . Because NF-κB is crucial for innate immune responses against bacterial infection , bacterial pathogens deploy multiple countermeasures to inhibit NF-κB activation . The invasion and replication of Shigella within host cells results in cellular damage and the production of bacterial components that trigger NF-κB activation . Here , we show that the Shigella type III secretion system ( T3SS ) effector IpaH0722 , a member of the IpaH E3 ubiquitin ligase family , inhibits NF-κB activation during Shigella infection . IpaH0722 preferentially targets the PKC–NF-κB pathway , which is activated in response to danger signals caused by disruption of the phagosomal membrane during the dissemination of Shigella into the cytoplasm . IpaH0722 inhibits NF-κB activation by targeting TRAF2 , which lies downstream of PKC , for ubiquitination and proteasome-dependent degradation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"bacterial",
"pathogens",
"host-pathogen",
"interaction",
"biology",
"microbiology"
] | 2013 | Shigella IpaH0722 E3 Ubiquitin Ligase Effector Targets TRAF2 to Inhibit PKC–NF-κB Activity in Invaded Epithelial Cells |
In the rabbit model of syphilis , infection phenotypes associated with the Nichols and Chicago strains of Treponema pallidum ( T . pallidum ) , though similar , are not identical . Between these strains , significant differences are found in expression of , and antibody responses to some candidate virulence factors , suggesting the existence of functional genetic differences between isolates . The Chicago strain genome was therefore sequenced and compared to the Nichols genome , available since 1998 . Initial comparative analysis suggested the presence of 44 single nucleotide polymorphisms ( SNPs ) , 103 small ( ≤3 nucleotides ) indels , and 1 large ( 1204 bp ) insertion in the Chicago genome with respect to the Nichols genome . To confirm the above findings , Sanger sequencing was performed on most loci carrying differences using DNA from Chicago and the Nichols strain used in the original T . pallidum genome project . A majority of the previously identified differences were found to be due to errors in the published Nichols genome , while the accuracy of the Chicago genome was confirmed . However , 20 SNPs were confirmed between the two genomes , and 16 ( 80 . 0% ) were found in coding regions , with all being of non-synonymous nature , strongly indicating action of positive selection . Sequencing of 16 genomic loci harboring SNPs in 12 additional T . pallidum strains , ( SS14 , Bal 3 , Bal 7 , Bal 9 , Sea 81-3 , Sea 81-8 , Sea 86-1 , Sea 87-1 , Mexico A , UW231B , UW236B , and UW249C ) , was used to identify “Chicago-“ or “Nichols -specific” differences . All but one of the 16 SNPs were “Nichols-specific” , with Chicago having identical sequences at these positions to almost all of the additional strains examined . These mutations could reflect differential adaptation of the Nichols strain to the rabbit host or pathoadaptive mutations acquired during human infection . Our findings indicate that SNPs among T . pallidum strains emerge under positive selection and , therefore , are likely to be functional in nature .
Syphilis continues to be a common and serious disease , affecting at least 25 million persons worldwide [1] . It is a recognized cofactor in the transmission and acquisition of HIV [2] , [3] , and is a major cause of stillbirth and perinatal morbidity particularly in the developing world [4] , [5] . The peculiar biology of the causative agent of syphilis , Treponema pallidum subspecies pallidum ( T . pallidum ) , along with the inability to grow this pathogen continually in vitro , has hindered progress in understanding the pathogenesis of this disease . Syphilis research however , greatly benefited from the elucidation of the T . pallidum Nichols strain genome sequence [6] . This 1 . 138 Mb genome is among the smallest characterized in prokaryotes . The lack of genes encoding for several metabolic pathways ( i . e . Krebs' cycle , glyoxylate shunt , amino acid and fatty acid synthesis , etc . ) , restriction-modification enzymes , transposons , or prophages [6] , strongly suggests that T . pallidum's evolution as a human pathogen exploited progressive genome reduction and loss of those functions now provided by the host . Since its isolation in 1912 from the cerebrospinal fluid ( CSF ) of a patient with secondary syphilis [7] , the Nichols strain of T . pallidum has been continually propagated in rabbits , and has become the reference strain in experimental syphilis . Thus , it was the obvious choice for the original T . pallidum genome project . The Chicago strain of T . pallidum , isolated in 1951 by Turner and Rodriguez from a primary chancre [8] and far less extensively propagated in rabbits , has become increasingly important in the study of the pathogenesis of syphilis . Despite the fact that Nichols and Chicago belong to the same T . pallidum molecular strain type ( 14a/a [9] ) , suggesting an elevated degree of genetic similarity , several phenotypic and genotypic differences have been highlighted between these strains during experimental infection . Important differences between the two strains were described regarding gene expression of candidate virulence factors [10] , [11] , as well as antibody and cellular responses against Nichols and Chicago antigens during experimental infection [10] , [11] . An example of the above differences involves the 12-membered tpr ( T . pallidum repeat ) gene family [12] . The tpr genes and the antigens they encode have been the focus of intense research by our group , leading to the characterization of the immune response against these antigens during experimental syphilis [10] , [11] , and their potential as protective antigens [12]–[16] . The study of transcriptional patterns of these genes and the mechanisms that control expression of several tpr genes resulted in the identification of phase variation as a mechanism for controlling expression of at least five tpr genes [11] , [17] . Another member of the tpr gene family , tprK , undergoes extensive sequence variation mediated by gene conversion during infection , resulting in changes in seven discrete variable ( V ) regions in the tprK ORF [18] . Chicago has been shown to diversify the sequence of tprK at a significantly higher baseline rate than Nichols , before onset of detectable specific immunity , during intratesticular ( IT ) passages and intradermal ( ID ) infections . The tprK gene in the Nichols strain remains virtually clonal , varying its sequence only after onset of an adaptive immune response against the initial TprK antigen [19] , while variants arise throughout infection with the Chicago strain . In the presence of an adaptive immune response against the TprK antigen , the difference in accumulation of variants in Chicago is even more striking . To investigate whether genomic differences could explain the biological differences between Nichols and Chicago , the genome of the Chicago strain was elucidated using next-generation Illumina sequencing , annotated , and compared to the published Nichols genome . Genomic differences were confirmed by dideoxy-terminator ( DT ) sequencing using template DNA from Chicago as well as the Nichols strain ( Houston ) used for the original T . pallidum genome project . All coding sequences carrying SNPs , as well as approximately one third of the loci carrying small indels , were amplified and sequenced , revealing a strikingly high frequency of sequencing errors in the available Nichols genome . Nonetheless , comparison of 16 Nichols and Chicago polymorphic loci with the corresponding genomic regions of 12 more recently isolated T . pallidum strains ( SS14 , Bal 3 , Bal 7 , Bal 9 , Sea 81-3 , Sea 81-8 , Sea 86-1 , Sea 87-1 , Mexico A , UW231B , UW236B , and UW249C ) suggested that the genetic differences between Chicago and Nichols were acquired under the action of positive selection , and allowed us to speculate on the pathoadaptive nature of these changes in these T . pallidum strains .
No investigations were undertaken using humans/human samples in this study . New Zealand white rabbits were used for T . pallidum propagation . Animal care was provided in accordance with the procedures outlined in the Guide for the Care and Use of Laboratory Animals under protocols approved by the University of Washington Institutional Animal Care and Use Committee ( IACUC ) . The Chicago strain of T . pallidum subsp . pallidum , initially supplied by Dr . Paul Hardy and Ellen Nell ( Johns Hopkins University , Baltimore , MD ) , was propagated intratesticularly in seronegative New Zealand white rabbits as previously reported [20] . Briefly , three rabbits were injected with 5×107 T . pallidum cells per testis and checked daily for disease progression . Animals were euthanized approximately 10 days after infection , at peak orchitis , to recover the highest number of organisms before the onset of immune clearance . Testes were minced in 20 ml of PBS for approximately 10 min and suspensions were centrifuged twice for 10 minutes at 1 , 000× G to remove large host cellular debris . The supernate was then centrifuged at 18 , 000× G for 15 minutes to pellet treponemes . Treponemes were resuspended in 1 ml of PBS and stored on ice as the gradients were prepared . Discontinuous sodium and meglumine diatrizoate ( Renografin-60 , Bracco Diagnostics , Princeton , NJ ) gradients were prepared at room temperature by first diluting Renografin-60 stock solution ( 60% ) to the desired concentrations with PBS [6] . To obtain the discontinuous gradient , a first layer of 60% Renografin-60 ( 1 . 5 ml total ) was deposited in the bottom of a 10 ml Ultra-Clear Thinwall centrifuge tube ( Beckman-Coulter , Fullerton , CA ) , followed by 1 ml each of 37 . 5% , 25% , and 19% Renografin-60 dilutions , respectively . Approximately 0 . 2 ml of ice cold treponemal suspension was carefully layered onto each gradient and tubes were centrifuged at 20°C for 45 min at 100 , 000× G in an Optima XL-100K ultracentrifuge ( Beckman-Coulter ) equipped with a SW-41 Ti rotor . Fractions of approximately 0 . 2 ml were recovered by drop from the bottom of the tube . Fractions containing high numbers of treponemes ( identified by dark-field microscopy ) were pooled together and treated with a total of 5 units of RQ1 DNaseI ( Promega , Madison , WI ) to reduce the contamination by rabbit DNA . After treatment and heat inactivation of the enzyme ( 10 min at 65°C ) , an appropriate volume of 50× lysis buffer for DNA purification ( final concentration: 10 mM Tris , pH 8 . 0; 0 . 1 M EDTA; 0 . 5% w/v sodium dodecyl-sulfate ) was added to the treponemal suspension . DNA extraction was performed using the QIAGEN Genomic-tip 100/G kit ( Qiagen Inc . , Chatsworth , CA ) , according to the manufacturer's instruction and the sample was stored at −20°C until use . The list of the additional strains used here can be found in Table 1 . Strain propagation , harvest and DNA isolation for amplification and DT-sequencing protocols were performed as previously described [11] , [17] . Although we cannot formally evaluate treponemal growth rates for the strains used in this study , differences were found regarding the time between strain passage into the rabbit hosts , and the yield of treponemes at the time of harvest . Some strains were transferred every 10–12 days ( Nichols , Chicago , Sea 81-8 , SS14 , and UW249C ) , others every 15–30 days ( Bal 7 , Sea 87-1 , Mexico A , Bal 9 , Sea 86-1 , Sea 81-3 , and Bal 3 ) , while some required ≥30 days ( Sea 87-1 , UW231B , and UW236B ) . Treponemal yields varied from approximately ≥108 treponemal cells/ml of testicular extract ( Nichols , Chicago , SS14 ) , ∼107 cells/ml ( Bal 7 , Bal 9 , Sea 86-1 , Sea 81-3 , Bal 3 , UW231B , and UW249C ) , and ∼106 cells/ml ( Sea 87-1 , Mexico A , Sea 81-8 , and UW236B ) . The percentage of rabbit genomic DNA in the Chicago sample was determined by quantitating the copy number of the rabbit ( Oryctolagus cuniculus ) cystic fibrosis conductance transmembrane regulator ( RCFTR ) gene and the T . pallidum TP0574 gene ( which encodes for the 47 kDa antigen ) by quantitative real-time PCR ( qRT-PCR ) . Primer sequence , amplification protocol and standard curve preparation for the TP0574 gene were previously reported in detail [11] . RCFTR-S ( 5′-gcgatctgtgagtcgagtctt-3′ ) and RCFTR-As ( 5′-cctctggccaggacttattg-3′ ) primers ( Oligos Etc . Inc . , Wilsonville , OR ) were used to determine the rabbit CFTR gene copy number . Amplification was carried on for 45 cycles in a Roche LightCycler 2 . 1 instrument ( Roche , Basel , Switzerland ) using the Masterplus SYBR green kit ( Roche ) according to the manufacturer's instruction . The reaction conditions for these amplifications included a 10 sec denaturation step at 95°C , an 8 second annealing step at 60°C , and an extension step for 10 sec at 72°C . Acquisition temperature was set at 83°C upon amplicon melting curve analysis . The standard curve for the rabbit CFTR gene was prepared as for the TP0574 gene [11] . The sizes of the rabbit and T . pallidum genomes were taken into account to determine the percentage of rabbit DNA in the sample . Genomic DNA isolated from the T . pallidum Chicago strain was further processed for Illumina-based sequence analysis using the Paired End DNA Sample Prep Kit ( Illumina Inc . , San Diego , CA ) following the provided protocol . Genome sequencing was performed at the Center for Genome Research and Biocomputing ( CGRB ) at Oregon State University ( Corvallis , OR ) using a Genome Analyzer IIx System ( Illumina Inc . ) . A first draft of the Chicago strain genome was assembled using the reference-guided assembly program Maq [21] with the T . pallidum Nichols strain genome [6] ( GenBank accession number for Nichols is NC_000919 ) as reference . Regions in the reference-guided assembled genome where Maq could not resolve sequence were then compared to contiguous sequences assembled through the use of the de novo assembly software VCAKE [22] , and a single contiguous draft sequence was then produced . Nucleotide differences between matched pairs were identified using the Diffseq program from the Emboss software suite . The locations and effects of individual differences were first determined using an in-house SNP parsing program ( not currently online but available upon contacting the authors ) and then re-evaluated after the annotation of the Chicago strain was completed . Regions containing nucleotide differences between the Chicago and Nichols ( Houston ) strain were targeted by PCR amplification and conventional DT-sequencing to confirm the high-throughput sequencing data . DNA from both Chicago and the Nichols-Houston strain sequenced in the original T . pallidum genome project were used as template . A subset of these regions were selected randomly , and others were selected to confirm differences in genes possibly implicated in generation of diversity in the tprK gene ( TPChic0897 ) or transcriptional control ( such as TPChic0924 , encoding the toxin expression gene , also known as tex ) . Overall , 41 loci ( 39 . 8% of the total originally reported [23] ) carrying small indels were sequenced in both strains . Twenty six additional regions carrying small indels were amplified using DNA from the Chicago strain and sequenced to further confirm the reliability of the high-throughput sequencing approach . Primers ( designed using the Primer 3 software , http://frodo . wi . mit . edu/primer3/ ) are in File S1 . All PCR amplifications were performed in 100 µl reactions containing 200 µM each dNTP , 20 mM Tris-HCl ( pH 8 . 4 ) , 1 . 5 mM MgCl2 , 50 mM KCl , 400 nM of each primer , and 1 . 0 U of Taq DNA Polymerase ( Promega , Madison , WI ) with approximately 100 ng of DNA template in each reaction . Cycling conditions were denaturation for 5 min at 95°C , followed by 1 min at 95°C , annealing for 1 min at 60°C and extension for 1 min at 72°C for a total of 45 cycles . A final extension of 10 min at 72°C was included . Amplicons were purified using the QIAgen PCR purification Kit ( Qiagen Inc . ) according to the provided protocol , and the concentration of each sample was determined using a ND-1000 instrument ( NanoDrop Technologies , Wilmington , DE ) . Sequencing was performed at the Department of Biochemistry DNA Sequencing Facility of the University of Washington , Seattle , WA . Electropherograms were analyzed using the BioEdit software ( http://www . mbio . ncsu . edu/BioEdit/bioedit . html ) . Amplification and sequencing of 16 ORF fragments found to carry authentic SNPs between the Chicago and Nichols strains were also performed on 12 additional T . pallidum strains ( SS14 , Bal 3 , Bal 7 , Bal 9 , Sea 81-3 , Sea 81-8 , Sea 86-1 , Sea 87-1 , Mexico A , UW231B , UW236B , and UW249C ) . Sequencing of the TP0924 ( tex ) gene region ( containing the C→A transversion that truncates the putative Tex protein in Chicago ) was also performed using DNA template from various Nichols isolates maintained in different laboratories over the last two decades ( Seattle , Farmington , Dallas , and UCLA ) , as described above . The Chicago strain genome sequence was submitted to the J . Craig Venter Institute ( JCVI ) Annotation Service ( http://www . jcvi . org/cgi-bin/annotation/service/submit/annengine . cgi ) , where it was processed through JCVI's prokaryotic annotation pipeline . Included in the pipeline are 1 ) a gene-finding function with Glimmer , HMM , and TMHMM ( Hidden Markov Models and Trans Membrane Hidden Markov Models , respectively ) searches; 2 ) frame shift mutation identification through Blast-Extend-Repraze ( BER ) searches; 3 ) SignalP predictions for identification of signal peptides; and 4 ) automatic annotations from AutoAnnotate . The manual annotation tool Artemis ( www . sanger . ac . uk/Software/Artemis/v11/ ) was used to manually review the output from the JCVI Annotation Service and compare it with the Nichols strain genome annotation ( Nichols GenBank accession number is NC_000919 ) . To assess the genome-wide nucleotide diversity of protein-coding genes in Chicago and Nichols genomes , each gene was subject to a modified version of ZPS [24] to perform in batch mode ClustalW-based sequence alignment [25] , followed by calculation of the rates of nonsynonymous ( dN ) and synonymous ( dS ) mutations using the mutation-fraction method of Nei and Gojobori [26] .
Paired-end sequencing yielded a single circular contig devoid of sequence gaps . The Chicago genome [23] was found to be 1 , 139 , 281 bp long , in contrast to 1 , 138 , 011 bp in the published Nichols genome , suggesting that genomic differences might contribute to explain the differences in infection phenotypes associated with the Nichols and Chicago strains . Next-generation Illumina sequencing was not adversely affected by residual rabbit DNA , corresponding to ∼18% of the total DNA content of the sample , and the coverage of the Chicago genome ranged from ∼50× to ∼100× ( average depth coverage was 64× ) . Based on the annotation service provided by the JCVI , there were some ORF assignment discrepancies between the Nichols and Chicago genomes that were due to differences in the annotation algorithm rather than any sequence differences . The Chicago genome annotation identified 96 putative ORFs not previously identified in Nichols ( File S2 ) . The size of these ORFs was relatively small , ranging from 111 to 399 bp ( average length = 180 bp ) . To facilitate direct ORF comparisons between T . pallidum strains , we named the new ORFs based on their proximity to a coding sequence shared by both strains . ( For example , according to our nomenclature , TPChic0005a is an ORF annotated only in Chicago and located immediately downstream , either on the plus or minus strand , of TPChic0005 that is homologous to Nichols TP0005 . If multiple new ORFs follow a shared annotation , their order is reflected by the alphabetical letter following the ORF . TPChic1025a and TPChic1025b , for instance , follow TPChic1025 and precede TPChic1026 . On the other hand , 21 published Nichols ORFs ( File S3 ) were not identified by the JCVI annotation software in the Chicago genome sequence despite nucleotide sequence conservation between the two strains . Also , the annotation service provided by the JCVI permitted the re-analysis of the possible functions of some T . pallidum ORFs shared by two genomes . Among a total of 842 genes with the same annotation , a total of 158 ORFs ( File S4 ) previously listed as hypothetical or conserved hypothetical proteins in the Nichols annotation were now assigned putative identities . Newly annotated possible functions include tyrosine kinases ( TPChic0024 , TPChic0139 ) , efflux pumps ( TPChic0901 , TPChic0965 , TPChic0988 ) , and permeases ( TPChic0301 , TPChic0302 ) . New putative lipoproteins ( TPChic0069 , TPChic0087 , TPChic0149 , TPChic0625 TPChic0646 , TPChic0693 ) , outer membrane lipoprotein carriers and permeases ( TPChic0333 , TPChic0580 , TPChic0582 ) , and metal transporters with outer membrane subunits ( TPChic0034 , TPChic0035 , TPChic0036 ) were also identified . Over 99% of all predicted protein-coding genes shared between Chicago and Nichols strains were syntenic ( having same relative position in both genomes ) , thereby arguing against any major role of gene shuffling in shaping the genotypic/phenotypic differences between these two strains . No gene inversions were identified . Because the Chicago strain tprK is hypervariable with respect to Nichols , a consensus sequence for the seven variable ( V ) regions of this gene could not be obtained and , thus , are not accounted for in the nucleotide-based comparative analysis . In the complete Chicago genome sequence found in GenBank , the tprK V1–V7 region sequences are replaced by N's . For the Chicago genome , comparison of Illumina sequencing data with traditional DT-sequencing of genomic regions carrying SNPs showed perfect agreement between the two sequencing methods . Although we previously reported [23] that preliminary comparison with the published Nichols genome [6] identified the presence of 44 SNPs between Chicago and Nichols , recent DT-sequencing of the regions carrying these SNPs in the Nichols ( Houston ) strain , revealed a high frequency of sequencing errors in the published Nichols genome sequence [6] . Overall , only 20 authentic SNPs are found between Chicago and the Nichols genome: four are located within intergenic regions and 16 , all non-synonymous , within ORFs coding for putative proteins ( Table 2 ) . The SNPs were evenly split between C/T and A/G transitions and were not clustered , but distributed more or less evenly along the genome ( File S5 ) . To further explore whether these genomic differences between Nichols and Chicago genomes could have been promoted by the extensive propagation of the Nichols strain in the rabbit host , we analyzed the identity of each ORF-associated mutation in 12 other T . pallidum strains ( Table 3 ) which , like Chicago , were propagated in rabbits far less extensively than Nichols . As a result of these 14 genome cross-examinations of 16 SNP regions , we identified only one SNP accumulated in Chicago ( in TPChic0746 , Table 3 ) . Because the other 12 T . pallidum strains were identical to Nichols for this nucleotide position , we define such a change as “Chicago-specific” . Interestingly , the remaining 15 SNPs were determined to be “Nichols-specific” , in that Chicago and the other 12 genomes had identical nucleotides in these polymorphic positions , with the exception of the tprJ gene ( TpChic0621 ) where one of the 12 other strains ( Bal 7 , Table 3 ) showed a sequence identical to Nichols . These findings clearly demonstrate that 12 other strains analyzed here are significantly more similar to Chicago at the DNA level . Overall , these data strongly suggest that “Nichols-specific” SNPs were acquired through mutation , and not recombination; furthermore , because all the “Nichols-specific” SNPs predict amino acid changes in their respective putative proteins , such a significant predominance of “Nichols-specific” changes suggests functional adaptation of Nichols in the rabbit host . Of the 16 polymorphic genes targeted in our analysis , 12 ( 75% ) genes were annotated with defined functions , equivalent to 729 ( 74% ) total genes with defined functions in the annotated Chicago genome . Although this small set of genes did not permit us to statistically evaluate over-representation of functional categories , at least four of these polymorphic genes are known to encode putative virulence factors , possibly contributing to the phenotypic differences seen during infection between Nichols and Chicago . These genes are TPChic0488 ( Methyl-Accepting Chemotaxis protein ) , TPChic0621 ( TprJ protein ) , TpChic0922 ( Tex protein , discussed later in more detail ) , and TpChic0978 ( LspA Signal Peptidase II ) . The most direct way to detect any action of positive selection in protein-coding genes is to evaluate whether the rate of amino acid replacement ( dN , nonsynonymous mutation per non-synonymous nucleotide site ) is significantly higher than the rate of silent , synonymous mutations ( dS , synonymous mutation per synonymous nucleotide site ) , assuming silent mutations to be , in general , of a neutral nature . Due to the small number of SNPs and because all changes were non-synonymous , the dN/dS rate could not be evaluated directly either for individual genes or for all the polymorphic genes concatenated . If , however , for the sake of analysis we incorporate a synonymous change in the concatenated genes with SNPs , the resulting dN/dS value of 4 . 5 ( 0 . 00081/0 . 00018 ) shows that dN was significantly higher ( P = 0 . 03 ) than dS . Therefore , the absence of any synonymous SNP in the observed dataset strongly indicates that the genetic changes are positively selected and , likely , of an adaptive nature . Apart from a single large event involving a 1204 bp insertion in an intergenic region ( position 148519–149723 in the Chicago genome ) , indel analysis at the time the Chicago genome was released on GenBank [23] identified 103 small ( ≤3 nt ) insertions/deletions between the two genomes ( involving a total of 109 nt , due to the presence of 4 di-nucleotide indels , and 1 tri-nucleotide indel ) . DT-sequencing of 41 loci carrying such indels in both Chicago and Nichols ( Houston ) revealed however that , with the exception of two loci ( TPChic0667 , and the IGR 3′ of TPChic0222 ) , the above result was due to sequencing errors in the 1998 Nichols genome . DT-sequencing of “indel-carrying” loci using template DNA from the Chicago strain never showed discrepancies with the Illumina-based sequencing results . A list of erroneous differences ( both SNPs and indels ) between the Nichols and Chicago strains is reported in File S6 . Although only 39 . 8% of the originally reported differences due to indels were re-analyzed using DT-sequencing in both strains , it is striking that only 2 indels out of 41 ( 4 . 8% ) were confirmed as real . This indicates that the total extent of differences due to true indels between the Nichols and Chicago strains is likely to be significantly more limited than originally reported . A single C nucleotide insertion within the TPChic0667 ORF ( coordinates: 730194–731009 ) caused a frame shift and an early termination of the ORF with respect to Nichols' paralogous gene . As a result , when Nichols and Chicago annotations are compared , Nichols' TP0667 ORF ( 555 codons ) encompasses both Chicago's TpChic0667 ORF ( 275 codons ) and TPChic0667a ( 271 codons ) . This indel was found to be “Chicago-specific” , based upon analysis by DT-sequencing of the same locus in 12 more T . pallidum strains ( data not shown ) . A single C insertion ( position: 228663 ) was also confirmed in the intergenic region 3′ of TpChic0222 ( Table 2 ) . Indels that were identified by comparative genomic analysis but are not currently confirmed by DT-sequencing using the Nichols ( Houston ) strain are reported in Table 4 . Indels falling within homopolymeric nucleotide sequences were found in three Chicago ORFs ( TPChic0127 , TPChic0479 , and TPChic0618 ) , and within 3 intergenic regions ( 3′ of TPChic0026 , TPChic0121 , TPChic0621 ) . Among the “Nichols-specific” indels , the only mutation targeting intergenic regions appeared to be the 1204 bp deletion corresponding to the region downstream of TPChic0126 and upstream of TPChic0127 ( spanning the location of TPChic0126a/b/c regions in the reverse strand and TPChic0126d in the plus strand ) . Šmajs et al . [27] previously reported that a subpopulation of the Nichols ( Houston ) strain used in the original T . pallidum genome project does not carry such deletion , suggesting that this genomic region might not be stable within a single treponemal isolate . The 1204 bp insertion lies between two direct repeats of 24 bp ( aatgtatttcagggtgtctttctc ) , suggesting a loop-out mechanism for this deletion .
Chicago and Nichols differ in their origins of isolation ( primary chancre vs CSF ) , durations of propagation in the rabbit host , gene expression levels , induction of antibody and cellular immune responses to some antigens , and rates of TprK variation , the latter being higher in Chicago than in the Seattle Nichols [19] . With respect to the published Nichols genome sequence , a 1204 bp insertion was found in the intergenic region downstream of TPChic0126 . This large insertion contains 19 putative donor sequences used by T . pallidum to generate variability within all of the seven tprK V regions , especially V3 and V6 [19] . Although this insertion might be speculated to be a reason for Chicago's higher tprK variability , this 1204 bp fragment is also present in the Nichols strain currently propagated in our laboratory [18] , which is slow to develop tprK variants . Therefore , the number of donor sites alone cannot explain the relative hypervariability of Chicago tprK . The Nichols strain has been extensively propagated in rabbits and this might have selected for a tprK sequence that is optimal for survival and rapid growth in rabbit tissues . Frequent passage of the Nichols strain ( every 9–12 days ) for routine propagation , virtually in the absence of an adaptive immune response , might have permitted the reduction in Nichols' propensity to vary tprK . Comparative analysis between the two strains did not show differences in the genes coding for the recombination machinery typically involved in gene conversion ( i . e . ruv and rec genes , genes encoding site-specific recombinases or hypermutation homologues; data not shown ) . Structural predictions of the TPChic0899 ORF obtained using the Bio Info Bank Metaserver ( http://meta . bioinfo . pl ) however , found the encoded protein to be similar to an AddB-like deoxyribonuclease , a component of the counterpart of the E . coli RecBCD enzyme in Gram positive bacteria . TPChic0899 spans Nichols' TP0899 and TP0900 ( originally annotated as separate hypothetical proteins ) [6] . The presence of two ORFs in Nichols is due to a single G deletion that puts in frame the TGA triplet introducing a premature stop codon . Because of the possible involvement of this enzyme in homologous recombination , we further explored this difference between Chicago and Nichols . DT-sequencing of the region containing the G insertion was performed in a total of 16 T . pallidum isolates , including Nichols strains obtained from several laboratories and the SS14 strain ( also reported carrying the deletion; GenBank accession number CP000805 . 1 ) [28] . Our sequencing data revealed that the G nucleotide is actually present in all isolates ( Figure 1 ) confirming that the annotation of two separate ORFs , TP0899 and TP0900 , in Nichols [6] and SS14 [28] is indeed erroneous . Because this gene appears to be functional in all T . pallidum strains , it is , therefore , likely not associated with the increased rates of tprK variation that Chicago exhibits with respect to Nichols . Nonetheless , this example underscores the likelihood , when comparative genome-wide studies among T . pallidum strains are pursued , of encountering inaccuracies in available sequences . TPChic0924 , which encodes the Tex transcriptional regulator , could potentially explain reported differences in transcription of some tpr genes in Chicago vs . Nichols [11] . The Chicago Tex protein is predicted to be 250 aa shorter than in Nichols . Tex was first isolated and characterized in Bordetella pertussis by virtue of its negative effect on the transcription and expression of toxin genes ptx and cyaA [29] . Tex paralogs were then identified in a wide variety of bacterial species [30] , [31] and were shown to contain domains involved in nucleic acid binding [31] . Interestingly , studies conducted on the Pseudomonas aeruginosa Tex protein showed that presence of the carboxyl-terminal domain ( present in Nichols but not in Chicago ) permits Tex to bind nucleic acids [31] and thus inhibit transcription . The presence or absence of a complete Tex protein in T . pallidum could affect a strain's ability to express virulence factors . To further support the “Nichols-specific” nature of this change , it is found that all examined non-Nichols T . pallidum isolates carry the same A/C transversion ( Figure 2 ) that would truncate the Tex protein in Chicago , in sharp contrast with the five Nichols isolates ( Seattle , Houston , Dallas , Farmington , and UCLA ) , where the ORF encoding the Tex protein would not be truncated . When the Chicago genome was first released [23] , we reported that 44 coding sequences , annotated as independent ORFs in Nichols , are fused in Chicago leading to 21 considerably longer genes . TPChic0006 , for instance , was predicted to be 417 aa long , and to span Nichols' TP0006-0008 ( 51 , 216 , and 89 aa , respectively ) . It is however evident now that these initial observations were a result of sequencing errors in the original Nichols genome , and not the result , as initially postulated , of gene inactivation of original longer sequences by frame shift or nonsense mutations . Recently , Šmajs and collaborators [32] suggested that genomic decay might have played a central role in T . paraluiscuniculi's adaptation to the rabbit host and loss of infectivity to humans [33] , and the hypothesis that gene inactivation in the Nichols strain could reflect its adaptation to rapid passage in rabbits for nearly a century , also appeared plausible . Resequencing of the Nichols ( Houston ) genomic regions containing mutations hypothetically responsible for inactivation of these genes , however , clearly revealed that these annotation differences are also due to sequencing errors in the Nichols genome . It is therefore very likely that reannotation of the resequenced Nichols genome will be significantly more similar to that currently reported for Chicago . Similar findings were described by Cejková et al . [34] . A complete list of predicted gene fusions is reported in File S6 . Indels falling within homopolymeric nucleotide sequences were found in three Chicago ORFs ( TPChic0127 , TPChic0479 , and TPChic0618 ) , and within 3 intergenic regions ( 3′ of TPChic0026 , TPChic0121 , TPChic0621 ) . Growing evidence suggests that changes in the length of these homopolymeric repeats , likely induced by slipped-strand mispairing during DNA replication , might be involved in transcriptional or translational control of T . pallidum genes . For example , the poly-G repeat upstream of TPChic0621 ( TprJ ) was shown to control transcription of this gene through a phase variation mechanism that allows transcription only when the poly-G tract is eight ( or fewer ) nucleotide-long [17] . The poly-G repeat upstream of TPChic0026 ( encoding the fliG1 gene ) could have a similar role , although evidence of intra-strain variability of this homopolymeric tract is currently not available . Furthermore , recent evidence suggests that changes in the poly-G repeat within TpChic0127 could either cause a frameshift that prematurely truncates the putative TP0127 protein , or change its reading frame , resulting in a novel protein of approximately equal length but with a different amino acid sequence ( unpublished data ) . Variation in the homopolymeric tracts associated with TPChic0479 , and TPChic0618 can also influence the annotation of these ORFs . Analysis of SNPs in protein-coding genes showed only nonsynonymous mutations , suggesting the presence of recent diversification favoring structural changes in T . pallidum genomes . Overall , significantly higher rates of nonsynonymous changes in the Nichols genome indicate positive selection pressures in 16 protein-coding genes throughout the genome . Limited frequency of polymorphic genes did not permit us to determine whether these genes with recent structural changes could be grouped into specific functional categories of proteins . However , we found a strong clustering of polymorphic genes into two functional groups – membrane proteins and DNA-binding proteins . Within the set of genes with defined functions , the single “Chicago-specific” SNP accumulated in an ATP-binding protein-coding gene , while most of “Nichols-specific” SNPs were found to be in membrane protein-coding genes mostly related to transport and proteolysis ( Table 3 ) . Our study suggests that genetic variability likely influences the phenotypic differences seen between the Nichols and Chicago strains of T . pallidum [10] , [11] , [35] , even though definitive evidence for the correlation between specific genomic change ( s ) and phenotypic differences will require further investigation . This study also raises an important concern regarding the selection process that led to these mutations , believed to result from the adaptation of the Nichols strain to the rabbit host . Our comparative analysis incorporating 12 more T . pallidum strains for the regions carrying SNP changes in Nichols and Chicago , indeed initially suggested that this might be the case , and that the SNPs identified in Chicago and Nichols might reflect pathoadaptive changes the Nichols strain acquired following years of growth in the laboratory animal where it has been propagated so far . Interestingly however , in the DAL-1 genome ( GenBank accession number NC_016844 ) [34] , a T . pallidum strain recently isolated from the amniotic fluid of a pregnant woman [36] , most of the Chicago/Nichols polymorphic loci were identical to Nichols sequences . Based on this evidence , we cannot exclude that Nichols and DAL-1 represent a separate naturally-occurring clonal lineage within T . pallidum . The significant predominance of non-synonymous polymorphisms between Chicago and Nichols strains strongly suggests the likelihood of a role of positive selection in microevolution of T . pallidum strains , whether due to differential adaptation during rabbit passage or pathoadaptation of individual strains in the human host . Support for the mutational evolution of Nichols from an ancestral T . pallidum lineage also comes from the published genome of T . paraluiscuniculi ( Cuniculi A strain , GenBank accession number NC_015715 . 1 ) , closely related to T . pallidum [37] . In the Cuniculi A strain , nine of the Chicago/Nichols polymorphic loci ( TP0051 , TP0265 , TP0430 , TP0443 , TP0488 , TP0584 , TP0748 , TP0790 , and TP0978 ) are identical to non-Nichols strains that were analyzed here , confirming the “Nichols-specific” nature of the mutations . Ongoing research in our laboratories using comparative genomics on a population-wide scale will provide an insight into phylogenetic relationships of T . pallidum clonal populations and likely will help explain the role of such sequence changes during syphilis infection . | During infection , the agent of syphilis , Treponema pallidum subsp . pallidum ( T . pallidum ) , successfully evades the host immune defenses and establishes a persistent infection that can cause blindness , paralysis , or even death in some individuals that progress to the tertiary stage of the disease . The study of the Nichols strain of T . pallidum , isolated over a century ago and continually propagated in rabbits , has been paramount to deepen our knowledge on the biology of the agent of syphilis and the pathogenesis of this complex disease . Nonetheless , when the more recent Chicago isolate of T . pallidum is compared to the Nichols strain , significant differences in gene expression , gene conversion rates , and antibody responses against virulence factor candidates are detected during experimental infection . To investigate whether differences at the genomic level between Nichols and Chicago might explain such phenotypic differences , we sequenced the Chicago strain genome and compared it to the previously sequenced T . pallidum Nichols strain . Our findings indicate that the genomic differences between these T . pallidum strains emerge under positive selection , and are likely to be functional in nature , thereby being involved in shaping the phenotypic diversity between the Chicago and Nichols strains . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"infectious",
"diseases",
"biology",
"genomics",
"microbiology",
"genetics",
"and",
"genomics"
] | 2012 | Footprint of Positive Selection in Treponema pallidum subsp. pallidum Genome Sequences Suggests Adaptive Microevolution of the Syphilis Pathogen |
Subsets and Splits
No saved queries yet
Save your SQL queries to embed, download, and access them later. Queries will appear here once saved.